Binocular visual function system

Abstract

Provided is a binocular visual function system for evaluating binocular single vision. The system comprises: a line-of-sight guiding unit that directs line-of-sight directions of the left eye and the right eye of a subject in a second direction different from a first direction while fixing the head of the subject in the first direction; and a processing unit that presents images spatially separately to the left eye and the right eye of the subject the line-of-sight directions of which have been guided in the second direction by the line-of-sight guiding unit. The images has feature amounts as indices indicating visual characteristics. The feature amount of a left eye image, which is the image presented to the left eye, is different from the feature amount of a right eye image, which is the image presented to the right eye.

Description

Binocular vision function system

[0001] The present disclosure relates to a binocular vision function system.

[0002] In the evaluation of vision correction lenses, measurements based on frontal vision are mainly conducted, and the evaluation of visual characteristics in lateral vision has not been sufficiently carried out at present.

[0003] International Publication No. 2021 / 065247

[0004] According to an aspect of the present disclosure, there is provided a binocular vision function system for evaluating binocular single vision, including a line-of-sight guiding unit that directs the line-of-sight directions of the left and right eyes of a subject to a second direction different from the first direction while fixing the head of the subject in the first direction, and a processing unit that spatially presents images separately to the left and right eyes of the subject whose line-of-sight directions are guided to the second direction by the line-of-sight guiding unit. The images have feature amounts as indicators showing visual characteristics, and the feature amount of the left-eye image, which is the image presented to the left eye, is different from the feature amount of the right-eye image, which is the image presented to the right eye.

[0005] This is a schematic diagram of the binocular vision function system according to this embodiment. This is a schematic diagram of the gaze guidance device according to this embodiment. This is a schematic diagram of the first example of the gaze guidance device according to this embodiment. This is a schematic diagram of the second example of the gaze guidance device according to this embodiment. This is a schematic diagram of the third example of the gaze guidance device according to this embodiment. This is a diagram showing an example of the head fixing part according to this embodiment. This is an example of a screen that displays each gaze point according to this embodiment. This is an example of a schematic block diagram of the information processing device according to this embodiment. These are the functional blocks of the processor according to this embodiment. This is a diagram showing the original image and stimulus image according to this embodiment. This is a diagram showing an example of a stimulus image created using the second example of the image creation method according to this embodiment. This is the calibration processing flow according to this embodiment. This is a diagram illustrating the method of presenting stimulus images in the right eye change processing according to this embodiment. This is a diagram illustrating the method of presenting stimulus images in the left eye change processing according to this embodiment. This is a diagram showing an example of response data according to this embodiment. This is a graph of the posterior probability distribution of the subjective equivalence point μ according to this embodiment. This is a graph showing the posterior probability distribution of the square root of the variance σ according to this embodiment. This is a graph showing the posterior probability distribution of the judgment boundary C according to this embodiment. This is a graph plotting the psychometric function (PF) according to this embodiment. This figure shows the results of measuring the perceptual characteristics of the right eye with the second direction set to 30 degrees according to this embodiment, and the measurement results of measuring the perceptual characteristics of the right and left eyes with the second direction set to 0 degrees. This figure shows the contrast ratio for the contrast image according to this embodiment. This figure shows an example of a contrast image according to this embodiment. This is a graph of the posterior probability distribution of the subjective equivalence point μ according to this embodiment. This is a graph showing the posterior probability distribution of the square root of the variance σ according to this embodiment. This is a graph showing the posterior probability distribution of the judgment boundary C according to this embodiment. This is a graph showing the results of psychophysical measurements according to this embodiment. This figure shows the results of measuring the perceptual characteristics of the right eye with the second direction set to 30 degrees according to this embodiment, and the measurement results of measuring the perceptual characteristics of the right eye with the second direction set to 0 degrees.

[0006] The present invention will be described below through embodiments, but these embodiments are not intended to limit the invention as defined in the claims. Furthermore, not all combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0007] Humans perceive slightly different retinal images in their left and right eyes as a single image in their brains. This phenomenon is called "binocular single vision," and the brain fuses the different images to produce stable vision. The binocular vision system 100 of this embodiment fixes the subject's line of sight at a certain angle and presents images with different characteristics to the left and right eyes for a certain period of time, causing the subject to perceive binocular single vision. Furthermore, the binocular vision system 100 can detect differences in sensitivity between the left and right eyes by acquiring perceptual characteristics such as the degree of perceptual change in the subject's binocular single vision when the feature quantities of the images presented to the left and right eyes are changed. Feature quantities are indicators that show visual characteristics.

[0008] For example, the binocular vision system 100 directs the subject's gaze to a specific angle for a certain period of time and presents images with different features to the left and right eyes. The binocular vision system 100 can measure perceptual characteristics according to the gaze angle by independently changing the feature quantities of the presented images for the left and right eyes. This measurement makes it possible to evaluate the differences between the left and right eyes in the subject's vision.

[0009] Eyeglasses that use corrective lenses may use lenses prescribed based on the wearer's vision test results. When the lenses are fixed in the correct position, the wearer's line of sight passes through the optical center of the lens, providing the appropriate optical properties. However, in reality, eyeglass frames often do not fit the face properly, and the lenses are not fixed in the correct position. Also, deformation of the frame can cause the lens position to shift. In such cases, when the wearer moves their eyeballs and shifts their gaze from side to side or up and down, they will see objects through the periphery of the lens.

[0010] In this regard, the optical properties of the peripheral part of the lens are also important, but prioritizing optical properties results in a thicker lens, sacrificing aesthetics and weight. Furthermore, progressive lenses are spectacle lenses that can focus on near and intermediate vision areas with different curvatures using a single distance lens, but the optical properties of the peripheral part of the lens are poor, causing differences in the perceived image when the line of sight passes through the periphery. For this reason, it is desirable not to use the peripheral part, but in reality, it is almost impossible for wearers to avoid using the peripheral part of the lens in daily life.

[0011] Furthermore, there is a difference in sensitivity between light entering from the center of the pupil and light entering from the periphery of the pupil, causing light distortion in the periphery and altering vision. Also, the wearer's line of sight does not always pass through the optical center of the lens, but often sees objects through other parts of the lens. However, since visual acuity tests are usually performed with the wearer facing forward, evaluation of lateral vision is not taken into consideration.

