Computer-implemented color vision test and method for calibrating a computer-implemented color vision test

Abstract

A method for calibrating a computerized color vision test and a method for testing color vision on a computer are provided. The invention relates to a method for calibrating a color vision test under given ambient lighting conditions, the color vision test being displayed on a color display (12) of a computer (10) having at least one input interface (14), characterized in that before starting the color vision test, a calibration test is displayed on the display (12) under given ambient lighting conditions to a person with normal color vision, a color determination task requiring user input as part of the calibration test is displayed in at least one measurement area (31) of the display (12), the user input is read via the at least one input interface (14) of the computer (10), the read user input is evaluated, and as a result of the evaluation, a display error caused by a combination of the color reproduction capability of the display (12) and the effect of the ambient lighting conditions on color vision is determined, and a modification of the color vision test is determined based on the display error, the modification correcting the color vision test with respect to the display error. The invention also relates to a method for testing color vision using the calibration method, and a computer and a computer program configured to perform this method. The invention also relates to a color discrimination test, a color recognition test, and a combination of the two tests.

Description

Technical Field

[0001] The invention relates to a method for testing the color vision of a patient with color vision deficiency, wherein the color vision test is intended to be displayed on a color display of a computer having at least one input interface.

[0002] The invention also relates to a method of calibrating such a color vision test, which is used to test the color vision of patients with color vision deficiencies.

[0003] The invention also relates to a computer program configured to perform such a method and to a device storing the program.

[0004] In the context of the present invention, a computer is understood to be an electronic device that includes hardware and software components for receiving, storing, processing, and transmitting data, as well as program code containing appropriate computer commands for performing these tasks. A computer typically has at least one processor, a memory, and peripherals, i.e., input and output interfaces, the latter of which may be provided in an integrated manner, for example in the form of a touch screen. In the context of the present invention, a computer is specifically a desktop PC (personal computer), a laptop, a tablet computer, a smartphone, etc.

[0005] Color vision is provided by certain receptors in the retina, namely the cones, which provide daylight vision. There are three types of these receptors, based on their spectral sensitivity. L cones are primarily sensitive to light in the long-wavelength (red) range of the spectrum. M cones are sensitive to light in the medium-wavelength range (green), while S cones are sensitive to light in the short-wavelength range (blue). The perception of color is based on the relative values of the stimuli transmitted by the color-sensing L, M, and S cone receptors.

[0006] The two most important characteristics of color vision are:

[0007] - the ability to discern the hue or tone of a color (color discrimination), that is, the ability to distinguish between two different colors; and

[0008] -Color recognition, which is a person's ability to correctly name certain colors or shades.

[0009] Color vision defects (also known as color blindness) are caused by a small or large deviation between the spectral sensitivity curve of a color-blind person and that of a normal (healthy) person. The most common forms of color vision defects are protanomaly, protanopia, deuteranomaly, and deuteranopia. Problems with red are primarily caused by the spectral sensitivity of the L cones being closer to that of the M cones than in people with normal color vision. Consequently, the difference between the stimulation of the L and M cones decreases, resulting in a decreased ability to distinguish colors. In extreme cases, the spectral sensitivity of the L cones shifts to the point where it aligns with the spectral sensitivity curve of the M cones, a condition known as protanopia.

[0010] In cases of deuteranomaly, the spectral sensitivity of the M cones is closer to that of the L cones than in people with normal color vision. The result is similar to the previous case: the difference between stimuli from the L and M cones decreases, meaning that the ability to distinguish colors in this case also decreases. Deuteranomaly occurs when the spectral sensitivity curve of the M cones shifts to the point where it aligns with that of the L cones.

[0011] Approximately 23% of people with color vision defects have protanomaly, 73% have deuteranomaly, and approximately 4% have severe color vision defects (red and green). The most severe color vision defect is total color blindness, which is extremely rare. Defects in the blue-sensitive S cones are rare. Damage to these receptors is usually caused by disease or poisoning; color vision returns to normal once the cause is resolved. Therefore, nearly 100% of people with color blindness have protanomaly, red-green blindness, deuteranomaly, or green-green blindness.

[0012] Computers are increasingly being used to test color vision. Patients are asked to perform various color tests on these computers to determine whether they have any color vision deficiency and, if so, to determine its type and severity. Such tests include screen versions of well-known pseudocolor test books (e.g., Ishihara, Rabkin, Dvorin, Velhagen, Hardy, HRR) and screen versions of various other color discrimination tests. Although the colors displayed on electronic screens vary depending on the screen type and differ from those in printed or pictorial tests, their use is expanding, partly due to their increasingly improved color reproduction capabilities and partly due to their easy availability. Another advantage is that electronic display devices allow people in remote locations around the world who may have impaired color vision to self-diagnose (or undergo remote diagnosis), based on which they can confirm the diagnosis and order appropriate corrective devices (color filters) online.

[0013] However, the different color reproduction capabilities of electronic displays represent a significant problem, since electronic displays produced by different manufacturers have different primary colors, including different characteristic wavelengths and spectra of the primary colors.

[0014] Another issue is how well the test subject's eyes adapt to the ambient light. It's crucial to determine whether the sensitivity of each of the three receptors in the human eye (red, green, and blue) is the same, or whether there are significant differences between them due to varying ambient lighting conditions.

[0015] The same eye sees color differently under different states of adaptation. For example, the human eye sees different shades of color in incandescent light, sunlight, fluorescent light, and light emitted by light-emitting diodes. Therefore, if a color vision test program is displayed on a computer monitor, laptop, tablet, or smartphone, the test may produce erroneous results, potentially leading to an incorrect diagnosis, depending on the device's color reproduction capabilities and the degree to which the person's eyes have adapted to the ambient lighting conditions.

[0016] There are other problems with current computerized (and non-computerized) color vision tests.

[0017] The most famous color vision test is the Ishihara test, named after Dr Shinobu Ishihara. This test consists of a number of colored plates, called Ishihara plates, each of which contains a circle of dots of random colors and sizes. The pattern consists of dots that form numbers or shapes that are clearly visible to people with normal color vision, but invisible or difficult to see for people with red-green color blindness. Other color plates are deliberately designed to display numbers only to those with red-green color blindness, while being invisible to those with normal red-green color vision. The complete test consists of 38 color plates, but only a few plates are needed to detect obvious and severe color vision deficiencies.

[0018] The Ishihara test exploits the fact that people with red-green color blindness tend to distinguish these colors based on brightness. The human eye has lower light sensitivity in the lower frequency range, so under the same lighting conditions, red (lower frequency) appears darker than green (higher frequency). People with red-green color blindness have difficulty distinguishing between red and green, so their brains learn to distinguish the two colors based on brightness rather than chromaticity. In the Ishihara test, the random dots in each color plate vary in brightness and color. The colors used are pseudo-metachromatic, meaning that they appear isochromatic (having the same chromaticity) to subjects with color deficiency, whereas people with normal color vision can distinguish at least two different chromaticities. In the Ishihara test, confusion between red and green is detected, indicating both red (for protanomaly and protanopia) and green (for deuteranomaly and deuteranopia) color vision deficiencies. The colors used in each color plate are mixed colors, possibly including blue as a component, but the information (numbers or symbols) is hidden either by the red and green components or by the brightness of the dots. People with normal color vision see different shades of the applied red and green components, and their brains automatically connect the different shades of red and independently connect the different shades of green to detect numbers and symbols, thus detecting information encoded by color. In contrast, people with red-green color blindness cannot (or can only use this information to a limited extent), and instead their brains connect points of equal or similar brightness, thus detecting information encoded by brightness.

[0019] Attempts have been made to implement the Ishihara test on a computer monitor; however, even with good calibration, the specific color and brightness of each individual dot cannot be reproduced on a computer monitor, which can result in suboptimal diagnosis.

[0020] The inventors have recognized additional problems with the Ishihara test and its computer implementation. The inventors have realized that the Ishihara test relies on pattern recognition, a learning process performed by the brain. Therefore, the Ishihara test not only measures the brain's ability to distinguish between red and green, but also automatically measures the brain's ability to connect similar points and interpret them as images (numbers or symbols). The influence of this latter effect cannot be removed from the measurement results, meaning that people with impaired pattern recognition or children who have not yet developed these skills will score poorly on these tests, regardless of their color vision.

[0021] The inventors have recognized that another problem with the Ishihara test is that dots containing different shades of red and green and having significantly different brightnesses from each other are packed closely together on each color plate, so that when a person attempts to detect a number or symbol hidden among a plurality of these dots, his or her gaze will move across the color plate filled with dots. The rapid changes in the sensed color and brightness as a person's gaze passes through the quasi-random dots can negatively affect the nervous system and may cause epilepsy in people who are susceptible to epilepsy. The inventors have found that this problem is exacerbated when the Ishihara test (or any other similar test) is displayed on a computer monitor, especially when the screen's brightness is set to a high level.

[0022] The inventors also recognized that, due to the limited color display capabilities of computer monitors, the severity of color vision deficiency cannot be reliably diagnosed when the Ishihara test is administered on a computer. Specifically, it is impossible to distinguish patients with severe deuteranopia, near-deuteranopia, and deuteranopia from patients with moderate (moderately severe) deuteranopia. Similarly, patients with severe protanopia, near-deuteranopia, and protanopia cannot be distinguished from patients with moderate (moderately severe) protanopia.

[0023] The inventors also realized that the Stone Prototype Test only measures a person's ability to distinguish colors. However, there may be cases where a person's color recognition ability is impaired while their color discrimination ability is not. This is particularly true for color blind people who have received corrective lenses (including color filters) designed to restore normal color vision. While these people may score well on color discrimination tests, most commercially available corrective lenses do not improve color recognition ability, but only improve color discrimination ability. This means that a person wearing the corrective lenses can distinguish red and green, but still cannot recognize these colors as red and green (i.e., the colors that a person with normal color vision would perceive). Therefore, it is also necessary to measure a person's ability to distinguish colors. Summary of the Invention

[0024] It is an object of the present invention to provide a method and a computer program that overcome the problems associated with the prior art, in particular the problems mentioned above.

[0025] A first aspect of the present invention is based on the recognition that a person with normal color vision can assist in making a correct diagnosis of a person with color deficiency, wherein the person with normal color vision is positioned in the same location as the person with color deficiency being tested and uses the same device under the same ambient lighting conditions, whereby the measurement results provide information about the device and the ambient lighting conditions. The same test as that subsequently performed on the patient to be tested can be performed on a reference subject with normal color vision, or a different test can be performed separately for calibration purposes.

[0026] Taking into account the calibration measurement results of people with normal color vision, color vision tests can be modified to account for the display's color reproduction capabilities and the effects of ambient lighting on color vision. Two things can be done with the measurement results of people with normal color vision: The results, such as the monitor's R / G / B settings, can be used to specifically modify and adjust the parameters of the software color vision test performed on the patient when performing a given task. In this way, the test used to check color vision is now displayed to the patient with the modified parameters. Alternatively, the display of the color vision test is not modified, but after performing the necessary corrections, a different diagnosis can be made based on its results.

[0027] According to the present invention, the above object is achieved by a calibration method according to claim 36 .

[0028] The invention also relates to a color vision testing method according to claim 44, which uses the calibration method according to the invention.

[0029] The invention further relates to a computer configured to perform the measuring method according to claim 47 , and a computer program according to claim 48 .

[0030] A second aspect of the invention is based on the recognition that it is possible to provide a color discrimination test that is independent of the test subject's image recognition abilities and in which a plurality of pseudoisochromatic points are presented in different areas of a display screen such that variations in chromaticity and luminance within a single area are within acceptable limits and thus do not induce epileptic seizures. The invention therefore also relates to a color vision test according to claim 1.

[0031] A third aspect of the present invention is based on the recognition that a color recognition test can be provided that detects severe deuteranomaly and severe protonomaly in a very reliable manner, whereby the test is suitable for supplementing color discrimination tests in order to better differentiate between moderately severe and severely severe deuteranomaly / protonomaly. The present invention therefore also relates to a color vision test according to claim 13.

[0032] A fourth aspect of the invention is based on the recognition that a color recognition test can be provided to supplement a color discrimination test in order to distinguish severe color vision deficiencies from mild and moderate color vision deficiencies.

[0033] Advantageous embodiments of the invention are defined in the appended dependent claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be explained in more detail by means of exemplary embodiments with reference to the accompanying drawings, in which:

[0035] Figure 1 A schematic diagram of an exemplary computer for implementing the method according to the present invention is shown;

[0036] Figure 2a A schematic diagram showing a sub-pixel of an exemplary display;

[0037] Figure 2b Schematic diagrams showing sub-pixels of different exemplary displays;

[0038] Figure 2c shows a schematic diagram of a sub-pixel of a third exemplary display;

[0039] Figure 3 A flow chart showing an embodiment of a calibration method according to the present invention is shown;

[0040] Figure 3a shows a flow chart presenting supplementary steps of an exemplary embodiment of a calibration method according to the invention;

[0041] Figure 4 Schematic screen images showing an exemplary embodiment of a calibration test according to the present invention;

[0042] Figure 5 A flow chart illustrating an exemplary embodiment of a color vision test according to the present invention;

[0043] Figure 6 Schematic screen images showing an exemplary embodiment of a color vision test according to the present invention;

[0044] Figure 7 Schematic screen images showing different exemplary embodiments of a color vision test according to the present invention;

[0045] Figure 7a shows the correction for Figure 7 An illustrative guide to the display of color vision tests;

[0046] Figure 8 is a flow chart of a preferred embodiment of a color vision test;

[0047] Figure 9 is a CIE chromaticity diagram showing the color gamuts of various displays;

[0048] Figure 10a is a CIE chromaticity diagram showing red-blind confusion lines;

[0049] Figure 10b is a CIE chromaticity diagram showing the deuteranopia confusion lines;

[0050] Figure 10c is a CIE chromaticity diagram showing tritanopia confusion lines;

[0051] Figure 11 is a CIE chromaticity diagram showing MacAdam ellipses;

[0052] Figure 12 is a schematic diagram of a screen image of an exemplary color discrimination test displayed on a computer monitor;

[0053] Figure 13 is a CIE chromaticity diagram showing target and test color pairs selected from the protanopia confusion line;

[0054] Figure 14 CIE chromaticity diagram showing target and test color pairs selected from the deuteranopia confusion line;

[0055] Figure 15 is a flow chart of another preferred embodiment of a color vision test; and

[0056] Figure 16 is an illustration of a screen image of an exemplary color recognition test displayed on a computer monitor. DETAILED DESCRIPTION

[0057] Figure 1 A schematic diagram of an exemplary computer 10 for implementing the method according to the present invention is shown. Reference numeral 10a denotes a desktop PC, which is equipped in a conventional manner with peripheral devices, such as a monitor 12a having a display 12 and input interfaces 14 (in this example, a keyboard 14a and a mouse 15a). Naturally, the PC 10a may also have other input interfaces 14 and peripheral devices.

[0058] Reference numeral 10b denotes a laptop computer having a built-in display 12, a built-in keyboard 14b, and a built-in mouse 15b as input interfaces 14. Figure 1 The exemplary laptop computer 10b shown also has an external mouse 15b'. Naturally, the laptop computer 10b may also have other input interfaces 14 and output interfaces.

[0059] Reference numeral 10c denotes a smartphone having a built-in display 12 in the form of a touch screen 13, which functions simultaneously as an input interface 14 and an output interface. In addition, other peripheral devices can be connected to the smartphone 10c.

[0060] Reference numeral 10d denotes a tablet computer, which also has a built-in display 12 in the form of a touch screen 13. Thus, the tablet computer also serves as both an input interface 14 and an output interface. Furthermore, the tablet computer 10d has an input interface 14 in the form of a stylus 15d, which enables more precise use of the touch screen. Of course, other peripheral devices are also possible.

[0061] At least one processor 16 and a storage medium 17 are located inside the computer 10. Figure 1 In the figure, a computer 10 is schematically indicated by a dashed line, in the present case the storage medium 17 is a memory 18. In the case of a desktop PC 10a, the processor 16 and the memory 18 are arranged in a separate housing 11. The central processing unit 16 is usually referred to as a central processing unit (CPU), and the text "CPU" in the text box in which the processor 16 is schematically indicated by a dashed line refers to the central processing unit. However, the at least one processor 16 can be provided in any known form, for example as part of a system on chip (SoC), which can also have an integrated memory 18. The latter component is usually abbreviated to SoC (System on Chip). There can be several processors in a computer 10, an example of which is a so-called multi-core processor, which is actually a chip in which the individual cores are themselves processors 16.