[0012] In daily life, wearers often look to the side to see objects, but until now, there has been no equipment that performs tests assuming lateral vision. The binocular vision function system 100 according to this embodiment can measure the visual characteristics of a subject in lateral vision and evaluate differences in perception in lateral vision. For example, the binocular vision function system 100 changes feature quantities such as the degree of blur as feature quantities of the image presented to the left and right eyes in lateral vision. Then, by examining how the left and right eyes are perceived in the brain in lateral vision, the binocular vision function system 100 can grasp the differences in sensitivity between the left and right eyes in lateral vision for each individual. How it is perceived in the brain includes, for example, information on which eye's information is preferentially perceived.

[0013] In particular, evaluation of progressive lenses is important when evaluating lenses that require lateral viewing. Due to the structure of progressive lenses, optical properties tend to deteriorate in the peripheral areas of the lens. Progressive lenses are characterized by a special design with different focal points for distance, near, and intermediate vision, and the central part of the progressive lens maintains relatively good optical properties. However, when the gaze is directed to the side, the line of sight passes through the peripheral area of ​​the lens, and optical problems such as aberrations and blurring may occur in this area.

[0014] Furthermore, when wearing progressive lenses and looking to the side, the left and right lenses pass through different peripheral areas, which can easily lead to differences in the feature quantities of visual information reaching each eye. Because this difference can affect the wearer's visual comfort and visibility, evaluating changes in perception during side-viewing and the state of information processing in the brain is important for the appropriate design and individual adjustment of progressive lenses.

[0015] The binocular vision function system 100 according to this embodiment can provide response data for evaluating the extent to which a progressive lens with an asymmetrical design, tailored to a specific subject, reflects the perceptual characteristics of that person's left and right eyes in lateral vision. This makes it possible to design lenses that take into account the perceptual differences between the left and right eyes.

[0016] Figure 1 is a schematic diagram of the binocular vision system 100 according to this embodiment. The binocular vision system 100 comprises a gaze guidance unit 200, a display device 300, and an information processing device 400.

[0017] The gaze guidance unit 200 comprises a gaze guidance device 201 and a head fixing unit 202.

[0018] The gaze guidance device 201 is a device for guiding a subject's gaze direction to a specific angle. The gaze guidance device 201, for example, fixes the subject's head in a first direction while directing the gaze directions of the subject's left and right eyes to a second direction different from the first direction. The first direction is, for example, a reference direction for directing the gaze to the second direction, and is the direction in front of the subject. The first direction is, for example, the direction in which the gaze direction is horizontal and indicates 0 degrees. The second direction is, for example, a direction other than in front (to the side). The second direction is, for example, a direction at an angle obtained by rotating the first direction counterclockwise or clockwise by an angle θ°.

[0019] The eye-tracking device 201 guides the subject's gaze to any direction to the side (direction of angle θ°) and causes the subject to fixate on an image displayed on a display device 300 installed in that direction. The eye-tracking device 201 is equipped with an optical mechanism for the subject to visually perceive different images with each of their left and right eyes.

[0020] Figure 2 is a schematic diagram of the gaze guidance device 201 according to this embodiment. Figure 3 is a schematic configuration diagram of a first example of the gaze guidance device 201 according to this embodiment. The second direction guided by the gaze guidance device 201 illustrated in Figure 3 is the direction at an angle obtained by rotating the first direction counterclockwise by an angle of 30°. As shown in Figures 2 and 3, the gaze guidance device 201 comprises a binocular unit 210 and two guidance body units 220.

[0021] The binocular unit 210 comprises a right-eye monocular 211R and a left-eye monocular 211L. The right-eye monocular 211R and the left-eye monocular 211L each have the same or similar configurations. In cases where the same or similar configurations are assigned the same reference number across multiple drawings, the explanation may be omitted. Also, when the right-eye monocular 211R and the left-eye monocular 211L are not distinguished, they may simply be referred to as "monocular 211".

[0022] The monocular 211 has a shape into which the subject looks. The monocular 211 is, for example, a cylindrical optical element and incorporates a vision correction lens 212. In this embodiment, the cylindrical monocular 211 that guides the direction of the left eye's line of sight and allows the subject to control their line of sight by looking into it is the left-eye monocular 211L, and the cylindrical monocular 211 that guides the direction of the right eye's line of sight and allows the subject to control their line of sight by looking into it is the right-eye monocular 211R.

[0023] The monocular 211 is provided with a fixing device (e.g., ring-shaped) for positioning a vision correction lens 212. The vision correction lens 212 is selected, for example, based on the subject's visual acuity measurement results and plays a role in providing a corrective effect. The mounting of the vision correction lens 212 employs a structure that allows it to be fixed at a predetermined angle, and the vision correction lens 212 can be replaced as needed. For example, the cylindrical side of the monocular 211 has an opening into which multiple vision correction lenses 212 can be inserted. The operator can insert and set a vision correction lens 212 according to the subject's visual acuity through this opening. The optical center of the vision correction lens 212 is set, for example, to coincide with the line of sight direction guided in a second direction.

[0024] The guidance unit 220 is a key component for guiding the subject TS's gaze in a second direction, and is provided independently on the left and right sides. The guidance unit 220 for the left eye is sometimes referred to as the "left eye guidance unit 220L," and the guidance unit 220 for the right eye is sometimes referred to as the "right eye guidance unit 220R." The left eye monocular 211L and the left eye guidance unit 220L are examples of left-eye gaze guidance units. The right eye monocular 211R and the right eye guidance unit 220R are examples of right-eye gaze guidance.

[0025] The guidance unit 220 includes a plurality of optical elements that reflect light at a certain angle. The guidance unit 220 includes a plurality of total reflection mirrors 230, 231, a half mirror 240, and an imaging unit 250. The half mirror 240 and the plurality of total reflection mirrors 230, 231 are examples of a plurality of optical elements that reflect light at a certain angle. The half mirror 240 and the plurality of total reflection mirrors 230, 231 work together to control the line of sight direction of the subject TS.

[0026] A monocular 211 is provided in the guidance body 220. For example, the guidance body 220 supports the monocular 211 so that it can move in a second direction. Specifically, the left eye guidance body 220L supports the left eye monocular 211L so that it can move in a second direction, and the right eye guidance body 220R supports the right eye monocular 211R so that it can move in a second direction.

[0027] The guidance unit 220 is movable. The position of each guidance unit 220 can be adjusted according to the interpupillary distance of the subject TS. That is, the distance between the left eye guidance unit 220L and the right eye guidance unit 220R can be adjusted according to the interpupillary distance of the subject TS. Therefore, by adjusting the distance between the left eye guidance unit 220L and the right eye guidance unit 220R, the distance between the left eye monocular 211L and the right eye monocular 211R can be adjusted. The adjustment of the distance between the left eye guidance unit 220L and the right eye guidance unit 220R may be automatic or manual.

[0028] The total reflection mirrors 230 and 231 are optical elements that guide the image light emitted from the display device 300 to the half mirror 240. The image light is the light that constitutes the image displayed on the display device 300 and is emitted along the second direction. The image light emitted from the display device 300 along the second direction first reaches the total reflection mirror 230, where the optical path is bent by 90 degrees by the reflective surface of the total reflection mirror 230 and reflected towards the total reflection mirror 231. The image light reflected by the reflective surface of the total reflection mirror 230 is then reflected in the second direction by the reflective surface of the total reflection mirror 231 and heads towards the half mirror 240.

[0029] The half-mirror 240 transmits a portion of the image light from the total reflection mirror 231 and reflects the remaining image light toward the imaging unit 250. The image light that has passed through the half-mirror 240 passes through the vision correction lens 212 (for example, the optical center of the vision correction lens 212) and is presented as an image to the subject TS's eye. That is, the image light from the second direction that has passed through the half-mirror 240 reaches the retina of the eye, and the subject TS perceives the image in the second direction while facing the first direction. The dashed line illustrated in Figure 3 shows the subject TS's line of sight.

[0030] The direction of the image light (second direction) that passes through the vision correction lens 212 coincides with the direction of the subject TS's line of sight. The subject TS's eyes are naturally and physically directed towards the second direction as they attempt to capture this image light. The subject TS's line of sight moves along the optical axis of the image light that passes through the vision correction lens 212 and is guided towards the second direction.

[0031] The imaging unit 250 captures the pupil of the subject TS based on the image light reflected by the half mirror 240. The image captured by the imaging unit 250 is transmitted to the information processing device 400. This allows the information processing device 400 to measure the pupil position and gaze direction of the subject TS in real time using the image transmitted from the imaging unit 250. The imaging unit 250 is, for example, an infrared camera. The image information captured by the imaging unit 250 may be used to improve the accuracy of gaze guidance, or to adjust the distance between the left eye guidance unit 220L and the right eye guidance unit 220R.

[0032] The guidance body 220 is not limited to having a configuration with a plurality of total reflection mirrors 230, 231 and a half mirror 240. The guidance body 220 only needs to have one or more optical elements that reflect light at a certain angle and guide the subject TS's line of sight in a second direction. Figure 4 is a schematic configuration diagram of a second example of the line of sight guidance device 201 according to this embodiment. The line of sight guidance device 201 may have a plurality of reflective prisms 260, 261 and a half mirror 240, as illustrated in Figure 4. In other words, the total reflection mirrors 230, 231 illustrated in Figure 3 may be replaced with reflective prisms 260, 261.

[0033] The principle by which the gaze guidance device 201 illustrated in Figure 4 guides the gaze direction of the subject TS to the second direction is the same as that of the gaze guidance device 201 illustrated in Figure 3. Furthermore, the gaze guidance devices 201 illustrated in Figures 3 and 4 are devices that enable the gaze direction to be directed for a certain period of time in a direction rotated counterclockwise by a specific angle. However, the device is not limited to this. For example, as illustrated in Figure 5, the gaze guidance device 201 may be a device that enables the gaze direction to be directed for a certain period of time in a direction rotated clockwise by a specific angle. In the gaze guidance device 201 illustrated in Figure 5, the total reflection mirrors 230 and 231 can of course be changed to reflection prisms.

[0034] The head fixing unit 202 is a device or instrument for improving the measurement accuracy of the subject's perceptual characteristics by stably holding the subject's head in a predetermined position and minimizing head movement. Figure 6 shows an example of the head fixing unit 202 according to this embodiment. Note that the head fixing unit 202 is omitted in Figure 2 for the sake of explanation.

[0035] The head fixing unit 202 has the function of fixing the head of the subject TS in a first direction. The head fixing unit 202 comprises, for example, two side supports 500, a forehead rest 510, and a chin rest 520. The side supports 500, the forehead rest 510, and the chin rest 520 make contact with various parts of the subject TS's head, thereby stabilizing the subject TS's head in three dimensions.

[0036] The two side supports 500 are structured to clamp the head from both sides in the left-right direction. The forehead rest 510 contacts the subject TS's forehead to prevent movement in the front-back direction. The chin rest 520 supports the subject TS's chin from below, and the height adjustment function of the chin rest 520 ensures that the subject TS's head is held in the optimal position.

[0037] The subject's head (TS) is securely fixed in a first direction (forward direction) by the head fixing part 202. The head fixing part 202 may also be equipped with cushioning material for added comfort. This cushioning material is designed to prevent excessive strain on the subject (TS) even during prolonged measurements. This configuration improves the accuracy of the line-of-sight measurement and ensures the reliability of the measurement results.

[0038] When subject TS looks through the front end of the left eye monocular 211L and the right eye monocular 211R, he adjusts the height of the chin rest 520 and rests his chin on the adjusted chin rest 520. Furthermore, by moving his head forward while his chin is resting on the chin rest 520, subject TS's head comes into contact with the front end of the forehead rest 510, fixing it in the front-to-back direction, and the two side supports 500 fix his head from both the left and right sides.

[0039] The display device 300 is connected to the information processing device 400. The display device 300 displays data transmitted from the information processing device 400 on its display screen. The display device 300 is a monitor for displaying data.