[0062] From the perspective of storing data, memory 18 can be of two types: volatile and non-volatile. Non-volatile memory 18 is a storage medium 17 (background storage) that is suitable for permanently storing program code (referred to as a program for short) containing computer instructions that can be executed (run) by processor 16. Very fast access memory 18, which is actively used by processor 16 during program execution, is usually volatile. While the program is running or triggered by it, computer 10 receives data (for example, computer 10 reads data via input interface 14), stores the data (temporarily in volatile memory 18 or long-term in background storage), processes the data, and transmits it (for example, to an output interface, for example, which displays the data on display 12).

[0063] Other conventional components may also be located in computer 10 in a manner known to those skilled in the art (eg, data buses, switches, connections, etc., not separately shown here).

[0064] The display 12 of the computer 10 used to implement the invention is a color display 12, which means that each pixel 20 forming an image point of the display 12 is composed of several so-called sub-pixels 21 of different colors, such as Figure 2a , wherein adjacent sub-pixels 21 forming one pixel 20 are framed by dashed lines. A pixel 20 of the display 12 typically includes sub-pixels 21 of three colors: red, green, and blue sub-pixels 21, which are arranged in a Figure 2a are marked by the letters R, G and B respectively. The technology used to display the individual colors depends on the type of display 12 and can be, for example, the use of organic light-emitting diodes (OLEDs) of appropriate colors, as in the case of OLED displays 12, or, for example, the use of controllable liquid crystal color filters and a white background lighting, as in the case of LCD-type displays 12. It should be noted that it is obvious to a person skilled in the art that other display technologies are conceivable, such as QLED or plasma technology. In addition to the traditional red, green and blue sub-pixels 21, there are also LCD screens that contain sub-pixels of other colors, such as in Quattron brand LCD screens, which have a fourth yellow sub-pixel 21 within each pixel 20. This is in Figure 2b In the Figure 2c In the case of the LCD screen shown using the so-called PenTile RGBW technology, the fourth sub-pixel 21 marked with W is controllable to become transparent, in such a way that it transmits the white background light without passing through the color filter. In other words, this type of display 12 uses white as the fourth color.

[0065] The calibration test and the color vision test according to the present invention are performed by a computer program on the computer 10 used by the user. The two types of tests can be implemented by separate computer programs or within the framework of a single computer program. The computer program can be stored in the non-volatile memory 18 of the computer 10 or can be downloaded from a separate non-volatile storage medium 17 (e.g. Figure 1 The computer 10 is read from a USB flash drive 19 shown, which is already shown next to the desktop computer 10a. Alternatively, the external storage medium 17 may be a CD, DVD, memory stick, etc. An embodiment is also conceivable in which the computer program is stored on a remote computer, typically a server, and the display of a given test is implemented on the display 12 of the local computer 10, for example, via an Internet connection, and the user input data read through the local input interface 14 is also transmitted via the Internet to the program running on the server for processing.

[0066] It is obvious to a person skilled in the art that other computers than the computer 10 presented here are also conceivable, but which may also be suitable for implementing the method according to the present invention.

[0067] Calibration test

[0068] use Figure 3 The shown flow chart presents an exemplary embodiment of a calibration method according to the invention.

[0069] During the method, a computer 10 is used, which has a color display 12 and at least one input interface 14. A color vision test for testing the color vision of a color-blind person will be displayed on the display 12 of the computer 10 under given ambient lighting conditions, but before this, a calibration is required in order to be able to take into account the influence of the color reproduction capabilities of the given display 12 on color vision and the influence of given ambient lighting conditions on color vision, both of which affect the color vision test, as described above. The given ambient lighting can be natural light (e.g. daylight in a room) or artificial light (e.g. a room illuminated with lamps). Calibration is related to the performance of the color vision test under given ambient lighting conditions, and therefore requires that the given ambient light conditions do not change between the time when the calibration test is displayed and the time when the color vision test is subsequently displayed. To this end, in step S10, preferably before the color vision test begins, the calibration test is displayed on the display 12 under the same ambient lighting conditions as for a person with normal color vision. Preferably, instructions for performing the calibration test by a user with normal color vision are displayed;

[0070] Anyone who knows they are not color blind (i.e., anyone with normal color vision) can be invited to perform the calibration test in the environment of the patient to be tested. If no one in the environment surrounding the patient to be tested knows their condition with certainty, it is best to invite two women to perform the calibration test, because the proportion of women with color vision deficiency (approximately 0.4%) is much lower than that of men, so the probability that both women have color vision deficiency is also negligible. Apart from this very rare situation, for example, if one of the people invited to perform the calibration test has color vision deficiency, then based on the different results of the two calibration tests, it can be determined that at least one person has color vision deficiency, and the calibration will therefore be unsuccessful, and another person or pair of people must be found to perform the calibration. However, all of this does not make the method according to the present invention unusable, all of this means that the method will only fail in very rare cases (for example, if the results of the calibration tests performed on two people are contradictory) or the method will lead to erroneous results (for example, if the person performing the calibration test believes they have normal color vision, but in fact they have color vision deficiency). People with normal color vision and their selection are not part of the method according to the invention. In other words, step S10 of the method according to the invention only extends to displaying the calibration test on the display 12, which is designed for people with normal color vision and should therefore be performed by people with normal color vision. In the following, the person participating in the calibration test will be referred to as the user.

[0071] During the calibration test displayed in step S10, preferably, in step S101, instructions are given to the user (in other words, a person with normal color vision) that, when executed, ensure that their eyes have adapted to the given ambient lighting conditions. For example, text instructions are displayed on a display for the user undergoing the calibration test, instructing the user to view an object 40 known to be white, such as a piece of white paper, for a period of several minutes, preferably at least two minutes, and optimally ten minutes. Color vision is partly a brain signal processing task, so even though the chromaticity of the ambient lighting may be yellow, blue, etc., the human brain is able to adjust the white balance. This effectively means that the sensitivity ratios of the S, M, and L cone receptors are adjusted so that, after a certain period of time, the user sees the color of the known white object 40 as white, even though the various wavelength components of the light reflected from the object 40 as measured by the instrument are present at different intensities. In other words, the light reflected from the object 40 has an objective chromaticity (e.g., a slightly yellowish hue in yellow light).

[0072] The instructions to the user may not only be displayed during the calibration test, nor may they only be displayed on the display 12. For example, it is conceivable that the instructions have been provided in advance as part of the instructions for use in electronic or paper form.

[0073] There is no guarantee that the eyes of the user performing the calibration test will have adapted to the ambient lighting conditions, but, for example, the above instructions can be used to increase the probability that the user's eyes have adapted appropriately, and thus the calibration will be more accurate. If the user is not required to perform training to adapt their eyes to the ambient lighting conditions, the eyes will still have a certain degree of adaptation, so calibration can still be performed and can be used to improve color vision testing, but the accuracy of the calibration will be reduced from the perspective of taking into account the effects of ambient lighting conditions on color vision.

[0074] The user is also preferably instructed to maintain and update the eye's adaptation while performing the calibration test, for example, by returning to looking at white (object 40) for a longer period of time than at display 12, to prevent the eye from becoming "unadapted" due to looking at colors on display 12 for a prolonged period of time during the test. If white paper or other known white object 40 is not used, and the eye's automatic adaptation to the environment is relied upon, it is still preferred to instruct the user to occasionally look away from display 12 and toward the environment while performing the calibration test. Naturally, it is more effective if the user looks at an object 40 that he or she knows to be white, because adaptation is faster this way.

[0075] During the display of the calibration test in step S10, in step S102 a color determination task requiring user input is displayed in at least one area of the display 12. Hereinafter, this area will be referred to as the "measurement" area. Figure 4A screen image of an exemplary color determination task displayed on the display 12 is presented. This may be, for example, a screen image of the monitor 12a of the desktop PC 10a. User-selectable text instructions 32 (marked with a dotted box) are displayed on the screen image, preferably in front of the entire background of the measurement area 31. In other words, the background of the text instructions 32 is also part of the measurement area 31. The instructions 32 may also include instructions related to eye accommodation, or may simply include instructions related to the execution of the calibration test.

[0076] In this example, sliders 33 are displayed for adjusting the intensities of the red, green, and blue sub-pixels 21. If the display 12 has other colored sub-pixels in addition to the red, green, and blue sub-pixels 12, their intensities are preferably set to zero for the purpose of performing the color determination task, and adjustment of these additional colored sub-pixels is unavailable to the user. The user can change the position of the sliders 33 via one of the input interfaces 14 of the computer 10 (e.g., a mouse 15a, 15b, 15b' or by using a finger or stylus 15d on the touch screen 13). By adjusting the intensities of the sub-pixels 21 in the measurement area 31, preferably according to the actual intensities determined by the sliders 33, the color of the measurement area 31 is adjusted, so that the user sees the color being mixed in the measurement area 31 in real time. Thus, in this case, the user input is the adjustment of the sliders 33, which the user provides in response to the color determination task and which is read in step S11 via one of the input interfaces 14 of the computer 10.

[0077] During the exemplary color determination task in step S101, the user must adjust the color of the white object 44 being viewed for an extended period within the measurement area 31 by adjusting the sliders 33, in other words, by adjusting the intensities of the red, green, and blue sub-pixels 21. Text instructions 32 displayed on the display 12 may be associated with this. Preferably, when setting the color of the white object 40, the user may also be instructed to view the object 40 for extended periods of time and the display 12 for shorter periods of time, so as to minimize the effect of the display 12 lighting on the degree of eye adaptation. If the user views the display 12 for extended periods of time, the degree of eye adaptation to ambient light will decrease, and calibration accuracy will be reduced from the perspective of considering the influence of ambient lighting conditions.

[0078] When setting the white color of the measurement area 31, it may happen that, although the color of the measurement area 31 is now similar to the color of the reference white object 40, it is still brighter or darker than the color of the reference white object 40. To correct this, a further slider 34 is preferably provided, with which the brightness of the measurement area 31 can be adjusted. Optionally, the hue of the display 12 and its saturation can also be adjusted.

[0079] The user input read in step S11 is evaluated in step S12, and as a result of the evaluation, a display error caused by a combination of the color reproduction capabilities of display 12 and the effect of ambient lighting conditions on color perception is determined in step S13. It is not intended to separate the effect of the color reproduction capabilities of display 12 on color perception and the effect of ambient lighting conditions on color perception from each other, or to determine them separately, as these together affect how the user views the colors appearing on display 12.

[0080] During the evaluation, it is preferred to take into account the fact that most problems with color vision manifest as mistaking red for green. Therefore, in the case of the example above, the evaluation preferably involves determining the values r and g for the intensities of the red and green sub-pixels 21, based on the positions of sliders 33 and, optionally, slider 34, set by the user. In the standard description, each slider position is a number between 0 and 255, and then taking their difference. Furthermore, the brightness value br read out, which indicates how bright the pixel of display 12 is, may also be taken into account. In step S13, the result of the evaluation will be an rg value, which is considered to be a display error caused by a combination of the color reproduction capabilities of display 12 and the effects of the ambient lighting conditions on color vision.

[0081] Naturally, a color vision test may be intended to measure very rare color vision defects associated with errors in the S cone cells. In this case, during calibration, the b value associated with the intensity of the blue sub-pixel 21 is taken into account, and from this, for example, the rb or bg value as a display error is also determined. Optionally, the evaluation result may also include rg, rb, and bg values as various components of the display error.

[0082] In step S14, a modification to the color vision test is determined based on the display error, the modification correcting the color vision test with respect to the display error.

[0083] For example, the results of the calibration test according to the above example will be used for a color vision test to measure red-green color blindness, in which case the patient being tested needs to set an accurate yellow (i.e., a yellow without orange or green traces) by using the slider 33 for adjusting the intensity of the red and green sub-pixels 21, while reducing the intensity of the blue sub-pixel 21 to 0 using the slider 33 for adjusting the intensity of the blue sub-pixel 21.

[0084] As a modification to the color vision test, two main things can be determined. On the one hand, modifications to the parameters of the color vision test can be determined in order to correct the parameters related to the intensities of those colored subpixels 21 that are accounted for in the display errors when the color vision test is displayed on display 12. In other words, in this example, when the color vision test is displayed, the RGB difference calculated during calibration is added to the intensity value of the red subpixel 21 to correct for the display errors. Thus, when the yellow color is set, a corrected yellow result is displayed to the patient being tested on display 12. In this way, the position of the patient's setting slider 33 can be read at the time of user input, and a correct diagnosis can be established as a result, since the patient's color vision deficiency is actually being measured, not a display error.

[0085] On the other hand, it is also possible not to modify the display of the color vision test; instead, as a modification of the color vision test, a modification of the evaluation of the color vision test is determined, which correction corrects the measurement result obtained by the color vision test for the display error according to the intensity of the colored sub-pixels 21 taken into account in the display error during the calibration when evaluating the color vision test. In the case of the present example, this means that the RG difference obtained during the calibration (i.e., the R2 value set for the intensity of the red sub-pixel 21) is then subtracted from the result belonging to the yellow set by the tested patient. Since the "result" is a value close to zero (precisely zero in the case of a person with normal vision and an ideal display 12), a correct diagnosis of a person with color vision deficiency is established by subtracting the display error from this result as a correction. If the evaluation of the color vision test is modified instead of the display of the color vision test, then the calibration test can also be performed after the color vision test has been performed.

[0086] A color vision test is also part of a preferred embodiment of the present invention. In this case, a calibration method is performed before the color vision test begins under given ambient lighting conditions, and then in step S19 the modifications determined during the calibration method are performed. According to the above, this may include modifications of display parameters or modifications of evaluation parameters of the color vision test.

[0087] In step S20, the modified color vision test is displayed on the display 12. In this example, the color vision test is very similar to the calibration test, so in order to avoid unnecessary repetition, reference will be made to the details of the calibration test, and if the color vision test does not contradict it, the description of the calibration test is also valid here.

[0088] The results of the calibration can be improved if care is taken to ensure that the patient's eyes are also adapted to the same ambient lighting conditions before he or she performs the color vision test displayed on the display 12. To this end, in step S201, during the color vision test (or optionally before), instructions are preferably given to ensure that the patient's eyes are adapted to the ambient lighting conditions, for example, by instructing the patient to look at an object that he or she knows is white for a few minutes, preferably at least 2 minutes, and optimally 10 minutes, preferably the same white object 40 used in the calibration test by a person with normal color vision. Similar to the calibration method, instructions to ensure eye adaptation can also be given to the patient in another form.

[0089] In the example case of step S20, during the display color vision test in step S202, a color determination task requiring user input for measuring red-green color vision deficiency is displayed on the display 12, a schematic screen image of which is as follows: Figure 6 As shown. In the case of this task, the patient needs to set a yellow color (i.e. a yellow color without traces of orange or green) in the measurement area 31 of the display 12 by using sliders 33 for adjusting the intensity of the red and green sub-pixels 21. Instructions 32 corresponding to this are displayed on the screen (marked with a dotted box). During the performance of the task, the intensity of the blue sub-pixel 21 is reduced to 0 and, in the case of this embodiment, there is no slider 33 with which the user can adjust the blue. The intensity of any further colored sub-pixels is also set to 0 and cannot be modified by the user. Preferably, in this case, a separate slider 34 is provided for adjusting the brightness.

[0090] In this case, the user input is an adjustment of the sliders 33 and 34 , which is read in step S21 via one of the input interfaces 14 of the computer 10 .

[0091] The user input read in step S21 is evaluated in step S22 and as a result of the evaluation the patient's color vision is determined in step S23. The latter also includes the type and severity of the color vision deficiency.

[0092] In the case of this example, the calibration test is different from the color vision test, however, examples can still be conceived in which the calibration test is the same as the color vision test. For example, during the calibration test, the task of a person with normal color vision as a user is also to mix yellow (preferably after resting his or her eyes on the white reference object 40).