[0040] The display device 300 displays an image based on the image data transmitted from the information processing device 400 on the display screen. The display screen of the display device 300 is arranged at an appropriate position in the line-of-sight direction of the subject TS and generates image light.

[0041] The display device 300 can present different images to each of the left and right eyes. The image presented to the right eye may be referred to as the "right-eye image GR", and the image presented to the left eye may be referred to as the "left-eye image GL". Display areas corresponding to each of the left and right eyes can be set on the display screen of the display device 300. The geometric center of an image (each of the right-eye image GR and the left-eye image GL) is defined in each display area, and the optical system is adjusted so that the line of sight of the subject TS passes through this center. With this design, the characteristics of the presented image are visually optimized.

[0042] FIG. 7 is an example of a screen for displaying the fixation points T of each of the left and right eyes using one monitor. The display device 300 shown in the figure is configured using one monitor, and this monitor can display the fixation points T corresponding to each of the left and right eyes. The display device 300 is designed so that the left and right eyes of the subject TS can perceive different images and fixation points T, enabling the subject TS to observe using one screen with both eyes simultaneously. Different fixation points T and images (right-eye image GR and left-eye image GL) are presented on the monitor of the display device 300, thereby making it possible to guide the left and right lines of sight to specific positions and directions. To achieve this, the optical system of the line-of-sight guidance device 201 is arranged between the monitor and the subject TS and is adjusted so that the image light is accurately transmitted. Specifically, a plurality of optical elements built into the line-of-sight guidance device 201 are combined and set so that the left and right eyes look at different areas on the monitor.

[0043] For each of the right-eye image GR and the left-eye image GL on the monitor, a geometric center is defined, and the optical system and the head fixing unit 202 of the gaze induction device 201 are adjusted so that the gaze of the subject TS passes through this center. As a result, the positions of the fixation points presented to the left and right eyes coincide with the gaze of the subject TS, and the subject TS can experience single vision with both eyes. Thus, the display device 300 can provide individual visual experiences for the left and right eyes while using a single monitor. Further, the display device 300 is configured using, for example, a high-resolution display, and can precisely control the respective feature amounts (such as contrast, blur degree, hue, etc.) of the presented right-eye image GR and left-eye image GL. Thereby, different image characteristics can be presented to the left and right eyes, and it becomes possible to evaluate the visual differences perceived by each eye.

[0044] The information processing device 400 creates the image data of the right-eye image GR and the left-eye image GL to be presented to each of the left and right eyes, and causes the display device 300 to display them. Further, the information processing device 400 can measure the pupil position and the gaze direction of the subject TS using the image from the imaging unit 250. FIG. 8 is an example of a schematic block diagram of the information processing device 400 according to the present embodiment.

[0045] As shown in FIG. 8, the information processing device 400 includes a communication I / F 410, a storage device 420, and a processor 430. The processor 430 is an example of a processing unit. Note that the storage device 420 may be an external storage device instead of being a component of the information processing device 400. When the storage device 420 is an external storage device, the information processing device 400 is connected to the storage device 420 by wire or wirelessly, and transmits and receives information to and from the storage device 420. The storage device 420 is an example of a storage unit.

[0046] The communication I / F 410 is a communication interface for communicating with an external device (for example, the display device 300, the gaze induction device 201, or an external communication terminal). The communication network through which the communication I / F 410 communicates may be wired, wireless, or both.

[0047] The storage device 420 includes, for example, non-volatile memory such as ROM (Read Only Memory), an HDD (Hard Disk Drive), and an SSD (Solid State Drive). The storage device 420 stores a processing program. The processing program may be provided by a computer-readable storage medium or by an external device via a wired or wireless communication network. The provided processing program is stored in the storage device 420 and executed by the processor 430.

[0048] Examples of computer-readable storage media mentioned above may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media may include diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (ERPOM or flash memory), electrically erasable programmable read-only memory (EERPOM), static random access memory (SRAM), compact disk read-only memory (CD-ROM), digital multipurpose disc (DVD), Blu-ray (RTM) disc, memory stick, integrated circuit card, etc.

[0049] The processor 430 includes, for example, at least one of a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). Alternatively, the processor 430 may be a microcontroller unit. The processor 430 is an example of a processing unit.

[0050] Figure 9 shows the functional blocks of the processor 430 according to this embodiment. The processor 430 comprises an image control unit 600 and an information processing unit 650. The image control unit 600 and the information processing unit 650 are software modules realized by the processor 430 executing a processing program stored in the storage device 420.

[0051] When the image control unit 600 performs perceptual characteristic measurement (hereinafter simply referred to as "perceptual characteristic measurement") necessary to evaluate the difference in sensitivity and perception between the left and right eyes of the subject TS in their vision, it spatially presents images separately to the left and right eyes of the subject TS, whose gaze direction has been guided to a second direction by the gaze guidance unit 200. In this embodiment, the image control unit 600 spatially presents images separately by displaying the left eye image GL in a first display area and the right eye image GR in a second display area different from the first display area on the display screen of the display device 300. Each of the left eye image GL and the right eye image GR has feature quantities as indicators of visual characteristics. The feature quantities of the left eye image GL displayed on the display screen of the display device 300 are different from the feature quantities of the right eye image GR displayed on the display screen. Feature quantities include, for example, the degree of blur, brightness, and shape.

[0052] The image control unit 600 is capable of performing a left-eye change process, which is a process of gradually or continuously changing the feature quantities of the left-eye image GL on the display screen of the display device 300 during perceptual characteristic measurement, and a right-eye change process, which is a process of gradually or continuously changing the feature quantities of the right-eye image.

[0053] In left-eye change processing, the feature quantities of the left-eye image GL are changed stepwise or continuously while the feature quantities of the right-eye image GR are fixed at a constant value. This processing makes it possible to evaluate how the left eye responds to changing feature quantities while the right eye is perceiving a fixed reference image. For example, by changing feature quantities such as "blur degree" or "contrast value" of the left-eye image GL, the change in the subject's visual perception TS can be measured.

[0054] For example, the image control unit 600 performs a process to change the left eye image GR to a "constant image having a feature quantity of 5.0" and to "gradually change the feature quantity of the left eye image GL from 3.0 to 6.0" as left eye change processing.