[0093] Figure 3aAnother possibility, shown in FIG, is to also require the patient being tested to perform a calibration test, which can simplify the color vision test. For example, when calibration is completed in step S15, the patient being tested is also shown a calibration test, during which, in step S151, the patient is instructed to view a white reference object 40, which a person with normal color vision also views for a few minutes, preferably approximately 10 minutes. Then, in step S152, the patient is also instructed to perform the white setting task described above, and in step S16, the user input associated with the setting of slider 33 for adjusting the intensity of green is read, in other words, the g value set by the patient is read. Using this g value, the color vision test associated with the setting of yellow is modified in step S17, so that the value of green subpixels 21 in the measurement area of the displayed color vision test is set to g, and the patient must, by contrast, set the intensity of red subpixels 21 in a manner that mixes with yellow. By fixing the intensity of green, approximately the same control range of display 12 with nonlinear characteristics can be used.

[0094] During the calibration test and color vision test, other types of color determination tasks may be displayed. In the case of the above example, the task involves a person with normal color vision or the patient being tested setting a given color in the measurement area 31 of the display 12 by adjusting the intensity of various colored sub-pixels 21. However, there may be many types of color determination tasks, which may involve, for example, adjusting the intensity of more or fewer colored sub-pixels 21, for example, the user only needs to adjust one type of colored sub-pixel 21, while the intensity of the other colored sub-pixels 21 is set to a specific value. The color determination task may include, for example, selecting one or more color samples from a set of color samples displayed in the measurement area 31 according to specific instructions (for example, selecting a color sample that the user considers to be more orange from two different shades of yellow).

[0095] Figure 7An illustration of a screen image for a more complex color determination task is shown, appearing on the display 12 of a smartphone 10c serving as a computer 10. In this example, the measurement area 31 consists of three circular surfaces 31a, 31b, and 31c, the colors of which can be changed using sliders 33a, 33b, and 33c displayed below each circular surface. Yellow is set in the first circular surface 31a by changing the position Y of the first slider 33a. In the case of an ideal display 12, the center position of slider 33a, indicated by the dashed line V1, would cause the first circular surface 31a to appear yellow. In this case, the intensities of the red and green sub-pixels 21, 21 within the circular surface 31a would both be 255 (on a standard scale of 0-255). At any position of slider 33a, the intensity of the blue sub-pixels 21 within the circular surface 31a is reduced to zero. By moving the slider 33a to the right (toward the terminal position marked by R) compared to the center position, the intensity of the red sub-pixel 21 remains unchanged, while the intensity of the green sub-pixel 21 decreases uniformly to 205 within the circular surface 31a, so in this case, more and more orange-yellow colors will appear because the proportion of the intensity of red will become larger and larger compared to green. By moving the slider 33a to the left (toward the terminal position marked by G), the intensity of the red sub-pixel 21 decreases uniformly to 205, while the intensity of the green sub-pixel 21 remains constant within the circular surface 31a, so in this case, more and more greenish yellow colors will appear because the proportion of the intensity of green becomes larger and larger compared to red.

[0096] The current position Y on the slider 33a is described using values of ±50, so that the center position marked by the dotted line is regarded as 0, at the right end position, the intensity of the red sub-pixel 21 is 50 greater than the intensity of the green sub-pixel 21, so this is regarded as a value of +50, and at the left end position, the intensity of the red sub-pixel 21 is 50 less than the intensity of the green sub-pixel 21, which is regarded as a value of -50.

[0097] Naturally, a slider 33a could be used, on which the intensities of the red and green pixels could be adjusted over a full scale from 0 to 255. However, in the case of displaying a color vision test on a small-sized display 12 (e.g., the display 12 of a smartphone 10c), this would result in a small movement of the slider 33a, causing significant chromaticity changes within the circular surface 31a. In other words, the patient would not necessarily be able to accurately mix the colors he or she sees as yellow. To overcome this problem, it is preferable to only allow the slider 33 to move within the expected range around the color yellow, which is why the aforementioned scale of ±50 was chosen in this example. Naturally, a wider scale could be chosen, for example, moving between values of ±100.

[0098] If, during the calibration method, the brightness is also adjusted using slider 34, this result can be used to provide a more accurate measurement in such a way that, for example, the 255-255 values for red and green are reduced according to the brightness value. For example, in a calibration test using slider 34, a person with normal color vision sets the brightness to 95%, and accordingly, the 255-255 values are reduced to 243-243 (this represents a correction of one component of the display error). In this case, at the right end position, the intensity value of green sub-pixel 21 will be 193, while at the left end position, the intensity value of red sub-pixel 21 will be 193. Naturally, many forms of scales are conceivable, so for the sake of simplicity, we will use the original example. In other words, at the center position marked by dashed line V1, the intensity of both red and green sub-pixels within circular surface 31a is 255, while the intensity of blue sub-pixel 21 is 0. If the brightness value set during calibration is also taken into account, then according to what is described above, the intensity corresponding to 255 will be proportionally reduced, and in the examples to be explained below, this reduced value will be used instead of the maximum value of 255.

[0099] Similarly, in the second circular surface 31b, the task is to set the purple color using the second slider 33b so that, when in its center position, marked by the dashed line V2, the intensity of the blue sub-pixel 21 and the red sub-pixel 21 within the circular surface 31b is 255, while the intensity of the green sub-pixel 21 is 0 at any position of the slider 33b. Slider 33b in the end position marked by B produces a purple color with a slight bluish tint, and the slider in the end position marked by R produces a purple color with a slight reddish tint, in this manner, similar to the above description, where the scale here is a value of ±50. The set position on the second slider 33b is marked by P.

[0100] In the third circular surface 31 c, the task is to set a cyan color, which is ideally created at the center position marked by the dotted line V3, with the values of the blue and red sub-pixels 21 within the circular surface 31 c both set to 225. In a similar manner to that described above, the third slider 33 c produces a bluish cyan color at the end position marked by B, while the end position marked by G will produce a greenish cyan color within the circular surface 31 c, with the values within the ±50 scale corresponding to the arbitrary position T.

[0101] The values belonging to the user-set Y, P, and T positions are represented by the letters y, p, and t, respectively.

[0102] Ideally, a person with normal vision would set all three sliders to 0, while a person with red or green color vision deficiency would set the y, p, and t values as follows:

[0103] People with red color vision deficiency set the following values

[0104] - On the yellow slider, y > 0

[0105] - On the purple slider, p>0

[0106] - On the turquoise slider, t=0.

[0107] People with green color vision deficiency set the following values

[0108] - On the yellow slider, y < 0

[0109] - On the purple slider, p = 0

[0110] - On the turquoise slider, t<0.

[0111] The numerical values (0-50) of the "+" and "-" values show the severity of a given type of color blindness.

[0112] In theory, the type and severity of protanopia or deuteranopia could be determined solely by the position of the yellow slider, but the redundancy created by testing for a variety of colors increases the reliability of the measurement.

[0113] In the above color vision test, the calibration method is used as follows.

[0114] If A / 4 paper is used for calibration, the Δ=rg value can be used as the display error (or as a component thereof if other factors such as brightness are determined in addition to this).

[0115] As an initial step in using a color vision test, this value must be provided to the color vision measurement software (e.g., the measurement software may ask for the value and the patient must type it in, or, for example, the measurement software and calibration software may communicate with each other and the calibration software immediately sends the Δ value to the measurement software).

[0116] Consider making corrections in the following manner.

[0117] If the correction is applied at display time, the corrected colors are displayed within the circular surfaces 31a, 31b, 31c for the positions Y, P, T of the sliders 33a, 33b, 33c as follows:

[0118] At each position Y of the slider 33a, the color corresponding to the value y'=y+Δ is displayed within the circular surface 31a. If y'≥0, the intensity of the red sub-pixel 21 of the displayed color is 255, and the intensity of the green sub-pixel 21 is 255-y'. If y'<0, the intensity of the green sub-pixel 21 of the displayed color is 255, and the intensity of the red sub-pixel 21 is 255+y'. This method has been used in Figure 7aThis is schematically shown in FIG, where the color corresponding to the value y' is displayed in the circular surface 31a. It has been assumed here that the values y>0 and Δ>0, but naturally, one or both of them can be negative, and from the display point of view, what is important is whether the sum of the two is positive or negative. If the value y' is greater than +50 or less than -50, then the rules continue to apply within the scale when displaying the intensities of the red and green sub-pixels 21. If the value exceeds +50, in other words y'>50, then the intensity of the red sub-pixel 21 is still set to the value 255, while the intensity of the green sub-pixel 21 is reduced to the value 255-y'. If y'<(-50), the intensity of the green sub-pixel 21 remains at the value 255, while the intensity of the red sub-pixel 21 is set to the value 255+y', where a negative number less than (-50) is added to the value 255, in other words, the intensity of the red sub-pixel 21 will be less than 205. With this arrangement, if the person being tested also has normal color vision, he or she will see yellow (neither reddish nor greenish) within the circular surface 31a when the slider 33a is at its center position (y=0) even if there is a display error.

[0119] If only those with the most common color vision deficiencies, namely protanopia and deuteranopia, are being tested, then the red and green color displays on the correction sliders 33b and 33c are sufficient (ignoring any errors in blue).

[0120] Similar to the above, according to the rules explained in the case of adjusting yellow, for each position P of slider 33b, a color corresponding to the value p' = p + Δ is displayed within circular surface 31b. In other words, with this setting, if the person being tested has normal color vision, then even if there is a display error, when slider 33b is at its center position (p = 0), he or she will see a purple hue (with neither a blue tint nor a red tint) within circular surface 31b.

[0121] According to the rules explained in the case of adjusting yellow, for each position T of slider 33c, the color corresponding to the value t'=t+Δ is displayed within circular surface 31c. With this setting, if the person being tested has normal color vision, then even if there is a display error, when slider 33c is at its center position (t=0), he or she will see a cyan color (with neither blue nor green tint) within circular surface 31c.

[0122] Using the above method, the values of y, p, and t set by the patient can serve as the basis for a diagnosis because the display errors on the display 12 have been corrected when the user enters these values.

[0123] Another possibility is that during the color vision measurement, the displayed color itself does not change, but rather the y, p, and t values corresponding to the reading positions Y, P, and T of sliders 33a, 33b, and 33c, respectively, change. The measurement results are then modified according to the formulas y' = y - Δ, p' = p - Δ, and t' = t - Δ, and these values are then used as the basis for diagnosis. Here, the display error Δ is subtracted from the y, p, and t values corresponding to the positions of sliders 33a, 33b, and 33c, because deviations from the zero position are partially or entirely caused by display errors, not by color blindness.

[0124] As mentioned above regarding the calibration method, an embodiment is conceivable in which the rb and bg values are also determined as other components of the display error. In this case, the formulas p' = p + (rb) and t' = t + (bg) can be used to determine the colors displayed at the P and T positions, or, if the display error is not corrected in the display, the formulas p' = p - (rb) and t' = t - (bg) can be calculated during diagnosis. In this way, even very rare S cone failures can be measured.

[0125] If, when performing the calibration, a person with normal color vision and a person with color vision deficiency take the same test, the y2, p2, t2 values for each single setting belonging to the person with color vision deficiency must be corrected with the same y1, p1, t1 values for the person with normal color vision.

[0126] This can also happen in two ways:

[0127] In this case, where the calibration software operates on the same, yet unique, principle as the measurement software, the calibration software's results y1, p1, t1 for positions Y1, P1, and T1, set by a person with normal color vision, are transferred to the measurement software. Instead of displaying the colors corresponding to the values y2, p2, and t2 for positions Y2, P2, and T2 when the patient sets the color, the colors corresponding to the values y2' = y2 + y1, p2' = p2 + y1, and t2' = t2 + y1 are displayed within circular surfaces 31a, 31b, and 31c, according to the aforementioned rules. In this way, display errors are corrected in the same manner as in the previous example by increasing the value Δ = rg, because the position Y1 of slider 33a itself encodes rg values that increase in increments of 1 between -50 and +50, as the red and green intensities vary according to a defined rule, but in such a way that a unit change in y1 always represents a unit change in rg with the same sign. In this case, the basis for diagnosis is the y2, p2, and t2 values set and actually entered by the patient.

[0128] The intensity of the blue sub-pixel 21 may also be taken into account by using the values of p1 and t1 in the above formulas, in other words, y2′=y2+y1, p2′=p2+p1 and t2′=t2+t1.

[0129] However, if the calibration software results y1, p1, t1 for people with normal color vision are only recorded, but the calibration results are not used to modify the measurement software, the final result must be corrected later. If the focus is only on display errors in red and green, the following formula can be used:

[0130] y=y2 y1

[0131] p=p2 y1

[0132] t=t2 y1

[0133] However, if we also want to consider the color blue, we can use the following formula:

[0134] y=y2 y1

[0135] p=p2 p1

[0136] t=t2 t1

[0137] in

[0138] y2, p2, and t2 are values measured for people with color vision deficiency.

[0139] y1, p1, and t1 are settings made by people with normal color vision.

[0140] And the y, p, and t values will be the quantities that serve as the basis for diagnosis.

[0141] In the above example, it is shown how to provide a task for testing color vision by mixing two primary colors (sub-pixel 21 colors). In the present case, the task of the user (patient) is to set a third color formed by mixing the two primary colors. In addition to color setting, the task can also include color selection, which are also collectively referred to as color determination tasks here. Color determination tasks can also be performed using more than two primary colors, for example, the task can be to set various shades mixed from two or more primary colors, or to distinguish them, or to arrange them in order, and so on. In the case of the above example, each color setting task requires setting a shade within the circular surfaces 31a, 31b, 31c belonging to a given task, but in the case of a similar Ishihara test, other embodiments can be conceived in which shapes of various shades are displayed in each task, such as tiny circular colored surfaces. It is obvious to a person skilled in the art that such more complex tasks can be derived from the above examples.

[0142] Color discrimination test

[0143] The present invention also relates to a computer-implemented method for testing color vision. According to a preferred embodiment, the color vision test is a color discrimination test, as will be explained in more detail below. Preferably, the color discrimination test is supplemented by performing a calibration method according to the present invention. The calibration test can be one of the calibration tests described above, the results of which can be used as described above. Alternatively, the calibration test can be the same test as the color discrimination test, as will be explained in more detail later in the disclosure of the color discrimination test. The color discrimination test is displayed on the same display 12 of the same computer 10 under the same given ambient lighting conditions as the calibration test.

[0144] Figure 8 The flowchart of FIGURE 1 shows a preferred embodiment of a color discrimination test. The reference numbers of the steps do not refer to the order of the steps unless otherwise stated or unless this is required by the nature of a given step.

[0145] According to a preferred embodiment, the test is started by displaying instructions on the display 12 of the computer 10 in step 800, which, when followed by the user, cause the user's eyes to adapt to the ambient lighting conditions. These user instructions of step 800 may be displayed when a computer program (e.g., a mobile application) containing a color vision test is started. For example, text instructions may be displayed at the start of the test instructing the user to look at an object 40 that he or she knows to be white, such as a piece of white paper, for several minutes, preferably for at least 2 minutes, and optimally for 10 minutes. The instructions preferably include advice on what ambient lighting conditions are best for performing the color vision test, preferably advising the user to perform the color vision test in daylight, without direct sunlight, such as outdoors or near a window. The instructions preferably also include advice on the color settings of the display, such as advising the user to use the default daylight color setting of the display.

[0146] In step 801, a target color 81 having a target hue 81a, a target brightness 81b, and a target saturation 81c is selected. The target hue 81a and the target saturation 81c together may be referred to as a target chromaticity 81d.

[0147] In step 802, which may be performed before, after, or concurrently with step 801, a test color 82 is selected having a test chromaticity 82a, a test lightness 82b, and a test saturation 82c. Test chromaticity 82a and test saturation 82c together may be referred to as test chromaticity 82d.

[0148] The target color 81 and the test color 82 are selected so that the target color and the test color are pseudo-metameric with respect to the selected dichroism type.

[0149] Dichromacy is a condition in which the eye has only two functional color receptors (cones). Dichromats can match any color they see with a mixture of no more than two pure spectral lights. The type of dichromacy depends on which color receptor is not functioning. If the L cones are not functioning (or if they shift to align with the spectral sensitivity curve of the M cones), the dichromat has protanopia. If the M cones are not functioning (or if they shift to align with the spectral sensitivity curve of the L cones), the dichromat has deuteranopia. If the S cones are not functioning, tritanopia results, but, as mentioned earlier, this is a very rare condition and is not usually permanent.