[0055] In the right-eye change processing, the feature quantities of the left-eye image GL are fixed at a constant value, while the feature quantities of the right-eye image GR are changed in a stepwise or continuous manner. This processing makes it possible to evaluate how the right eye responds to changing feature quantities while the left eye is perceiving a fixed reference image.

[0056] For example, the image control unit 600 performs a process to change the right eye image GL, fixing the left eye image GL to a "constant image having a feature quantity of 5.0", and "gradually changing the feature quantity of the right eye image GR from 3.0 to 6.0".

[0057] The image control unit 600 may independently perform left-eye change processing and right-eye change processing when evaluating the difference in sensitivity between the left and right eyes. In human vision, there may be differences in the sensitivity and perceptual characteristics of the left and right eyes. To accurately understand this, it is sometimes desirable to gradually or continuously change the feature quantities of the other eye using one eye as a reference. For example, in left-eye change processing, by changing the feature quantities of the left eye while the right eye is perceiving a certain image, it is possible to evaluate how the left eye responds to the change in feature quantities. On the other hand, in right-eye change processing, by changing the feature quantities of the right eye while the left eye is perceiving a certain image, it is possible to evaluate the response of the right eye. This process makes it possible to independently clarify the difference in sensitivity between the left and right eyes.

[0058] Furthermore, to analyze the integration process in binocular monopsia, it is important to perform processing that alters the left and right eye feature quantities. When images with different feature quantities are presented to the left and right eyes, the human brain integrates these images to form a single visual image. In this integration process, it is important to understand how the influence of the feature quantities of each eye manifests in visual perception. By using left-eye alteration processing and right-eye alteration processing, it is possible to evaluate in detail the roles that the left and right eyes play in monopsia perception.

[0059] Furthermore, separate feature vector processing for the left and right eyes is required from the perspective of evaluating asymmetry in visual perception. In visual perception, the left and right eyes often have different perceptual characteristics. Evaluating this asymmetry is extremely important, especially in the design of corrective lenses and progressive lenses. By applying individual feature vector changes to the left and right eyes, it becomes possible to accurately measure the asymmetrical characteristics of each eye, and this can be reflected in the design of visual correction devices and lenses.

[0060] The image control unit 600 creates image data for the left eye (GL) and the right eye (GR) when performing right eye change processing and left eye change processing. The method for creating the left eye (GL) and the right eye (GR) (hereinafter referred to as the "image creation method") will be described below. First, as the first example of the image creation method, the method for creating the left eye (GL) and the right eye (GR) images in which the feature quantity is the degree of blur will be described.

[0061] When creating a display image, the image control unit 600 first creates a square image with dimensions of 600 pixels vertically and 600 pixels horizontally, and uniformly sets the pixel values ​​of this square image to 255 (a value close to white). In other words, the image control unit 600 uniformly sets the pixel values ​​of the background portion of the image to 255. Next, the image control unit 600 places characters on this image, each with a side length of 104 pixels and a pixel value of 100 (dark gray). These characters are arranged using the alphabet from "A" to "Z", with five characters in each of the five columns from left to right. In this way, the image control unit 600 creates an original image with a pixel value of 255, and 25 characters in total, arranged in five columns of five characters each from the alphabet from "A" to "Z" with a pixel value of 100, from left to right. The original image is an image in which a total of 25 characters are arranged in a regular pattern within a single image.

[0062] The image control unit 600 performs image processing on the original image. For example, the image control unit 600 creates an array with dimensions of 29 x 29. The image control unit 600 also generates a matrix that follows a Gaussian distribution with mean 1 / 29 and variance [1.4, 2.4, 3.0, 3.8, 4.7, 5.8, 6.9, 8.6, 11.0, 15.0]. The image control unit 600 inserts the generated matrix into the 15th row of the 29 x 29 array to create a filter. The image control unit 600 applies this filter to the original image to perform filtering and create 10 images (hereinafter referred to as "stimulus images"), each with different feature quantities.

[0063] Figure 10 shows the original image and 10 stimulus images according to this embodiment. Each of the 10 stimulus images shown in Figure 10 is labeled with an identification number from "01" to "10". This label indicates the order in which the feature quantities of the images change, with the image having the fewest feature quantities being labeled "01", and the numbers increasing stepwise from "02", "03", and so on. This makes it easy to identify which stage of feature quantities each presented image has, and allows for correspondence with the order of the images in experimental procedures and result analysis.

[0064] The feature quantity of the stimulus image shown in Figure 10 is the degree of blur. The degree of blur is a numerical measure that indicates the extent to which the clarity of edges (contours and boundaries) in the image is reduced. The image control unit 600 can use this stimulus image as the left eye image GL and the right eye image GR. Stimulus images in which the feature quantity is the degree of blur are sometimes referred to as blurred images.

[0065] The image control unit 600 can quantify the feature quantities (e.g., degree of blur) of each stimulus image. For example, the image control unit 600 processes the stimulus image using feature quantity detection filters (Equations 1 and 2). The feature quantity detection filter of Equation (1) detects changes in brightness in the horizontal direction (X-axis) of the stimulus image. The feature quantity detection filter of Equation (2) detects changes in brightness in the vertical direction (Y-axis).

[0066] The image control unit 600 applies a feature detection filter to the stimulus image and calculates the change in brightness in the horizontal and vertical directions for each pixel. The image control unit 600 expresses the calculation results as vectors for the horizontal and vertical components, respectively, and calculates the magnitude of the vectors for each pixel. Next, the image control unit 600 statistically processes the magnitudes of the vectors obtained for all pixels in the image and calculates them as features for the entire image. These features are a numerical representation of the degree of blur in the image. The processing by the feature detection filter models the function of neurons that detect changes in brightness in the human visual system. Therefore, it accurately quantifies the visual features of the image and enables an objective evaluation of the degree of blur.

[0067]

[0068]

[0069] Next, as a second example of the image creation method by the image control unit 600, we will describe how to create the left-eye image GL and the right-eye image GR, where the feature quantity is the degree of deformation due to skew transformation. The image control unit 600 first creates an original image. The method for creating the original image is the same as in the first example of the image creation method, so we will omit the explanation.

[0070] The image control unit 600 generated a new image by applying an affine transformation to the coordinates (x and y coordinates) of the original image, thereby changing the arrangement and shape of the entire string. An affine transformation is a method that uses a 3x3 matrix to achieve geometric deformations such as scaling, translation, rotation, and skew (shearing) of an image. Furthermore, it is possible to perform more complex transformations by combining multiple matrices. As an example of this embodiment, the image control unit 600 created a new image by applying an x-axis skew transformation and a y-axis skew transformation to the original image as affine transformations. Equation (3) below is a 3x3 matrix used for the x-axis skew transformation. Equation (4) below is a 3x3 matrix used for the y-axis skew transformation. By changing the value of tanθ in these matrices, the degree of deformation due to the skew transformation can be controlled as a feature, representing the degree of tilt and distortion of the arrangement and shape of the entire string.

[0071]

[0072]

[0073] The image control unit 600 creates 10 stimulus images, each with different feature quantities, by setting θ in the matrices of equations (3) and (4) to θ1, θ2, θ3, θ4, θ5, θ6, θ7, θ8, θ9, and θ10. Figure 11 shows an example of 10 stimulus images created using the second example of the image creation method. Each of the 10 stimulus images shown in Figure 11 is labeled with an identification number from "01" to "10," similar to Figure 10. This label indicates the order in which the image feature quantities (degree of deformation due to skew transformation) change, with the image having the smallest feature quantity being labeled "01," and the numbers increasing step by step being "02," "03," and so on. This makes it easy to identify which stage of feature quantities a presented image has, and allows for correspondence with the order of the images in experimental procedures and result analysis. The image control unit 600 can use one or more of these stimulus images shown in Figure 11 as the left eye image GL and the right eye image GR.

[0074] Next, the calibration according to this embodiment will be described. Calibration is performed before the measurement of perceptual characteristics. Calibration is a process that adjusts the position of the left and right boxes of the measuring instrument based on the interpupillary distance of the subject TS and aligns the geometric center of the image on the monitor with the line of sight of the subject TS. Figure 12 shows the processing flow of calibration according to this embodiment.

[0075] First, before starting the perceptual characteristics measurement, the purpose and procedure of the measurement are explained to the subject TS (Step S101). For example, the operator gives the subject TS specific instructions such as, "We will now present different images to your left and right eyes. Please tell us whether you can perceive the difference or not."

[0076] The operator measures the subject TS's visual acuity and, if necessary, selects an appropriate corrective lens 212 based on the results. The operator then sets the selected corrective lens 212 into the eye-tracking device 201 (step S102). This ensures that the subject TS can accurately view the image.

[0077] The information processing device 400 adjusts the positions of the left eye guide unit 220L and the right eye guide unit 220R based on the interpupillary distance of the subject TS measured by the interpupillary distance meter, and fixes the positions precisely to match the position of the subject TS's eyes (step 103). This adjustment may be performed by an operator.

[0078] Subject TS sits in the chair, adjusts the height of the chin rest 520, and places his chin on the chin rest 520 while simultaneously moving his head forward so that his head comes into contact with the tip of the forehead rest 510. As a result, Subject TS's head is fixed in the first direction by the head fixing part 202 (step S104).

[0079] The information processing device 400 presents a calibration image (for example, an image including a fixation point "+") to each of the subject TS's left and right eyes by displaying the calibration image on the display screen of the display device 300. The subject TS is asked to gaze at the fixation point of the calibration image for several seconds (for example, 5 seconds). The imaging unit 250 captures the pupils of the subject TS while the calibration image is being presented or while the subject TS is gazing at the fixation point (step S105). For example, the imaging unit 250 captures approximately 570 images at 115 frames per second. The images captured by the imaging unit 250 are transmitted to the information processing device 400.

[0080] The information processing device 400 calculates the center position of the left and right pupils based on the image of the subject TS captured by the imaging unit 250. The information processing device 400 then calculates the interpupillary distance from the center positions of the pupils (step S106) and determines whether the error between the calculated interpupillary distance and the interpupillary distance measured by the interpupillary distance meter is below the error threshold (step S107). If the error exceeds the error threshold, the information processing device 400 proceeds to step S103 to readjust the positions of the left eye guide unit 220L and the right eye guide unit 220R. If the error is below the error threshold in step S107, the calibration is completed.

[0081] Next, the measurement method for measuring perceptual characteristics according to this embodiment will be described. In this embodiment, sensitivity measurement using a psychophysical measurement method is performed as the measurement of perceptual characteristics. In this measurement, the subjective equivalence point (PSE) and the judgment threshold (JND) are determined.

[0082] Psychophysical measurement methods include the method of limits, the method of constants, and the adjustment method. In this embodiment, the method of limits is used to explain the case where the method of limits is employed. However, the method of constants or the adjustment method can also be used. In the method of limits, a constant standard stimulus and a variable stimulus whose magnitude and intensity change in stages are presented to the subject TS. The subject TS looks at these images, compares them, and answers the result. In this case, the variable stimulus has features (e.g., degree of blurring) that change in stages. The subject TS judges using either a dichotomy ("harder to see (+)", "clearer (-)") or a ternary ("harder to see (+)", "clearer (-)", "cannot tell or same (?)"). In this embodiment, the case in which the subject TS judges the difference between images using a ternary method is explained as an example using the method of limits. The information processing device 400 then calculates two parameters in the psychometric function, the subjective equivalence point and the discrimination threshold, based on the subject TS's response data.

[0083] Here, perceptual characteristic measurements performed on the subject TS reveal that comparative judgments regarding stimuli near the subjective equivalence point or discrimination threshold do not always result in the same judgment for the same physical stimulus, but may fluctuate probabilistically. This probabilistic fluctuation, expressed as a physical stimulus value function, is called a psychometric function. Typical psychometric functions include the cumulative normal distribution, logistic distribution, Weibull distribution, and Campbell distribution. However, since the cumulative normal distribution function is the most theoretically justified function, this embodiment will explain the case where the subjective equivalence point and discrimination threshold are calculated using the cumulative normal distribution function as an example.