[0150] Pseudo-metameric colors appear metameric (having the same chromaticity) only to a dichromatic eye; these colors appear different to a person with normal color vision. In the context of the present invention, the term "pseudo-metameric" refers to colors of different chromaticities that appear to have the same chromaticity (metameric) to a dichromatic eye, but appear to have different chromaticities (non-metameric) to normal color vision, i.e., to a subject with normal color vision. As will be explained in more detail later, certain colors with different chromaticities may appear metameric even to a person with normal color vision. For the sake of clarity, such colors are not referred to as pseudo-metameric colors.

[0151] There are various ways to select pseudo-metameric color pairs, such as the Ishihara color palette, which uses pseudo-metameric colors. However, it should be noted that some colors of the Ishihara color palette cannot be reproduced on a computer display 12 because the computer display 12 can only produce a limited chromaticity of colors. Chromaticity is an objective measure of color quality that is independent of the brightness (lightness) of the color. Chromaticity consists of two independent parameters, called hue and saturation.

[0152] Another way to define target color 81 and test color 82 as pseudo-metachromatic colors relative to a selected dichroic type is to select the target color and test color so as to correspond to two colors of different chromaticities that lie substantially on a common confusion line of the selected dichroic type in a chromaticity diagram of a color space.

[0153] In many technical applications, perceived color is represented by coordinates in a color space. Typically, such a space has three dimensions, corresponding to three distinct color properties sufficient to uniquely describe each color. If one of the dimensions corresponds to brightness, it determines the perceived lightness. The remaining two dimensions are then interpreted as chromaticity, defined by the two coordinate values: hue and saturation.

[0154] For example, the well-known CIE chromaticity diagram may be used, where normalized coordinates x and y are calculated from XYZ values of a CIEXYZ color space (eg, the CIE 1931 XYZ color space, however, other CIE XYZ color space standards may also be used). Figure 9 The CIE chromaticity diagram depicted in the chromaticity diagram can be thought of as a plane (of a specific brightness) in the three-dimensional CIE XYZ color space. The visible chromaticity values in this coordinate system form a horseshoe-shaped region with spectrally pure (monochromatic) colors at the curved boundary.

[0155] If in Figure 9 If the colors of sub-pixels 21 of display 12 are represented in a CIE chromaticity diagram, then these colors will correspond to the three vertices of a triangle, and display 12 can only produce colors within this triangle. Therefore, the CIE chromaticity diagram can also be used to visualize the color gamut of display 12 (i.e., the range of colors that can be produced). Figure 9 Two such triangles 91, 92 are shown, corresponding to two different types of displays 12. Triangle 91 corresponds to an LED display 12, and triangle 92 corresponds to the sRGB (standard red, green, and blue) color space of the display 12, which is typically the "default" color space for images that contain no color space information, especially if the image's pixels are stored as 8-bit integers per color channel.

[0156] The white points corresponding to blackbody radiation at different temperatures lie along the so-called Planckian locus 93, which is well known in the art. The Planckian locus 93 is the path that the color of an incandescent blackbody follows in a specific chromaticity space as the blackbody temperature changes. It ranges from deep red at low temperatures to orange, yellow-white, white, and finally to bluish-white at very high temperatures. The most commonly used white point is the isoenergetic white point W corresponding to a 6500K blackbody radiation, located at [x, y] = [1 / 3, 1 / 3]. The CIE standard illuminant D65 white point, denoted by the reference symbol C, corresponds to a blackbody radiation of 6504K, which also roughly corresponds to the average midday light in Western / Northern Europe (including direct sunlight and light scattered from a clear sky), and is therefore also called a daylight illuminant.

[0157] Once the target color 81 and the test color 82 are selected from the common aliasing lines 94, 95, and 96 of the CIE XYZ color space, the coordinates of the colors 81 and 82 can be mathematically converted to RGB coordinates for a given display 12 if the coordinates of the red, green, and blue sub-pixels 21 are known. The red, green, and blue sub-pixels 21 define triangles 91 and 92 of the color space of the display 12 within the CIE chromaticity diagram. Therefore, the preferred embodiment includes the step of obtaining the type of display 12 being used and the coordinates of the red, green, and blue sub-pixels 21 of the given display 12. For example, the type of display 12 can be automatically queried or detected, or can be input by a user via input interface 14, after which the required coordinates can be obtained from a lookup table containing the coordinates of the red, green, and blue sub-pixels 21 for various types of displays 12. If the type of display 12 cannot be determined, or the type is not found in the lookup table, or no such lookup table is available, the coordinates of the sRGB triangle 92 can be used as a good approximation to obtain the RGB coordinates of the color 81 and 82 selected from the common aliasing lines 94, 95, or 96. This will introduce some error, as the colors displayed on a given display 12 using the obtained RGB coordinates will generally not lie exactly on the selected common confusion lines 94, 95, or 96, but rather the displayed colors will correspond to color points in the CIE chromaticity diagram that are only very close to the selected common confusion lines 94, 95, or 96. In the context of the present invention, the chromaticities of the displayed colors obtained by such a conversion are understood to correspond to the chromaticities 81a and 82a of the target color 81 and the test color 82, respectively. The effects of the errors introduced by the imperfect conversion can be compensated for by calibration and / or by combining color discrimination tests with color identification tests, as will be explained later.

[0158] Various other color space representations are also known in the art, in which chromaticity diagrams can be defined in planes. For example, the HSL color space represents color in a cylindrical coordinate system, where brightness is measured along the z-axis, saturation corresponds to the distance r from the z-axis, and chromaticity corresponds to an angle measured relative to the x-axis. In HSL space, each plane perpendicular to the z-axis contains a chromaticity diagram for a given brightness. Different color space representations can be converted into each other.

[0159] Confusion lines are lines in a chromaticity diagram (e.g., the CIE chromaticity diagram) where chromaticities are indistinguishable for the corresponding dichroic properties. Each type of dichroic property has its own confusion line. Figure 10a The protanopia confusion line 94 is shown (L cone dysfunction, protanopia), Figure 10b Deuteranopia confusion line 95 (M cone cell dysfunction) is shown. Figure 10cA tritanopia confusion line 96 (S cone dysfunction) is shown, with a protanopia confusion line 94 and a deuteranopia confusion line 95 also shown passing through a white dot W. As can be seen, each type of confusion line 94, 95, 96 intersects at a common point, the so-called common point.

[0160] If target color 81 and test color 82 lie on (or substantially lie on) a common confusion line in a chromaticity diagram of a color space (e.g., a CIE XYZ chromaticity diagram), this means that the luminance 82b of test color 82 is the same as target luminance 81b. Preferably, the luminance is selected so that target color 81 and test color 82 correspond to two points on the CIE XYZ chromaticity diagram at which the luminance is set to approximately 35% to 65%, more preferably approximately 40% to 60%, for example approximately 50%, so that white point W appears light gray. This selection is more comfortable for the eye than using the CIE XYZ chromaticity diagram at the luminance at which white point W appears white on the screen of a given display 12.

[0161] In the CIE XYZ color space, saturation depends on how close the selected color 81 , 82 is to the border of the chromaticity diagram, as saturation is highest near the border and decreases towards the white point W.

[0162] Preferably, the target color 81 and the test color 82 are selected so that they lie on a common confusion line 94, 95, 96 that passes through the white point W of the chromaticity diagram. To have different chromaticities, the target color 81 and the test color 82 are selected so that they lie on opposite sides of the confusion lines 94, 95, 96 relative to the white point W. This is also possible in color spaces other than the CIE XYZ color space.

[0163] According to a particularly preferred embodiment, the target saturation 81c and the test saturation 82c are equal or nearly equal.

[0164] If color vision testing is performed during the day, the isoenergetic white point W, corresponding to 6,500K blackbody radiation, can be used. This is well-suited for daylight measurements. The standard white point C, a standard white light approximation of sunlight, is also a good choice for daylight measurements. However, for the purposes of this example, the isoenergetic white point W will be used.

[0165] Note that if the ambient lighting conditions differ substantially from daylight, the results of any of the calibration tests described above can be used to determine which color temperature is most appropriate for measurement, and in this case, the white point corresponding to that color temperature can be used. This is because the eye adapts to the ambient lighting conditions, and if the ambient lighting conditions differ significantly from a hypothetical 6,500K blackbody radiation, the adapted eye will perceive a slightly different color as white, thus shifting the white point along the Planckian locus. The location of the actual white point is also affected by the color reproduction capabilities of display 12, which is automatically accounted for when using the results of the calibration tests to define the actual white point.

[0166] Preferably, a white point corresponding to the ambient lighting conditions is used, however, if the subjects are instructed to perform the color vision test in daylight, an isoenergetic white point W corresponding to blackbody radiation of 6,500 K may be used. If any other point around the 6500 K white point is used, the test may still be sufficiently accurate.

[0167] According to the present embodiment, the method is performed under daylight and uses the isoenergetic white point W of the CIE chromaticity diagram corresponding to 6500K blackbody radiation, with coordinates [x, y] = [1 / 3, 1 / 3]. As described above, any other white point in its vicinity (perceived as white by subjects with normal color vision) can be used, which can be defined, for example, by an x, y coordinate pair where [x, y] = [1 / 3 ± 0.05, 1 / 3 ± 0.05].

[0168] Confusion lines 94, 95, and 96, which pass through white point W, are divided in half by white point W. One half is on one side of the confusion line relative to the white point, and the other half is on the other side. If white point W substantially corresponds to the ambient lighting conditions, then the chromaticity is essentially the same on each side, with only the saturation increasing as the position of the dots moves from the white point to the circumference. However, the chromaticity on one side of white point W differs from the chromaticity on the other side. This means that target color 81 is selected from one side, and test color 82 is selected from the other side.

[0169] It should also be noted that the target color 81 and the test color 82 can lie on any confusion lines 94, 95, 96, even confusion lines that do not pass through the white point W. However, in any case, the target chromaticity 81a and the test chromaticity 82 must be different for normal color vision, in the sense that they can be distinguished by a person with normal color vision. As is well known in the art, even colors of different chromaticity can appear indistinguishable to a person with normal vision. This phenomenon is best explained by Figure 11 98. MacAdam ellipse 98 is a region on a chromaticity diagram that includes all colors that a person with normal color vision cannot distinguish from the color at the center of ellipse 98. Therefore, the outline of ellipse 98 represents a just noticeable difference in chromaticity.

[0170] The target color 81 and the test colors 82 are chosen to be pseudo-metameric, meaning that by definition these colors 81, 82 appear metameric only for certain types of dichroism, but these colors 81, 82 have different chromaticities 81a, 82a both in an objective sense and for normal color vision (i.e., the two chromaticities 81a, 82a can be distinguished by a person with normal color vision).

[0171] For example, the CIE chromaticity diagram of the CIE XYZ color space is used as a reference (for example, the CIE 1931 XYZ color space or other CIE XYZ color space standards may be used). The target color 81 and the test color 82 are selected from the common confusion lines 94, 95, and 96 so that the target color 81 and the test color 82 do not lie within the same MacAdam ellipse 98. For example, if the target color 81 and the test color 82 are along the same MacAdam ellipse 98, the target color 81 and the test color 82 are not located within the same MacAdam ellipse 98. Figure 11 If the deuteranopia confusion line 95 is not shown, the test color 82 must lie outside the MacAdam ellipse 98 (specifically, any MacAdam ellipse 98) that contains the target color 81, otherwise even a test subject with normal color vision would not be able to distinguish between the two colors 81, 82. Furthermore, both the target color 81 and the test color 82 must lie within the triangles 91, 92 that define the color gamut of the display 12, otherwise the colors 81, 82 cannot be produced by the given display 82. It should be understood that any other chromaticity diagram may be used, as the confusion lines 94, 95, 96 and the MacAdam ellipse 98 of the CIE chromaticity diagram can be mathematically converted to any other chromaticity diagram.

[0172] Once the target color 81 and the test color 82 are selected, a target image 83 is provided in step 803, and a plurality of test images 84 are provided in step 804. These images 83, 84 may be provided by a database or generated for testing, optionally based on calibration results (e.g., by determining an actual white point W corresponding to the ambient lighting conditions and the color reproduction capabilities of the display 12).

[0173] Figure 12 8 shows an exemplary target image 83 and an exemplary test image 84 on a common image background 87. Image background 87 is preferably a neutral color, meaning low saturation (less than 30%, preferably less than 15%, more preferably close to 0%, and most preferably 0%) and medium brightness (preferably 35-65%, more preferably 45-55%, for example 50%), i.e., gray. The advantage of gray is that when the user's eyes scan the screen of display 12, gray is less likely to cause the user's eyes to adapt to the ambient lighting conditions and then need to readjust.

[0174] Target image 83 includes a set of target markers 85 having a chromaticity corresponding to target chromaticity 81a and an average brightness corresponding to target brightness 81b. Corresponding means that the chromaticity of target markers 85 is equal to or substantially equal to target chromaticity 81a, and the average brightness is equal to or substantially equal to target brightness 81a. Target markers 85 preferably have a saturation corresponding to target saturation 81c of target color 81, i.e., the saturation of target markers 85 is equal to or substantially equal to target saturation 81c. This means that the chromaticity of target markers 85 corresponds to chromaticity 81d of target color 81. Alternatively, only the average saturation of target markers 85 corresponds to chromaticity 81d of target color 81.

[0175] Preferably, the target mark 85 is a fuzzy disc-like shape of random diameter and quasi-random brightness randomly located within the target image 83. The disc-like shape can be a circular disc or other shapes, such as a triangle, square, star, etc. Fuzzy means that the disc-like shape does not have a clear outline, and the boundaries of the shape blend into the common image background 87 in terms of hue, saturation and brightness. Optionally, the target image 85 can have its own background, which is preferably also a neutral color (low saturation and medium brightness) like the common image background 87, for example different shades of gray. Figure 9 In the illustration, target marker 85 is represented by a shaded disk, the direction of the shading represents hue, the angle of the shading represents saturation (larger angles relative to the horizontal correspond to higher saturation), and the density of the shading represents brightness (denser shading corresponds to lower brightness). The blurred outline of target marker 85 is not shown (not shown in this illustration).

[0176] The diameter of each disc-like shape is understood to be the diameter of the smallest circle that encloses the disc-like shape. The random diameter is preferably within a range of + / - 50% of the average diameter, i.e. the smallest diameter is 50% smaller than the average diameter and the largest diameter is 50% larger than the average diameter. A wider or narrower range can also be selected. Preferably, each disc-like shape has a diameter such that when viewed from the center (when it is in the center of the view), it is seen at an angle of 10-20 degrees when viewing the display 12 from a normal distance, the normal distance representing the average distance at which a user typically views a display 12 of a given type (size). Another way to define the diameter is by pixels. According to a preferred embodiment, the average diameter is between 10 and 20 pixels, preferably approximately 15 pixels, and the diameter of each individual target marker 85 is between 5 and 30 pixels. A wider or narrower range can also be selected. Preferably, the disc-like shapes can overlap.

[0177] The brightness of each target mark 85 is quasi-random, meaning it appears random. However, the average brightness within the target image 83 has a preselected value, which is the target brightness 81b. According to a preferred embodiment, the average brightness is calculated as the average of the brightness values of each target mark 85. According to another preferred embodiment, the average brightness is calculated as a weighted average of the brightness values, where the weighting factor is the area of the corresponding target mark 85. The brightness of each target mark 85 is preferably within a range of at least + / - 10% of the target brightness 81b, for example, approximately + / - 30%. That is, the brightness of the darkest target mark 85 can be 30% less than the target brightness 81b, and the brightness of the brightest target mark 85 can be 30% greater than the target brightness 81b. A wider or narrower range, or an asymmetric range, can also be selected. The brightness distribution of the target marks 85 is preferably Gaussian. Maintaining the brightness of each target mark 85 in the target image 83 within a limited range is more relaxing for the eye, compared to a situation where the brightness of the target marks 85 within the target image 83 varies greatly.

[0178] At least one target marker 85 is provided within the target image 83, which means that the set of target markers 85 may consist of a single target marker 85. However, preferably there is more than one target marker 85, preferably there are at least 10, more preferably at least 50 target markers 85.

[0179] Each test image 84 comprises a set of test marks 86, which are preferably also formed by fuzzy disk-like shapes of quasi-random diameter and random brightness randomly located within one of the test images 84. The test marks 86 may be located on a common background 87, or each or some of the test images 84 may have their own background, which, however, is also a neutral color, such as different shades of gray.

[0180] The random diameters of the test marks 86 preferably fall within the same range as the target marks 85, i.e., within a range of + / - 50% of the average diameter, i.e., the smallest diameter is 50% smaller than the average diameter and the largest diameter is 50% larger than the average diameter. A wider or narrower range can also be selected. Preferably, each disc-like shape has a diameter such that when viewed from the center (when it is in the center of the view), it is seen at an angle of 10-20 degrees when the display 12 is viewed from a normal distance, which represents the average distance at which a user typically views a display 12 of a given type (size). Another way to define the diameter is by pixels. According to a preferred embodiment, the average diameter is between 10 and 20 pixels, preferably about 15 pixels, and the diameters of the individual target marks 85 are each between 5 and 30 pixels. A wider or narrower range can also be selected. Preferably, the disc-like shapes can overlap.

[0181] The brightness of each test mark 86 is quasi-random, meaning it appears random; however, the average brightness within each test image 84 has a preselected value. According to a preferred embodiment, the average brightness is calculated as the average of the brightness values of each test mark 86 within a given test image 84. According to another preferred embodiment, the average brightness is calculated as a weighted average of the brightness values, where the weighting factor is the area of the corresponding test mark 86 within a given test image 84. Furthermore, the brightness of each test mark 86 within a given test image 84 is preferably within a range of at least + / - 10%, for example, approximately + / - 30%, of the average brightness of the test marks 86 within the given test image 84. A wider or narrower range, or an asymmetric range, may also be selected. The brightness distribution of the test marks 86 may be, for example, a Gaussian distribution. Maintaining the brightness of each test mark 86 within a given test image 84 within a limited range is more relaxing for the eye, compared to test marks 86 having brightness that varies wildly within the same test image 84.

[0182] At least one test mark 86 is provided within each test image 84, which means that the set of test marks 86 within each test image 84 may consist of a single test mark 86. However, preferably, there is more than one test mark 86 within each test image 84, preferably at least 10, more preferably at least 50 test marks 86 within each test image 84.

[0183] As follows, test images 84 of varying chromaticity and luminance are provided.

[0184] A first test image 84a is provided, which includes a set of first test marks 86a having a chromaticity corresponding to target chromaticity 81a and an average brightness different from target brightness 81b. First test marks 86a preferably have a saturation corresponding to target saturation 81c of target color 81. This means that the chromaticity of first test marks 86a corresponds to chromaticity 81d of target color 81. Alternatively, only the average saturation of first test marks 86a corresponds to chromaticity 81d of target color 81.

[0185] Second test image 84b is composed of a group of second test marks 86b having a chromaticity corresponding to test chromaticity 82a and an average luminance corresponding to target luminance 81b. Second test marks 86b preferably have a saturation corresponding to test saturation 82c of test color 82, wherein the chromaticity of second test marks 86b corresponds to chromaticity 82d of test color 82. Alternatively, only the average saturation of second test marks 86b corresponds to chromaticity 82d of test color 82.

[0186] Any number of third test images 84c may be provided, including zero, but preferably at least one, more preferably at least two, and even more preferably at least four, third test images 84c are provided to reduce the probability of accidentally selecting the correct image, as will be explained later. Each third test image 84c includes a set of third test marks 86c having a chromaticity corresponding to test chromaticity 82a and an average luminance different from target luminance 81b. Third test marks 86c preferably have a saturation corresponding to test saturation 82c of test color 82, with the chromaticity of third test marks 86c corresponding to chromaticity 82d of test color 82. Optionally, only the average saturation of third test marks 86c of each third test image 84a corresponds to chromaticity 82d of test color 82.

[0187] An additional test image 84 (not shown) consisting of additional test marks 86 having a different hue and / or saturation than the target image 81 and the test image 82, and an average brightness of any value may be used.

[0188] In step 805, target image 83 and test image 84 are displayed such that target image 83 is displayed in a continuous target region 88 (e.g., a circle) of display 12, and multiple test images 84 are displayed in multiple continuous test regions 89 (e.g., a circle) of display 12. These continuous test regions 89 are distinct from continuous target region 88 and from each other. Preferably, target region 88 and test regions 89 are also spaced apart from each other. Preferably, target region 88 is centered and test regions 89 are positioned around target region 88, equidistant from target region 88 and from each other. First, second, and third test images 84a, 84b, 84c are preferably randomly arranged within multiple test regions 89 (one test image 84 in each test region 89).

[0189] By displaying the target and test images 83, 84 in this manner, target and test marks 85, 86 having different chromaticities 81a, 82a and significantly different brightnesses are separated from one another so that there are no abrupt transitions in chromaticity or brightness between closely spaced marks 85, 86 (arranged within the same image 83, 84). Consequently, the displayed test of the present invention does not have the disturbing vibration effect of a stone prototype test. Furthermore, the neutral (preferably gray) background 87 helps to mitigate the varying neural stimulation produced by changes in chromaticity and brightness as the user's gaze scans from one image 83, 84 to the other. These measures effectively reduce the chances of triggering an epileptic seizure compared to conventional stone prototype tests.

[0190] In step 806, instructions (specifically, text instructions) are displayed, instructing the user (test subject) to select one of the multiple test images 84 based on which one has the highest chromatic similarity to the target image 83. Preferably, the user is instructed to ignore any differences in brightness and / or saturation. Step 806 may precede step 805, or the target image 83, test image 84, and user instructions may be displayed simultaneously. User instructions related to the color discrimination task and user instructions related to the user's eye accommodation may also be displayed, so that steps 800 and 806 do not need to be distinct from each other.

[0191] Image selection may be performed in any known manner using one of the user input interfaces 14, for example by allowing the user to select one of the test images 84 by clicking on a corresponding test area 89 of the touch screen 13, or by allowing the user to select one of the test images 84 by clicking with a mouse 15a, 15b, 15b' etc.

[0192] In step 807 , user input related to the selection of the selected test image 84 is read in via the input interface 14 of the computer 10 .

[0193] In step 808, the test is evaluated based on the user's selection, such that if the selected test image 84 corresponds to the first test image 84a, the selection is determined to be correct, and if the selected test image 84 corresponds to one of the second or third test images 84a, 84b, the selection is determined to be incorrect. An incorrect selection typically indicates a disturbance in the type of cones affected by the selected dichromacy type. For example, if the selected dichromacy type is protanopia, the affected cones are L cones, and color vision defects are associated with disturbances in L cone cells. Generally, this means red-green deficiency, and in rare cases, this condition can be red-green blindness (which can be considered an extreme case of red-green deficiency). Note that if the selection is determined to be correct, this is understood to mean that the test subject (user) can distinguish between the target color 81 and the test color 82, and therefore has normal color vision for both colors 81, 82. This does not necessarily mean normal color vision. The user may have other types of color vision deficiency (e.g., if testing for red-tingedness, they may also have green-tingedness or, more rarely, tritanomania), or the user's color vision deficiency may be less severe (e.g., if testing for red-tingedness), which is not apparent in selecting target color 81 and test color 82, but is detectable using another pair of target color 81 and test color 82. It should also be noted that statistically, there is a probability that the user may accidentally select first test image 84a (i.e., not actually being able to distinguish between target chromaticity 81a and test chromaticity 82a). This statistical error can be significantly reduced by using second test image 84b, as people with color vision deficiencies tend to rely on brightness to distinguish colors they easily confuse. Thus, for a user with a given type of color vision deficiency, second test image 84b, having the same average brightness as target image 83, will appear more similar to target image 83 than first test image 84a, having a different average brightness.

[0194] The evaluation step 808 can be performed by the computer 10, on which the test is performed, or the test results can be sent to a remote computer (e.g., a server) via a data communication channel established on the Internet or other communication networks (e.g., GSM), and step 808 can be performed by software running on the remote computer. In the context of the present invention, the remote computer can be provided in the form of a cloud server, and step 808 can be performed by any one or more computers constituting the cloud server.

[0195] The severity of the color vision deficiency can be determined based on the distance between the target color 81 and the test color 82 measured along a common confusion line 94, 95, or 96 for the target color 81 and the test color 82 (the target color 81 and the test color 82 have been selected to lie substantially on the common confusion line 94, 95, or 96). The distance indicates a certain degree of severity. A severe case of color vision deficiency means that the user cannot even distinguish between colors 81 and 82 that are far apart on the common confusion line 94, 95, or 96. If the target color 81 and the test color 82 are far apart (thus making the correct selection for this pair of colors 81, 82, indicating normal vision), the user may correctly identify the first test image 84a, indicating a good condition. However, if the target color 81 and the test color 82 are close together, the user may select one of the second image 84b or the third image 84c, which is an incorrect selection for these two colors 81, 82, indicating a color vision deficiency.

[0196] Therefore, in order to determine the severity of color vision deficiency, it is preferable to conduct multiple tests on the same user (test subject).

[0197] According to a preferred embodiment, this is accomplished by repeating the above steps with a plurality of color pairs 80 of a target color 81 and a test color 82, wherein three color pairs 80 are Figure 13 . In this embodiment, all color pairs 80 are selected so that both the target color 81 and the test color 82 lie on a protanopic confusion line 94 that passes through a white point W. Some target colors 81 are selected to lie on one side of the confusion line 94 relative to the white point W, while other target colors 81 are selected to lie on the other side of the confusion line 94 relative to the white point W. The test color 82 of each pair 80 is selected from the side opposite the target color 81 of a given color pair 80, so that the test chromaticity 82a is always different from the target chromaticity 81a. Preferably, the test color 82 of each color pair 80 is selected to have a test saturation 82c that is similar to the target saturation 81c of the corresponding target color 81. As can be seen, the distance between the target color 81 and the test color 82 measured along the common confusion line 94 of the target color 81 and the test color 82 is different for each depicted color pair 80. This distance is referred to as the color distance. In this embodiment, the color distance is determined by the target saturation 81c and the test saturation 82c, the higher the two saturations 81c, 82c, the greater the color distance (this is not necessarily the case if other protanopia confusion lines 94 are used).

[0198] The test is repeated using multiple color pairs 81, 82, so that a new target image 83, a new first test image 84a, a new second test image 84b, and any number of new third test images 84c are provided in each test based on a given target color 81 and test color 82. Note that when the test is repeated, step 806 of displaying user instructions may be performed only once, for example, at the beginning of the first test in the test series. According to another preferred embodiment, the user instructions are permanently displayed when images 83, 84 are changed in each test.

[0199] To increase the reliability of the test, more than one test can be generated using the same color pair 80, so that more than one test has the same color distance between its target color 81 and test color 82. Successive pairs of target color 81 and test color 82 for each test can have increasing color distances along a common protanopia confusion line 94, or the order of the tests can follow any other rule, or can be randomized with respect to the distances between the target color 81 and test color 82 of the color pair. Images 83, 84a, 84b, 84c are displayed, and the user is prompted to select from the test images 84 based on their colorimetric similarity to a given target image 83.

[0200] Multiple tests are evaluated based on all correct and incorrect user selections, with a correct selection being the one in which the first test image 84a is selected by the user, and every other selection being an incorrect selection. The severity of the color vision deficiency (in this case, protanopia) can be determined based on the number of incorrect selections (correct selections may also be considered). Because users with a certain degree of protanopia will confuse chromaticities below a critical color distance, all tests using color pairs 80 with color distances below that critical color distance will result in incorrect selections, with some correct selections having a statistically low but non-zero probability. Note that the correct selection may be due to pure luck, meaning that the user cannot see the difference in chromaticity and is simply guessing from the six possible answers (if the four third test images 84c are displayed). There are a number of ways to eliminate this risk, one being to use a second test image 84b with the same average luminance as the target image 83. Another way to reduce the probability of accidentally correct selections is to increase the number of tests, including the possibility of multiple tests at the same color distance.

[0201] Preferably, the severity of color vision deficiency (in this case, protanomaly) is classified as mild, moderate, and severe based on the number of incorrect answer thresholds provided for each category. For example, in a test series consisting of 20 tests for testing protanomaly, in which the color distances applied vary uniformly within a range starting from the minimum color distance discernible by a person with normal color vision (corresponding to the width of the MacAdam ellipse 29 at the selected white point W) and the maximum color distance possible along a given confusion line 94 (corresponding to the maximum saturation of the target color 81 and the test color 82), thresholds can be defined such that 3 to 8 incorrect choices are classified as mild protanomaly, 9 to 14 incorrect choices as moderate protanomaly, and 15 to 20 incorrect choices as severe protanomaly. Classification can also be performed based on the ratio of incorrect to correct choices.

[0202] The severity of a color vision deficiency can be determined using a more complex function that correlates with the number of incorrect selections. Specifically, the severity of a color vision deficiency can be determined by considering the distance between the target color 81 and the test color 82 in those color pairs 80 for which an incorrect selection was made. For example, multiple tests can be evaluated based on a weighted percentage of incorrect user selections, with the weighting factor corresponding to the color distance between the target color 81 and the test color 82 in the corresponding color pair 80.

[0203] The above test series can be performed with color pairs 80 selected to lie along the deuteranopia confusion line 95 passing through the white point W, as shown in FIG. Figure 14 This test series determines whether the user suffers from deuteranomaly.

[0204] Similarly, one of the tritanopia confusion lines 96 may be used in this test series.

[0205] More than one color vision deficiency can be tested in a single test series. According to a preferred embodiment, a first plurality of color pairs 80 of target colors 81 and test colors 82 are selected such that the common confusion line is the protanopia confusion line 84, and a second plurality of color pairs 80 of target colors 81 and test colors 82 are selected such that the common confusion line is the deuteranopia confusion line 95. The test is repeated using the first plurality of color pairs 80 and the second plurality of color pairs 80. Multiple tests are evaluated based on the number of incorrect user selections within the first and second plurality of color pairs 80. If the number of incorrect user selections within the first set exceeds the number of incorrect user selections within the second set by a given first threshold, a color vision deficiency associated with a disorder of the L cones is determined; if the number of incorrect user selections within the second set exceeds the number of incorrect user selections within the first set by a given second threshold, a color vision deficiency associated with a disorder of the M cones is determined. The first and second thresholds can be determined based on calibration for users with normal color vision, as will be explained later. Without calibration, both the first and second thresholds are at least 1, and preferably 2 to 3.

[0206] According to another preferred embodiment, a third plurality of color pairs 80 of target colors 81 and test colors 82 are selected such that the common confusion line is the tritanopia confusion line 96, e.g. Figure 10c The confusion line depicted in FIG passes through the white dot W. The test is repeated using a third plurality of color pairs 80, and the plurality of tests in the third set are evaluated by determining the number of incorrect user selections in the third set. The number of incorrect user selections in the third set is preferably used to calibrate the evaluation of the plurality of tests in the first and second sets. Because color vision defects in the S cones are extremely rare, the number of incorrect responses indicates the user's general sensitivity to color differences. Return to Reference Figure 11 , wherein the depicted MacAdam ellipse 98 is determined for average color vision. However, each person's MacAdam ellipse 98 may differ from the average based on their individual ability to discern color. Therefore, the incorrect selections in the third group indicate that the user is less able to distinguish different colors than the average person, and the results obtained for the first and second groups can be corrected using this information. For example, the number of incorrect selections in the third group can be subtracted from the number of incorrect selections in the first and second groups.

[0207] However, if the number of incorrect choices in the third group is found to exceed the number of incorrect choices in the first and second groups, this may indicate a very rare color vision defect associated with S cones (tritanomania).

[0208] The color discrimination test described above can also be used for calibration purposes. When performing a color discrimination test as a calibration test, at least one, preferably two, people with normal color vision are invited to perform (one after the other) a color discrimination test on the same display 12 that will be used to test subjects with color vision deficiency and under the same ambient lighting conditions, which are preferably daylight (outdoors without direct sunlight, or indoors near a window but without direct sunlight). The calibration test can be performed before or after the test subject performs the color discrimination test. The calibration test preferably starts with displaying instructions on the display 12 that, when followed by the user, will ensure these conditions. Preferably, instructions are also displayed that, when followed by the user, will ensure that the eyes adapt to the ambient lighting conditions, as described above.