[0084] The image control unit 600, when presenting the left eye image GL and the right eye image GR, displays two images for each eye on the same coordinate axis of the display screen, superimposed like front and back. Specifically, for the presentation of the left eye image GL, the image control unit 600 displays two images superimposed like front and back in the first display area. Similarly, for the presentation of the right eye image GR, the image control unit 600 displays two images superimposed like front and back in the second display area.

[0085] In this way, the image control unit 600 uses two images to superimpose and display them as the left eye image GL, and uses two images to superimpose and display them as the right eye image GR. The image control unit 600 can gradually change the alpha value (transparency) of each pixel in the back image (hereinafter referred to as the "back image") and the front image (hereinafter referred to as the "front image") from 0 to 255, or from 255 (completely opaque) to 0 (completely transparent). This allows the image control unit 600 to switch from the back image to the front image and from the front image to the back image. As a result, the subject TS can fixate on the back image or the front image without moving their gaze. Furthermore, by controlling the rate of change of the alpha value, the image control unit 600 can apply multiple changes within the presentation time. In addition, by changing the alpha value on a pixel-by-pixel basis, the image control unit 600 can also apply partial changes to the image rather than the entire image.

[0086] When performing right-eye change processing in perceptual characteristic measurement, the image control unit 600 fixes the feature quantities of the left-eye image GL to a constant value and changes the feature quantities of the right-eye image GR in steps or continuously. For example, as the right-eye image GR in right-eye change processing, the image control unit 600 uses a standard image with constant feature quantities as the back image and displays stimulus images as the front image, changing them in steps from label "01" to "10". The image control unit 600 also sets and displays the standard image as both the back and front image for the left eye image in right-eye change processing. In other words, in right-eye change processing, only the front image of the right-eye image is a changing image in which the feature quantities change. The standard image is an image with constant feature quantities used as a reference value, and for example, a stimulus image with label "05" can be used.

[0087] In the right-eye transition processing, the image control unit 600 may change the alpha value of each pixel of the front image of the right-eye image GR from 0 to 255 so that the back image gradually becomes visible over a period of 5 seconds. Figure 13 is a diagram illustrating the method of presenting the stimulus image in the right-eye transition processing according to this embodiment. At the start of presentation, the alpha value (transparency) of each pixel of the front image for the right eye is set to 0, so the front image is completely transparent. In this state, the standard image set on the back side of the right-eye image is presented to the subject TS's vision. After the start of presentation, the image control unit 600 changes the alpha value of each pixel of the front image for the right eye from 0 to 255 in steps or continuously. As a result, the standard image on the back side gradually becomes invisible, and the front image is gradually presented to the subject TS.

[0088] This switching process takes 5 seconds, eventually reaching a state where the front image is fully displayed (alpha value 255). Meanwhile, the left eye has a standard image set for both the front and back images from the start of presentation, and no change occurs in the left eye's image. This configuration makes it possible to accurately evaluate the subject's (TS's) visual response to changes in feature quantities (e.g., degree of blur) in the front image for the right eye.

[0089] When performing left-eye change processing in perceptual characteristic measurement, the image control unit 600 fixes the feature quantities of the right-eye image GR to a constant value and changes the feature quantities of the left-eye image GL in steps or continuously. For example, as the left-eye image GL in left-eye change processing, the image control unit 600 uses a standard image with constant feature quantities as the back image and displays the stimulus image as the front image, changing it stepwise from label "01" to "10". Also, as the right-eye image in left-eye change processing, the image control unit 600 sets the standard image as both the back image and the front image and displays it. In other words, in left-eye change processing, only the front image of the left-eye image becomes the change image.

[0090] In the left-eye transition processing, the image control unit 600 may change the alpha value of each pixel of the front image of the left-eye image GL from 0 to 255, so that the front image gradually becomes visible from the back image over a period of 5 seconds. Figure 14 is a diagram illustrating the method of presenting the stimulus image in the left-eye transition processing according to this embodiment. At the start of presentation, the alpha value (transparency) of each pixel of the front image for the left eye is set to 0, so the front image is completely transparent. In this state, the standard image set on the back side is presented to the subject TS's vision. After the start of presentation, the image control unit 600 changes the alpha value of each pixel of the front image for the left eye from 0 to 255 in steps or continuously. As a result, the standard image on the back side gradually becomes invisible, and the front image is gradually presented to the subject TS.

[0091] This switching process takes 5 seconds, eventually reaching a state where the front image is fully displayed (alpha value 255). Meanwhile, the right eye has a standard image set for both the front and back images from the start of presentation, and no change occurs in the image for the right eye. This configuration makes it possible to accurately evaluate the subject's (TS's) visual response to changes in feature quantities (e.g., degree of blur) in the front image for the left eye. The switching process achieves a smooth image transition by controlling the transparency (alpha value) of the front and back images, allowing the subject (TS) to perceive the visual change while keeping their gaze fixed.

[0092] In both the left-eye and right-eye change processing, the subject TS is instructed to gaze at the presented image for 5 seconds. Depending on the differences in the features of the presented images, the following responses are expected: When the features of the back image are greater than those of the front image (back > front), the subject TS perceives the image they are gazing at as gradually becoming clearer. Conversely, when the features of the back image are smaller than those of the front image (back < front), the subject TS perceives the image as gradually becoming harder to see. Furthermore, when the difference in features between the back and front images is small, the subject TS is expected to be unable to perceive any difference between the two images and judge that there is "no change." These responses (response data) from the subject TS will be used to evaluate the subject TS's visual sensitivity to the features of the presented images.

[0093] In both the left-eye change processing and the right-eye change processing, the image control unit 600 displays a random dot image before presenting the test images (left-eye image GL and right-eye image GR), allowing the subject TS to perform visual adaptation. Subsequently, the image control unit 600 presents the left-eye image GL and right-eye image GR, as illustrated in Figures 13 and 14, to the subject TS for a certain period of time (for example, 5 seconds). The subject TS fixates their gaze on the left-eye image GL and right-eye image GR with their line of sight fixed in the second direction and responds using one of the three-point scale described above: "difficult to see (+)", "clearer (-)", or "cannot tell or same (?)".