[0209] If a single test is performed, i.e. a single color pair 80 is selected, the selection made by the user performing the calibration test is used to validate the test itself: if a user with normal color vision makes an incorrect selection, the test is preferably discarded, since an incorrect selection does not indicate any color vision deficiency.

[0210] If a series of tests are performed, the number of incorrect choices in each group (corresponding to the protanopia, deuteranopia, and tritanopia confusion lines 94, 95, or 96) made by the calibrating user is preferably subtracted from the number of incorrect choices made by the color vision test subject in the corresponding group. Thus, if the user performing the calibration test makes NP incorrect choices in the first (red) group, ND incorrect choices in the second (green) group, and NT incorrect choices in the third (blue) group, then these numbers can be used to normalize the color vision test subject's test results, i.e., the number of incorrect choices P, D, and T in the first (red), second (green), and third (blue) groups, respectively. If the test subject makes fewer incorrect choices in the corresponding group than the calibrating user, then the number of incorrect choices can be considered to be zero. According to these rules, the normalized number of incorrect choices P', D', and T' of incorrect choices by the test subject is:

[0211] If P>NP, P'=P-NP, otherwise P'=0,

[0212] If D>ND, D'=DN D, otherwise D'=0,

[0213] If T>NT, T'=TN T, otherwise T'=0.

[0214] This standardization can affect the severity of the color vision deficiency determined because the number of incorrect choices made by the test subject is reduced to compensate for incorrect choices not due to the color vision deficiency. It can also prevent the display of tests in which users with normal color vision make incorrect choices.

[0215] Preferably, a threshold is set for the normalized number of incorrect selections P', D', and T' for the test subject, and a color vision defect is diagnosed only when one of the normalized numbers P', D', and T' exceeds this threshold (e.g., 2). The type of color vision defect (L cones, M cones, or S cones associated with the color vision defect) is determined based on which normalized number P', D', or T' is highest. Another threshold can be set to determine the minimum difference between the highest normalized number P', D', or T' and the second highest normalized number, e.g., a difference of 2 is required to determine the type of color vision defect. Alternatively, a different threshold difference can be set for each type of color vision defect, e.g., the normalized number of incorrect selections P' in the first (red) group must exceed the normalized number of incorrect selections D' and T' in the second and third groups by a first threshold, and the normalized number of incorrect selections D' in the second (green) group must exceed the normalized number of incorrect selections P' and T' in the first and third groups by a second threshold. Similarly, for cases where no calibration is performed, a threshold for a critical number of incorrect selections and a critical difference between each group can be determined.

[0216] Note that if color pairs 80 are selected outside any common MacAdam ellipse, the number of incorrect selections NP, ND, NT made by a user with normal color vision is expected to be close to zero and close to equal. If this is not the case, a higher number of incorrect selections during calibration (e.g., more than 2 in each group), or a higher difference in incorrect selections between different groups during calibration (e.g., a difference of more than 2) typically indicates poor ambient lighting conditions or poor color reproduction capabilities of display 12, or that the user for whom calibration is being performed and who is believed to have normal color vision suffers from a color vision deficiency.

[0217] Therefore, also for calibration purposes, a critical number of incorrect selections (e.g., 2 out of 60 tests) and a critical difference in incorrect selections between different groups (e.g., a difference of 2) can be determined, so that if the number of incorrect selections or the number of differences exceeds the predetermined critical number, this indicates that the user does not have normal color vision, or the color reproduction capability of display 12 is too poor, or the ambient lighting conditions are not suitable for color vision measurement. In this case, the calibration test results are preferably discarded and the possible source of error is displayed to the user.

[0218] According to another preferred embodiment, a series of tests is provided in all three groups (red, green, blue), and the number T of incorrect choices made by the color vision test subject in the third group of tests (blue) is used to calibrate the measurements in the first and second groups (red, green), for example by subtracting the number T of incorrect choices in the third group from the number P of incorrect choices in the first group (red) and the number D of incorrect choices in the second group (green). This type of calibration can be used in addition to or as an alternative to calibration performed by a user with normal color vision. The reason for this is that, for the third group of tests (blue), almost any test subject can be considered to have normal color vision, since any degree of tritanomania is extremely rare.

[0219] As previously mentioned, due to the inaccurate conversion of the R, G, and B values of the red, green, and blue sub-pixels 21 of the target color 81 and the test color 82, which are selected from the common confusion lines 94, 95, and 96 of the chromaticity diagram, the measurement results may contain some error. Generally, the further the selected colors 81 and 82 are from the white point, the greater this error. Therefore, the error caused by the inaccurate coordinate conversion can make it difficult to distinguish between cases of severe and moderate color vision deficiency. Note that other factors also contribute to a degree of uncertainty when distinguishing between severe and moderate cases. This can be improved by applying the following color recognition test, which is specifically designed to detect severe color vision deficiency.

[0220] The color discrimination test is implemented in the form of one or more computer programs. The computer program is software containing computer instructions that, when executed by the processor 16 of the computer 10, causes the computer 10 to perform some or all of the steps of the color discrimination test. Because the term "computer" is broadly defined, the term "computer program" also includes various software, such as desktop computer programs, tablet and smartphone applications, etc.

[0221] The computer program may be dedicated software or may be a web application (e.g., a JAVA script application) that is run by a web browser on the computer 10. According to a preferred embodiment, only the steps preceding the assessment test are performed by the computer program running on the user's computer 10, and the steps of the assessment test are performed by a separate computer (e.g., a remote server or a cloud server) running remotely from the computer 10.

[0222] The invention also relates to a computer 10 having such a computer program stored in its non-volatile storage medium 17 .

[0223] Color recognition test

[0224] The present invention also relates to a computer-implemented method for testing color vision by means of a color recognition test. Preferably, the color recognition test is supplemented by performing a calibration test according to the present invention. The calibration test may be Figures 3 to 7a One of the aforementioned calibration tests is shown, the results of which can be used as previously described. According to a preferred embodiment, the calibration test is the same test as the color recognition test, which will be explained in more detail following the disclosure of the color recognition test. If the calibration test supplements the color recognition test, the color recognition test is displayed on the same display 12 of the same computer 10 under the same given ambient lighting conditions as the calibration test.

[0225] Now refer to Figures 15 to 16 Explain the color recognition test.

[0226] Figure 15 The flowchart of FIGURE 1 shows a preferred embodiment of a color recognition test. The reference numbers of the steps do not imply an order of the steps unless otherwise stated or unless this is required by the nature of a given step.

[0227] According to a preferred embodiment, before or at the start of the color recognition test, one or more instructions are displayed that, when executed by the user, ensure that the user's eyes adapt to the given ambient lighting conditions, as explained in conjunction with the calibration test. For example, text instructions may be displayed at the start of the test instructing the user to look at an object 40 that he or she knows to be white, such as a piece of white paper, for several minutes, preferably for at least 2 minutes, and optimally for 10 minutes.

[0228] According to a preferred embodiment, the test is initiated by displaying instructions on the display 12 of the computer 10 in step 900, which, when followed by the user, cause the user's eyes to adapt to the ambient lighting conditions. The user instructions for step 900 may be displayed when a computer program (e.g., a mobile application) containing a color recognition test is launched. For example, text instructions may be displayed at the start of the color recognition test, instructing the user to look at an object 40 that he or she knows to be white, such as a piece of white paper, for several minutes, preferably for at least 2 minutes, and optimally for 10 minutes. The instructions preferably include advice on what ambient lighting conditions are best for performing the color recognition test, preferably advising the user to perform the color recognition test in daylight, without direct sunlight, such as outside or near a window. The instructions also preferably include advice regarding the color settings of the display, such as advising the user to use the display's default daylight color setting.

[0229] In step 901, a plurality of color samples 101 are provided, each color sample 101 being within the color gamut of the display 12 and having a chromaticity between 0 and 220. In practice, the color samples 101 are provided by selecting properties of the color samples 101, such as hue, saturation, and brightness, or intensity values of sub-pixels 21 of pixels 20 of a given display 12. Note that chromaticity is also defined by the intensity values of the sub-pixels 21 relative to each other and corresponds to the dominant wavelength when displayed.

[0230] The color sample 101 may be generated, for example, by the computer 10 , or the color sample may be stored in and retrieved from a database, for example.

[0231] In step 902, a plurality of color samples are displayed on the display 12 of the computer 10 by setting the hue, saturation, and brightness corresponding to the intensity values of the sub-pixels 21 of the pixels 20 of the display 12 so as to obtain the color of each color sample 101. The intensity values of the sub-pixels 21 of the pixels 20 that make up the color sample 101 are set so that the chroma of the color sample 101 has a value between 0 and 220 on a standard scale of 0 to 255. Note that chroma is usually expressed in degrees from 0 to 360 degrees. In this representation, it is preferred to use only the chroma between 0 and 310 degrees because the chroma range between 310 and 360 degrees has no equivalent in the visible spectrum and can only be simulated by mixing high-frequency light (blue or violet) with red light, which results in other colors between magenta and red on the violet line. In the context of the present invention, whenever chroma is given as a number rather than degrees, the value is understood to be relative to the chroma scale of 0 to 255.

[0232] The brightness of each color sample 101 is preferably between 25% and 75%, more preferably between 40% and 60%, and the saturation is preferably between 25% and 90%, more preferably between 40% and 80%. Reducing the saturation increases the difficulty of the color identification test. If the brightness increases above 75% or decreases below 25%, the difficulty level also increases. Note that saturation and brightness can also be expressed using a scale of 0 to 255, in which case the saturation and brightness values are the percentages given above multiplied by 256 minus 1.

[0233] When displayed, color sample 101 may have any shape. Figure 16 An exemplary screenshot of display 12 is schematically shown, in which color samples 101 are displayed as disk-shaped areas against a neutral background 100, meaning low saturation (near zero, preferably less than 30%, more preferably less than 15%) and medium brightness (preferably 35-65%, more preferably 45-55%, for example 50%), appearing as a light grey shade. This has the advantage that the grey colour is less likely to cause the user's eyes to need to readjust after adapting to the ambient lighting conditions as the user's gaze scans across the screen of display 12. Preferably, the position of each color sample 101 on the neutral background 100 is randomly selected. This helps to reduce the negative impact of any residual images of previously displayed color samples 101, because, on the one hand, the color samples 101 are subsequently displayed at different (random) positions, and, on the other hand, the user's gaze must move from the position of the previous color sample 101 to the position of the subsequent color sample 101 on the neutral background 100, which helps to neutralize residual images.

[0234] In step 902, a group of identifiers 102 corresponding to different colors (shades) is also displayed. Preferably, the group includes identifiers 102 corresponding to the colors "purple", "blue", "cyan", "white", "green", "yellow", "orange" and "red". The identifiers 102 may actually be the color names "purple", "blue", "cyan", "white", "green", "yellow", "orange" and "red". It will be understood that the color names may be in any suitable language, such as a language selected by the user. The identifiers 102 may correspond to colors in other ways, for example, the identifiers 102 may be abbreviations of the color names "purple", "blue", "cyan", "white", "green", "yellow", "orange" and "red", for example, "P", "B", "T", "W", "G", "Y", "O" and "R" may represent "purple", "blue", "cyan", "white", "green", "yellow", "orange" and "red" respectively, as Figure 16 As shown. The identifier 102 may also correspond to a given color in other ways. In the context of the present invention, the displayed identifier 102 may be presented in a non-text form. For example, the identifier 102 may be in the form of a symbol or image indicating an object that is typically associated with the colors "purple", "blue", "turquoise", "white", "green", "yellow", "orange", and "red". For example, an image of purple may be used in association with the color "purple", an image of water may be used in association with the color "blue", an image of turquoise (mineral) may be used in association with the color "turquoise", an image of a snowman may be used for the color "white", an image of grass may be used for the color "green", an image of an orange may be used for the color "orange", and an image of fire may be used for the color "red". Other color names may be displayed in addition, such as "black", "grey". According to a preferred embodiment, only the names of pure colors are displayed, which means that the names of mixed colors are not displayed (i.e., there is no "leaning ..." color name, such as "leaning greenish yellow"). However, if the names of mixed colors are also displayed, the difficulty of the test will increase. In this case, mixed colors are preferably considered to be "leaning ..." colors, such as "leaning greenish yellow" will be considered to be "yellow".

[0235] In step 903, instructions 103 are displayed instructing the user to select, for each color sample 101, an identifier 102 from the set of identifiers corresponding to the color (chroma) of the color sample 101. Note that chroma is a color appearance parameter of a color, whereby this set of names may be considered to correspond to chroma, or they may equally be considered to correspond to a color. Most users identify chroma (i.e., a color appearance parameter) with the color itself, so to make the instructions 103 more comprehensive, it may refer to selecting a "color" rather than a "chroma". Therefore, the terms "color" and "chroma" are used as synonyms in the description, as long as it is clear that only the color appearance property of the color is relevant. The instructions 103 may be displayed together with the color sample 101 and the identifier 102, as shown in FIG. Figure 16 As shown, or instruction 103 may appear separately, for example before any color sample 101 is displayed, ie before step 902 .

[0236] The color sample 101 , identifier 102 , and instructions are preferably all displayed on a neutral background 100 .

[0237] More than one color sample 101 may be displayed simultaneously, however, preferably, the color samples 101 are displayed individually, one at a time, e.g. Figure 16 As shown, the user thus has no comparison when selecting the corresponding identifier 102 , which makes the task more difficult for users with color vision deficiencies.

[0238] If a single color sample 101 is displayed (e.g. Figure 16 ), the user is then allowed to select one of the displayed identifiers 102 via the input interface 14, the selection is read in step 904, and the subsequent color samples 101 are then displayed. Preferably, the color samples 101 are displayed in a random or quasi-random order according to their chromaticity, which means that the chromaticity of the color samples 101 is not in an increasing or decreasing order when presented. In the event that the user cannot recognize the color (chromaticity), the user may also be allowed to skip the color sample 101.

[0239] If more than one color sample 101 is displayed simultaneously, it is preferred to allow selection of each color sample 101 via the user input interface 14, or to automatically select one of the displayed color samples 101 so as to allow the user to select one of the displayed identifiers 102 for the selected color sample 101. After selecting an identifier 102, subsequent color samples 101 are selected automatically or by the user. When all displayed color samples 101 have been selected and the user has selected a corresponding identifier 102, a new image containing previously unshown color samples 101 (if any) is displayed.

[0240] In step 906, the color recognition test is evaluated. This can be performed on the same computer 10 that performs the test, or the data obtained related to the user's selection can be transmitted to a remote computer via any conventional data communication channel established on the Internet or other communication network (e.g., GSM), in which case step 808 is performed by software running on the remote computer. In the context of the present invention, the remote computer can be provided in the form of a cloud, and step 808 can be performed by any one or more computers that constitute the cloud.

[0241] Evaluation of the test includes determining whether the user has made any diagnostically significant selections. The term "diagnostically significant selections" is used herein to refer to selections where:

[0242] The identifier 102 corresponding to the color "orange" or the color "red" is selected for display color samples 101 having chromaticity values between 60 and 220; or

[0243] selecting the identifier 102 corresponding to the color "green" for a color sample 101 having a chroma value between 0 and 35; or

[0244] selecting the identifier 102 corresponding to the color "white" for the color sample 101 having a chromaticity value between 105 and 150; or

[0245] selecting the identifier 102 corresponding to the color "purple" for the color sample 101 having a chroma value between 105 and 135; or

[0246] For a color sample 101 having a chroma value between 190 and 220, an identifier 102 corresponding to the color "cyan" is selected.

[0247] The number of diagnostically significant options can be used to determine whether the user has severe color vision deficiency, a condition bordering on color blindness or, in extreme cases, color blindness. The type of color blindness can be any type, i.e., red-green color blindness, or tritanopia. To determine severe color vision deficiency, it is usually sufficient to identify at least one diagnostically significant option.

[0248] It is also possible to test the user in only one or a few color bands where users with severe color vision deficiency are more likely to make choices that are important for diagnosis. Preferably, the following color bands are relevant:

[0249] The relevant first chromaticity band lies between 60 and 105;

[0250] The associated second chromaticity band lies between 0 and 35;

[0251] The associated third chromaticity band lies between 105 and 135; and

[0252] The associated fourth chromaticity band lies between 190 and 220.

[0253] One or more of these associated chromaticity bands may be used to select the chromaticity of each color sample 101. Chromaticities other than those in these chromaticity bands may also be used as the chromaticity of additional color samples 101.

[0254] Preferably, to perform a color identification test, at least two, and more preferably at least three, color samples 101 are provided within each relevant chromaticity band, such that the chromaticities of the at least two, and more preferably at least three, color samples 101 divide the given chromaticity band into band portions of equal width, with each band portion containing the chromaticity of one of the selected color samples 101. This means that if two color samples 101 are provided within a given chromaticity band, the chromaticity of one of the two color samples 101 is located in the band portion corresponding to the first half of the given chromaticity band, while the chromaticity of the other of the two color samples 101 is located in the band portion corresponding to the second half of the given chromaticity band. In the case of three color samples 101, the chromaticities of the color samples 101 are located in the first third, second third, and last third of the given chromaticity band, respectively. Note that this provision does not preclude the presence of additional color samples 101 within the same band portion. For example, the chromaticity of the fourth color sample may also be located in the first third of the given chromaticity band. More than one color sample 101 having the same chromaticity may also be provided.

[0255] Preferably, at least 5, more preferably 5 to 15 first color samples 101 are provided, each first color sample 101 having a chromaticity selected from the first chromaticity band, so that the first chromaticity band is divided into at least five first band portions of equal bandwidth, and each first band portion contains at least one selected chromaticity.

[0256] Preferably, at least 5, more preferably 5 to 15 second color samples 101 are provided, each second color sample 101 having a chromaticity selected from the second chromaticity band, so that the second chromaticity band is divided into at least five second band portions of equal bandwidth, and each second band portion contains at least one selected chromaticity.

[0257] Preferably, at least 3, more preferably 4 to 10 third color samples 101 are provided, each third color sample 101 having a dominant wavelength selected from the third chromaticity band, so that the third chromaticity band is divided into at least three second band portions of equal bandwidth, and each third band portion contains at least one selected dominant wavelength.

[0258] Preferably, at least 3, more preferably 4 to 10 fourth color samples 101 are provided, each fourth color sample 101 having a chromaticity selected from the fourth chromaticity band, so that the fourth chromaticity band is divided into at least three second band portions of equal bandwidth, and each fourth band portion contains at least one selected chromaticity.

[0259] According to a preferred embodiment, the color samples do not have a uniform sampling rate across the entire visible spectrum, but rather increase the sampling rate within the chromaticity band of the visible spectrum where users with severe color vision deficiency are more likely to make choices that are important for diagnosis. Here, the color samples 101 are selected more densely (i.e., their respective chromaticities), which means a higher number of color samples per unit chromaticity. The sampling rate can also be increased by providing multiple color samples 101 with the same dominant wavelength. Preferably, the sampling rate for chromaticity values between 60 and 100 and 20 and 35 is higher than the average sampling rate within the measurement range.

[0260] For example, between chroma values 0 to 10 and 140 to 220 (measured on a scale of 0 to 255), a conventional sampling rate of 1 sample per 10 chroma units may be used, between chroma values 135 to 105, an increased sampling rate of 1 sample per 5 chroma units may be applied, and the sampling rate may be further increased by providing the plurality of color samples 101 in increments of 5 chroma units between 15 and 100. The plurality of color samples 101 taken at each selected chroma preferably consists of two or three color samples 101, however more than three color samples 101 may have the same chroma value.

[0261] Unless the color reproduction capability of display 12 is very poor, color samples 101 within the first chromaticity band between 60 and 105 typically have dominant wavelengths that are perceived as green by an observer with normal color vision and whose eyes are adapted to daylight. Color samples 101 within the second chromaticity band between 0 and 35 typically have dominant wavelengths that are perceived as red or orange, color samples 101 within the third chromaticity band between 105 and 135 typically have dominant wavelengths that are perceived as cyan, and color samples 101 within the fourth chromaticity band between 190 and 220 typically have dominant wavelengths that are perceived as purple by an observer with normal color vision and whose eyes are adapted to daylight.

[0262] Even a chromaticity defined by a dominant wavelength that falls into one of the relevant chromaticity bands may be perceived as having a color different from the color normally associated with a given dominant wavelength if the observer's eyes are adapted to ambient lighting that differs significantly from daylight (e.g., the room in which the test is conducted is illuminated using fluorescent lighting).

[0263] If combined Figure 3As explained, by calibrating the color recognition test with users having normal color vision, errors caused by poor color reproduction capabilities of display 12 or by non-ideal ambient lighting conditions can be eliminated.

[0264] According to a particularly preferred embodiment, the calibration test is identical to the color recognition test. In this case, the calibration test can be performed before or after the test subject undergoes the color recognition test. When the color recognition test is used for calibration purposes, Figure 3 Step S101 of displaying instructions corresponds to step 900 of displaying instructions to ensure that the user's eyes are adapted to the ambient lighting conditions. Step S102 of displaying color determination training during calibration testing corresponds to steps 902, 903, 904 of displaying color sample 101, color identifier 102, and user instructions 103. Step S11 of reading user input corresponds to step 905 of reading the selection of color identifier 102 associated with color sample 101. Step S12 of evaluating user input corresponds to step 906.

[0265] In step S13 of the calibration, the display error caused by the combination of the color reproduction capabilities of the display 12 and the effect of the ambient lighting conditions on color vision is determined as the difference between the expected identifier 102 based on the expected dominant wavelength of the chromaticity and the identifier 102 actually selected by the user with normal color vision. Based on the identifier 102 selected by the user, a modification of the color vision test is determined in step S14 so that the selections made by the user (test subject) who may have color vision deficiency are identified based on the responses of the user with normal color vision, regardless of the chromaticity value. Thus, in step 906, the color vision test is evaluated by identifying the selections that are important for diagnosis if:

[0266] selecting an identifier corresponding to the color "orange" or the color "red" for a displayed color sample during a color vision test and selecting an identifier corresponding to the color "green" for a displayed color sample during a calibration test; or

[0267] selecting an identifier corresponding to the color "green" for the color sample during the color vision test and selecting an identifier corresponding to the color "orange" or the color "red" for the color sample during the calibration test; or

[0268] selecting an identifier for the color sample that corresponds to the color "white" or the color "violet" during the color vision test and selecting an identifier for the color sample that corresponds to the color "turquoise" during the calibration test; or

[0269] The identifier corresponding to the color "turquoise" is selected for the color sample during the color vision test, and the identifier corresponding to the color "purple" is selected during the calibration test.

[0270] Additionally, options that are important for diagnosis can be identified if:

[0271] During a color vision test, select the identifier that corresponds to the color "orange" or "red" for the color sample displayed, and during a calibration test, select the identifier that corresponds to the color "turquoise," "blue," or "purple" for the color sample displayed; or

[0272] The identifier corresponding to the color "white" is selected during the color vision test, and the identifier corresponding to the color "blue" is selected during the calibration test.

[0273] Preferably, the expected color of the color sample 101 is considered during the calibration test, thereby defining a critical number of deviations from the expected color within an allowable range (preferably at least 5%, more preferably at least 10%, and even more preferably 10% to 20% of the number of color samples 101), none of which would correspond to a diagnostically significant selection without calibration. In the context of the present invention, the expected color of the color sample is defined as the color at the dominant wavelength of the chromaticity of the color sample calculated using the color gamut triangle 92 of the color space of the sRGB display 12 using CIE standard illuminant D65 as the white point C. If the selection made by the user performing the calibration test exceeds the critical number of allowable deviations, or if the user makes a diagnostically significant selection, this indicates that the user does not have normal color vision, or that the color reproduction capabilities of the display 12 are too poor, or that the ambient lighting conditions are not suitable for color vision measurement. In either case, the results of the calibration test are preferably discarded and the possible sources of error are displayed to the user.

[0274] The color recognition test is implemented in the form of one or more computer programs. A computer program is software containing computer instructions that, when executed by processor 16 of computer 10, causes computer 10 to perform some or all of the steps of the color recognition test. Because the term "computer" is broadly defined, the term "computer program" also includes various software, such as desktop computer programs, tablet and smartphone applications, etc.

[0275] The computer program may be dedicated software or may be a web application (e.g., a JAVA script application) that is run by a web browser on the computer 10. According to a preferred embodiment, only the steps preceding the assessment test are performed by the computer program running on the user's computer 10, and the steps of the assessment test are performed by a separate computer (e.g., a remote server or a cloud server) running remotely from the computer 10.

[0276] The invention also relates to a computer 10 having such a computer program stored in its non-volatile storage medium 17 .

[0277] The color recognition test described above is preferably used to supplement the color discrimination test described above or any other color discrimination test, which uses at least one target mark having a target chromaticity, a set of first test marks, and a set of second test marks, each first test mark having a chromaticity corresponding to the target chromaticity, and each second test mark having a test chromaticity that lies on a common confusion line for cone receptors having the target chromaticity in a chromaticity diagram. For example, the Ishihara test is such a color discrimination test, as well as the color discrimination test described above according to the present invention. It should be understood that any number of additional test marks having chromaticities different from the target chromaticity and the test chromaticity can be used. The brightness of at least one of the target mark and the test mark can also be varied, for example as explained above in conjunction with the color discrimination test of the present invention.

[0278] Color discrimination tests are used to determine whether a test subject has a color vision deficiency in one of the cone receptors (typically L cones or M cones) associated with the color discrimination test. Color recognition tests are used to determine whether the color vision deficiency is severe, i.e., a condition characterized by color blindness or near-color blindness. The inventors of the present invention have discovered that color discrimination tests are generally suitable for reliably determining whether a test subject has a color vision deficiency. However, tests based on chromaticity pairs on a common confusion line (pseudoisomeric chromaticity pairs) are less reliable or fail to distinguish severe cases from moderate and mild cases. This is because color discrimination tests require the test subject to distinguish two chromaticities that lie on a common confusion line on a chromaticity diagram, typically a line that passes through the white point of the chromaticity diagram. The fact that chromaticities that are far apart from each other along the common confusion line are confused indicates a severe case of color vision deficiency. Specifically, if a confusion line that passes through the white point is used, the two chromaticities must be located on opposite sides of the white point, and a severe color vision deficiency is indicated by confusion between chromaticities that are far from the white point and from each other. However, due to the errors associated with displaying theoretically determined chromaticity pairs, the actual chromaticities displayed are only close to the selected confusion lines, rather than lying exactly on them. The further the chromaticities are from the white point, the greater the deviation from the selected confusion lines, which can result in the actual displayed chromaticities no longer being pseudo-isochromatic, whereby a user with severe color vision deficiency (or even color blindness) may be able to distinguish the actual displayed chromaticities, even though the same user cannot distinguish the theoretical chromaticities lying on the common confusion lines. This problem can be overcome by the color recognition test of the present invention, which is designed to detect symptoms of severe color vision deficiency.

[0279] According to a preferred embodiment, the color vision test comprising the color discrimination test and the color identification test is supplemented by a calibration test corresponding to the color identification test.The calibration test is performed as described above and is used to modify the evaluation of the color identification test as described above.

[0280] According to another preferred embodiment, the color vision test is supplemented by a calibration test corresponding to the color discrimination test, thereby subtracting the number of incorrect choices made by users with normal color vision for each cone receptor type from the number of incorrect choices made by the same cone receptor type by test subjects with potential color vision deficiency. For example, if the L cones are tested by the calibration test and the color discrimination test, the number of incorrect responses made by users with normal color vision in the color discrimination task that tests for L cone color vision deficiency is subtracted from the number of incorrect responses made by the test subject in the color vision test for the same color discrimination task. If the calibration test and the color discrimination test also test for M cones, the number of incorrect responses made by users with normal color vision in the color discrimination task that tests for M cone color vision deficiency is subtracted from the number of incorrect responses made by the test subject in the same color discrimination task. In essence, L cone color vision deficiency is tested using chromaticity pairs selected from common protanopia confusion lines, and M cones are tested using chromaticity pairs selected from common deuteranopia confusion lines. Similarly, S cones can be tested with chromatic pairs selected from the common tritanopia confusion lines, however, this color discrimination task is preferably used for calibration purposes because tritanopia is exceptionally rare.

[0281] According to another preferred embodiment, the color recognition test and the color discrimination test are calibrated using calibration tests corresponding to the color recognition test and the color discrimination test as described above.

[0282] Color vision tests, including a combination of color discrimination tests and color identification tests, are also implemented in the form of one or more computer programs. The computer program can be dedicated software or a web application (e.g., a JAVA script application) that is run by a web browser on the computer 10. According to a preferred embodiment, only the steps preceding the evaluation test are performed by the computer program running on the user's computer 10, while the evaluation test steps are performed by a separate computer (e.g., a remote server or a cloud server) running on a remote computer 10. The present invention also relates to a computer 10 that stores such a computer program on its non-volatile storage medium 17.

[0283] Various modifications to the above-disclosed embodiments will be apparent to those skilled in the art without departing from the scope of protection defined by the appended claims.

Claims

1. A computer-implemented method for testing color vision by displaying a color vision test on a display (12) of a computer (10) having at least one input interface (14), the method comprising: - selecting a target color with a target chromaticity and a target luminance and a test color with a test chromaticity such that the target color and the test color correspond to two colors of different chromaticities lying on a common confusion line of a selected dichroic type in a chromaticity diagram of a color space; - providing a target image consisting of a set of target marks, the target marks having a chromaticity corresponding to the target chromaticity and having an average brightness corresponding to the target brightness; - Provides multiple test images, including: - a first test image comprising a set of first test marks having a chromaticity corresponding to the target chromaticity and having an average luminance different from the target luminance; - a second test image comprising a set of second test marks having a chromaticity corresponding to the test chromaticity and having an average luminance corresponding to the target luminance; - any number of third test images, each third test image comprising a set of third test marks having a chromaticity corresponding to the test chromaticity and having an average luminance different from the target luminance; - displaying the target image in a target area of the display (12) and displaying the plurality of test images at a plurality of test areas of the display (12), the plurality of test areas being spaced apart from the target area and the plurality of test areas being spaced apart from each other; - displaying instructions to select one of the plurality of test images based on a highest similarity in colorimetry compared to the target image; - reading in user input related to the selection of the selected test image via at least one input interface (14) of the computer (10); and - evaluating the test such that if the selected test image corresponds to the first test image, the selection is determined to be correct, and if the selected test image corresponds to one of the second test image or the third test image, the selection is determined to be incorrect.

2. The method of claim 1 , comprising selecting the target color and the test color so that the confusion line passes through a white point of the chromaticity diagram and the target color and the test color are on opposite sides of the confusion line relative to the white point.

3. The method of claim 1 , comprising calibrating the color vision test by performing the following calibration test: - displaying instructions for performing said calibration test by a user with normal color vision; - displaying the target image in a target area of the display (12), and displaying the plurality of test images in a plurality of test areas of the display (12); - displaying instructions to select one of the plurality of test images based on a highest similarity in colorimetry compared to the target image; - reading in user input related to the selection of the selected test image via at least one input interface (14) of the computer (10); and evaluating the read user input such that if the selected test image corresponds to one of the second test image or the third test image, a selection error is determined as a result of the calibration test, and if the result of the calibration test is a determination of a selection error, the result of the calibration test is used to modify an evaluation of the color vision test by discarding the erroneous user selection made in the color vision test.

4. The method according to claim 2 or 3, comprising: A plurality of pairs of target and test colors that lie on a common confusion line of a selected dichroism type are selected such that for at least two of the color pairs, the distances between the target and test colors measured along the common confusion line of the selected dichroism type are different, the testing is repeated using the plurality of color pairs, and the plurality of tests is evaluated by determining a number of incorrect user selections.

5. A method according to claim 2 or 3, comprising selecting a plurality of pairs of target and test colors lying on a common line of confusion for a selected type of dichroism such that the distances of the target and test colors measured along the common line of confusion for the selected type of dichroism are different for at least five of the color pairs, repeating the test using a plurality of color pairs, and evaluating the plurality of tests by determining the number of incorrect user selections.

6. The method according to claim 4, comprising: Selecting a first plurality of pairs of target colors and test colors, wherein the selected dichromacy type is protanopia and the common confusion line is the protanopia confusion line, and selecting a second plurality of pairs of target colors and test colors, wherein the selected dichromacy type is deuteranopia and the common confusion line is the deuteranopia confusion line, repeating the testing with the first plurality of color pairs and the second plurality of color pairs, and evaluating the plurality of tests by determining a number of incorrect user selections within the first plurality of color pairs and a number of incorrect user selections within the second plurality of color pairs.

7. The method according to claim 6, comprising: Selecting a third set of multiple pairs of target colors and test colors, wherein the selected dichromacy type is tritanopia and the common confusion line is a tritanopia confusion line, repeating the testing using the third set of multiple color pairs; and evaluating the multiple tests in the third set by determining the number of incorrect user selections in the third set.

8. The method according to claim 7, comprising: The number of erroneous user selections in the third group is used to preferably calibrate multiple tests in the first and second groups by normalizing the number of erroneous user selections in the first group of color pairs by subtracting the number of erroneous user selections in the third group of color pairs from the number of erroneous user selections in the first group of color pairs, and normalizing the number of erroneous user selections in the second group of color pairs by subtracting the number of erroneous user selections in the third group of color pairs from the number of erroneous user selections in the second group of color pairs.

9. The method of claim 6, comprising calibrating the color vision test by performing the following calibration test: - displaying instructions for performing said calibration test by a user with normal color vision; - for each pair of target color and test color, displaying a corresponding target image in a target area of the display (12) and displaying a plurality of corresponding test images in a plurality of test areas of the display (12), wherein the plurality of test areas are spaced apart from the target area and the plurality of test areas are spaced apart from each other; - displaying instructions to select one of the plurality of test images based on a highest similarity in colorimetry compared to the target image; - reading in user input related to the selection of the selected test image via at least one input interface (14) of the computer (10); as well as - evaluating the read user input such that a selection error is determined if the selected test image corresponds to one of the second test image or the third test image; for each color pair lying on a common confusion line for a given dichroic type, determining the number of erroneous selections and using the determined number of erroneous selections to modify the evaluation of a color vision test for the given dichroic type, preferably by normalizing the number of erroneous user selections of the color vision test for the given dichroic type by subtracting the number of erroneous user selections of a calibration test for the given dichroic type from the number of erroneous user selections of the color vision test for the same color pairs.

10. The method according to claim 1, comprising: 10 to 100 target marks are provided as blurred shapes of random diameter and random brightness randomly located within the target image, and 10 to 100 test marks are provided for each test image as blurred shapes of random diameter and random brightness randomly located within a given test image.

11. The method according to claim 1, wherein The brightness of each target mark and each test mark within the second test image is within a range of at least + / - 10% of the target brightness.

12. The method according to claim 1, wherein The brightness of each target mark and each test mark in the second test image is within the range of + / - 30% of the target brightness.

13. A computer-implemented method for displaying a color vision test on a display (12) of a computer (10) having at least one input interface (14), the method comprising: - providing a plurality of color samples, each color sample having a chromaticity between 0 and 220 on a scale of 0 to 255; - displaying the plurality of color samples on the display (12); - displaying a set of identifiers on the display (12), each identifier corresponding to a different color; - displaying on said display (12) instructions for selecting, for each color sample, from the set of displayed identifiers, an identifier corresponding to the chromaticity of the displayed color sample, - reading in, for each color sample, via at least one input interface (14) of the computer (10), a user input associated with the selection of an identifier corresponding to the hue of the color sample.

14. The method of claim 13, comprising displaying one color sample of a plurality of color samples at a time on the display.

15. The method of claim 14, comprising evaluating the color vision test by identifying options that are important for diagnosis if: The identifier corresponding to the color "orange" or the color "red" is selected for display color samples having chromaticity values between 60 and 220; or For color samples with chroma values between 0 and 35, select the identifier corresponding to the color "green"; or For color samples with chroma values between 105 and 150, select the identifier corresponding to the color "white"; or For color samples with chroma values between 105 and 135, select the identifier corresponding to the color "purple"; or For color samples with chroma values between 190 and 220, select the identifier corresponding to the color "turquoise".

16. A method according to claim 14 or 15, comprising providing the color samples at a varying sampling rate within a measurement range of chromaticity values between 0 and 220, such that the sampling rate for chromaticity values between 60 and 100 and 20 and 35 is higher than the average sampling rate within the measurement range.

17. The method of claim 14, comprising calibrating the color vision test by performing the following calibration test: - displaying instructions for performing said calibration test by a user with normal color vision; - displaying the plurality of color samples on the display (12); - displaying the group identifier on the display (12); - displaying on the display (12) instructions for selecting, for each color sample, from the displayed set of identifiers, an identifier corresponding to the color of the displayed color sample, - reading in, for each color sample, via at least one input interface (14) of the computer (10), a user input associated with the selection of an identifier corresponding to the color of the color sample.

18. The method of claim 17, comprising evaluating the color vision test by identifying diagnostically important choices if: an identifier corresponding to the color "orange" or the color "red" was selected for the color sample displayed during the color vision test, and an identifier corresponding to the color "green" was selected for the color sample displayed during the calibration test; or an identifier corresponding to the color "green" was selected for the color sample during the color vision test, and an identifier corresponding to the color "orange" or the color "red" was selected for the color sample during the calibration test; or an identifier corresponding to the color "white" or the color "violet" was selected for the color sample during the color vision test, and an identifier corresponding to the color "turquoise" was selected for the color sample during the calibration test; or An identifier corresponding to the color "turquoise" is selected for the color sample during the color vision test, and an identifier corresponding to the color "purple" is selected for the color sample during the calibration test.

19. The method according to claim 13, comprising: providing at least three first color samples, each first color sample having a first chromaticity selected from a first chromaticity band between 60 and 105, such that the first chromaticity band is divided into at least three first band portions of equal bandwidth, and each first band portion contains at least one selected first chromaticity; - displaying the at least three first color samples on the display (12); - displaying a set of identifiers on the display (12), each identifier corresponding to a different color, including at least one identifier corresponding to the colors "green", "yellow", "orange" and "red"; - displaying on the display (12) instructions for selecting, for each color sample, from the displayed set of identifiers, an identifier corresponding to the color of the displayed color sample; - reading in, for each first color sample, a user input associated with the selection of an identifier corresponding to the color of the displayed first color sample via at least one input interface (14) of the computer (10).

20. The method of claim 19, comprising evaluating the color vision test by identifying a diagnostically significant selection if an identifier corresponding to the color "orange" or the color "red" is selected for the displayed first color sample.

21. The method according to claim 19 or 20 comprises providing at least 5 first color samples, each first color sample having a first chromaticity selected from the first chromaticity band, so that the first chromaticity band is divided into at least five first band portions of equal bandwidth, and each first band portion contains at least one selected first chromaticity.

22. The method according to claim 19 or 20 comprises providing 5 to 15 first color samples, each first color sample having a first chromaticity selected from the first chromaticity band, so that the first chromaticity band is divided into at least five first band portions of equal bandwidth, and each first band portion contains at least one selected first chromaticity.

23. The method according to claim 13, comprising: - providing at least three second color samples, each second color sample having a second chromaticity selected in a second chromaticity band between 0 and 35, such that the second chromaticity band is divided into at least three second band portions of equal bandwidth, and each second band portion contains at least one selected second chromaticity; - displaying the at least three second color samples on the display (12); - displaying a set of identifiers on said display (12); - for each second color sample, reading in, via at least one input interface (14) of the computer (10), a user input associated with the selection of an identifier corresponding to the color of the displayed second color sample.

24. The method of claim 23, comprising evaluating the color vision test by identifying a selection that is diagnostically significant if an identifier corresponding to the color "green" is selected for the displayed second color sample.

25. The method according to claim 23 or 24 includes providing at least 5 second color samples, each second color sample having a second chromaticity selected from the second chromaticity band, so that the second chromaticity band is divided into at least five second band portions of equal bandwidth, and each second band portion contains at least one selected second chromaticity.

26. The method according to claim 23 or 24, comprising providing 5 to 15 second color samples, each second color sample having a second chromaticity selected from the second chromaticity band, so that the second chromaticity band is divided into at least five second band portions of equal bandwidth, and each second band portion contains at least one selected second chromaticity.

27. The method of claim 13, comprising: providing at least two third color samples, each third color sample having a third chromaticity selected from a third chromaticity band between 105 and 135, such that the third chromaticity band is divided into at least two third band portions of equal width, and each third band portion contains at least one selected third chromaticity; - displaying the at least two third color samples on the display (12); - displaying on the display (12) the set of identifiers, including identifiers corresponding to the colors "turquoise", "purple" and "white"; - for each third color sample, reading in, via at least one input interface (14) of the computer (10), a user input associated with the selection of an identifier corresponding to the color of the displayed third color sample.

28. The method of claim 27, comprising evaluating the color vision test by identifying a diagnostically significant selection if an identifier corresponding to the color "white" or the color "purple" is selected for the displayed third color sample.

29. The method of claim 13, comprising: - providing at least two fourth color samples, each sample having a fourth chromaticity selected from a fourth chromaticity band between 190 and 220, such that the fourth chromaticity band is divided into at least two fourth band portions of equal width, and each fourth band portion contains at least one selected fourth chromaticity; - displaying the at least two fourth color samples on the display (12); - displaying on the display (12) the set of identifiers, including identifiers corresponding to the colors "turquoise" and "purple"; - for each fourth color sample, reading in, via at least one input interface (14) of the computer (10), a user input associated with the selection of an identifier corresponding to the color of the displayed fourth color sample.

30. The method of claim 29, comprising evaluating the color vision test by identifying a diagnostically significant selection if an identifier corresponding to the color "turquoise" is selected for the displayed fourth color sample.

31. A computer-implemented method for displaying a color vision test on a display (12) of a computer (10) having at least one input interface (14), the method comprising: A color discrimination test for preparing a color vision deficiency of at least one cone receptor is provided by the following steps: - presenting the color discrimination test using at least one target mark having a target chromaticity, a set of first test marks, and a set of second test marks, wherein each first test mark has a chromaticity corresponding to the target chromaticity and each second test mark has a test chromaticity, the test measure lying on a common confusion line for at least one type of cone receptor having the target chromaticity in a chromaticity diagram; - displaying instructions to solve said color discrimination test; - reading in user input related to a solution of the color discrimination test via at least one input interface (14) of the computer (10); Color recognition testing is provided in the following ways: - Provide multiple color samples, each color sample has a chroma between 0 and 220; - displaying the plurality of color samples on the display (12); - displaying a set of identifiers on the display (12), each identifier corresponding to a different color; - displaying on the display (12) instructions for selecting, for each color sample, from the displayed set of identifiers, an identifier corresponding to the chromaticity of the displayed color sample; - reading in, for each color sample, via at least one input interface (14) of the computer (10), a user input associated with the selection of an identifier corresponding to the color of the color sample.

32. The method of claim 31 , comprising calibrating the color vision test by performing the following calibration test: - displaying instructions for performing said calibration test by a user with normal color vision; - displaying the plurality of color samples on the display (12); - displaying the group identifier on the display (12); - displaying on the display (12) instructions for selecting, for each color sample, from the displayed set of identifiers, an identifier corresponding to the color of the displayed color sample, - reading in, for each color sample, via at least one input interface (14) of the computer (10), a user input associated with the selection of an identifier corresponding to the color of the color sample.

33. A method for calibrating a color vision test for testing color vision under given ambient lighting conditions and for displaying the color vision test on a color display (12) of a computer (10) having at least one input interface (14), characterized in that: displaying a calibration test on the display (12) for a user with normal color vision under given ambient lighting conditions; displaying a color determination task requiring user input as part of the calibration test in at least one measurement area (31) of the display (12), the user input being read via at least one input interface (14) of the computer (10); evaluating the read user input and, as a result of the evaluation, determining a display error caused by a combination of the color reproduction capabilities of the display (12) and the effect of ambient lighting conditions on color perception; as well as Based on the display error, a modification to the color vision test is determined that corrects the color vision test for the display error.

34. The method according to claim 33, characterized in that displaying a color determination task requiring a user to determine, via a user input interface (14) of the computer (10), the intensity of at least a first color sub-pixel selected from red, green, and blue sub-pixels (21) of pixels (20) of the measurement area (31); reading, via the at least one input interface (14), user input related to determination of the intensity of at least a first colored sub-pixel (21); The read user input is evaluated and, as a result of the evaluation, a display error is determined with respect to at least a first colored sub-pixel (21), the display error being a deviation from a first reference color corresponding to the at least first color, the deviation being due to a combination of the color reproduction capabilities of the display (12) and the effect of ambient lighting conditions on the perception of the first color.

35. The method according to claim 34, characterized in that A display color determination task is provided during which a user must set or select the intensity of at least a first colored sub-pixel (21) by using the user input interface (14), and data related to the intensity setting or selection of the at least first colored sub-pixel (21) is read via the at least one input interface (14) and used as user input to set or select the color indicated in the color determination task within the measurement range of the display (12).

36. The method according to claim 34, characterized in that A display color determination task is provided during which a user has to set the intensity of at least the red and green sub-pixels (21) by using the user input interface (14), set white or yellow within a measurement area (31) of the display (12), and during evaluation, the intensity difference of the red and green sub-pixels (21) within the measurement area (31) of the display (12) is determined based on the read setting data and used as a display error.

37. The method according to claim 34, characterized in that A display color determination task is performed during which a user must set purple or cyan within a measurement area (31) of the display (12) by setting the intensities of the red and blue or green and blue sub-pixels (21) using the user input interface (14), and during evaluation, the intensity difference of the red and blue or green and blue sub-pixels (21) within the measurement area (31) of the display (12) is determined based on the read setting data and is used as a display error.

38. The method according to claim 33, wherein During a calibration test prior to performing the color determination task, a person with normal color vision is instructed to view an object (40) known to be white for at least 2 minutes.

39. The method according to claim 33, wherein During a calibration test prior to performing the color determination task, people with normal color vision were given instructions to look at a white paper for at least 2 minutes.

40. The method according to claim 38, characterized in that A display color determination task is provided, during which a user must set the color white corresponding to the color of the object (40) in the measurement area (31) of the display (12) by setting the intensity of at least one red and green sub-pixel (21) selected from the red, green and blue sub-pixels (21) of the pixels of the measurement area (31) using the user input interface (14), setting data related to the intensity of the red and green sub-pixels (21) are read as user input through the at least one input interface (14), and during the evaluation, the intensity difference of the green and red sub-pixels (21) in the measurement area (31) of the display (12) is determined based on the read setting data and is used as a display error.

41. The method according to claim 33, wherein Checking whether the display (12) has sub-pixels (21) of another color besides red, green and blue sub-pixels (21) and, if so, setting their intensity to zero at least within the measurement area (31) during display of the color determination task.

42. A method for testing color vision, during which a color vision test for testing color vision is displayed on a display (12) of a computer (10) having at least one input interface (14) under given ambient lighting conditions, characterized in that A method for performing a calibrated color vision test according to claim 33: Modifications to the color vision tests determined during the calibration method, displaying the color vision test on the display (12), As part of the color vision test, a color determination task requiring user input is displayed in at least one measurement area (31, 31a, 31b, 31c) of the display (12), reading said user input via at least one input interface (14) of said computer, The read user input is evaluated and as a result of the evaluation the color vision of the user performing the color vision test is determined.

43. The method according to claim 42, characterized in that When the color vision test is displayed on the display (12), the modification of the display parameters of the color vision test is determined as a modification of the color vision test to correct the color vision test for the display error, and when the color vision test is modified, the display parameters of the color vision test are modified accordingly, or the modification of the evaluation parameters of the color vision test is determined as a modification of the color vision test to correct the measurement results obtained with the color vision test for the display error when evaluating the color vision test, and when the color vision test is modified, the evaluation parameters of the color vision test are modified accordingly.

44. The method according to claim 42 or 43, characterized in that The calibration test is the same as the color vision test.

45. A computer (10), comprising: A color display (12), at least one input interface (14), at least one processor (16) and a non-volatile storage medium (17) storing a computer program containing computer instructions, the computer program being configured to cause the computer to perform a method according to any one of claims 1 to 44 when the computer instructions are executed by the at least one processor (16).

46. A computer program comprising computer instructions, the computer program being configured to cause a computer (10) to perform a method according to any one of claims 1 to 44 when the computer instructions are executed by at least one processor (16) of the computer (10), the computer (10) having at least one processor (16) operatively connected to a color display (12) and at least one input interface (14).

Images