[0094] The image control unit 600 or the operator may determine the next change image to present based on the response result. The determined change image is then presented to the subject TS by the display device 300 of the measuring device, and responses from the subject TS are repeatedly obtained. Through this series of processes, the visual responses of the subject TS are sequentially acquired, making it possible to measure parameters such as the subjective equivalence point (PSE) and the discrimination threshold (JND) with high accuracy. The subject TS may also directly input the content of their response into the information processing device 400 by using an input device connected to the information processing device 400. Alternatively, the subject TS may answer verbally to the operator, and the operator may input the content of that response into the information processing device 400.

[0095] A triplicate experiment was conducted using the cumulative normal distribution function as the psychometric function. In this case, the probability model that models the subject TS response has three parameters: μ (subjective equivalence point), σ (square root of variance), and C (judgment boundary). The subject TS response to the feature quantity Xt of the stimulus image falls into three patterns: when the image becomes difficult to see (+), when it is indistinguishable or feels the same (-), and when it becomes easier to see or clearer (-). The probability models for each response pattern are shown below. Note that μ is the subjective equivalence point (PSE), C is the distance between the subjective equivalence point and the judgment criterion (judgment boundary), and σ is the square root of variance.

[0096] Equation (5) shows a probabilistic model of the response when the image being watched becomes difficult to see, Equation (6) shows a probabilistic model of the response when the image being watched is indistinguishable or appears the same, and Equation (7) shows a probabilistic model of the response when the image being watched becomes easier to see or appears clearer.

[0097] F(Xt;σ,C1,C2)=Φ((Xt-C2) / σ)…(5)

[0098] F(Xt;σ,C1,C2)=Φ((Xt-C1 / σ)-Φ((Xt-C2) / σ)...(6)

[0099] F(Xt;σ,C1,C2)=1-φ((Xt-C1) / σ)...(7)

[0100] C1=μ−C…(8)

[0101] C2=μ+C...(9)

[0102] PSE=(C1+C2) / 2...(10)

[0103] The subjective equivalence point (PSE) is considered to be a value indistinguishable from the standard stimulus (hereinafter also referred to as the "standard image"), and this value is represented by μ.

[0104] PSE=μ...(11)

[0105] The discrimination threshold (JND) is expressed by the following formula, based on the square root of the variance σ, as the value corresponding to a discrimination probability of 75%.

[0106] JND=0.6745×σ…(12)

[0107] The boundary parameters C1 and C2 for category judgments depend on the judgment criteria and are distinct from the discriminative power of the senses. The judgment process and the sensory process are modeled by signal detection theory. Note that when analyzing 3-point scale data using the whole-series method, it can be difficult to separately evaluate the judgment criteria and the discriminative power of the senses. On the other hand, by performing analysis based on a probabilistic model, it is possible to appropriately distinguish between the judgment process and the sensory process. Furthermore, if the total number of trials is N, and N data points are grouped together, the response data D = {R １ , R ２ , … R ＮThis is expressed as}. Here, the N data points represent the response (a triplicate response using "+", "-", and "?") to the image features presented to the subject TS in each trial. In this case, the likelihood L(θ|D) is given by the following equation. The likelihood L(θ|D) is the probability of obtaining the observed response data D given the parameter θ. It is expressed as θ = {μ, σ, C}.

[0108]

[0109] In a trial, let Xi be the feature of the i-th image presented, and Ri be the response of subject TS to it. The probability that subject TS's response Ri occurs under the conditions of the presented feature Xi is given by P(Ri|Xi). The likelihood L(θ|D) based on subject TS's response data D across all trials is the sum of the conditional probabilities P(Ri|Xi) for all trials.

[0110] For example, suppose that right-eye change processing is performed and the response data shown in the table in Figure 15 is obtained. Note that the response data shown in Figure 15 is used in the calculation of the likelihood, P(θ|D), described later, and the order of the data does not affect the analysis results. The response R of the subject TS is obtained from the response of the subject TS, with "1" indicating clearer vision, "2" indicating not knowing or the same vision, and "3" indicating worsening vision.

[0111] In the field of modern social science, Bayesian statistical analysis is widely recognized as a standard analytical method. The Psychological Society, a scientific psychology society in North America, also recommends Bayesian statistical analysis as an alternative to conventional null hypothesis testing. Furthermore, many software programs have been developed that allow for easy implementation of Bayesian statistical analysis, such as PyMC3, NumPyro, and PyStan. In this embodiment, Bayesian statistical analysis is employed for parameter estimation of the psychometric function. Compared to the conventional maximum likelihood method, Bayesian statistical analysis enables highly accurate parameter estimation even with a small number of trials, and furthermore, by using the obtained posterior distribution (equation (14)), flexible and diverse analyses become possible. Specifically, the Markov-Chain-Monte-Carlo method (MCMC method) is used to calculate point estimates from the posterior distribution of the parameters. These point estimates are obtained as representative values ​​of the posterior distribution, such as the mean, median, or mode.

[0112]

[0113] P ０ (θ) is the prior distribution, P(D|θ) is the likelihood, and P(θ|D) is the posterior distribution. The prior distribution needs to be appropriately selected so as not to impose a substantial bias on the posterior distribution. For example, when setting the prior distribution, the information processing unit 650 considers the reliability of the prior information and selects an appropriate distribution that does not impose a substantial bias on the posterior distribution. Specifically, if prior information is scarce, a non-informative prior distribution (e.g., a uniform distribution) is selected to prevent the analysis results from being overly dependent on the prior information. Furthermore, the information processing unit 650 also needs to set the parameters of the selected prior distribution (e.g., mean and variance) in the same way. In this embodiment, the information processing unit 650 used a known analysis library using the Python language to obtain the posterior distribution of the parameters (the posterior distribution of the three parameters (μ, σ, C) of the psychometric function) from the response data.

[0114] A uniform distribution, which is a non-informative prior distribution, was adopted as the prior distribution for these parameters. The information processing unit 650 uses the response data as a likelihood function and calculates the posterior distribution based on this. Subsequently, the information processing unit 650 performs sampling using the MCMC method from the posterior distribution to obtain representative values ​​of the parameters. These representative values ​​include the mean, median, and mode. In this embodiment, the information processing unit 650 uses the median or mode as representative values ​​for point estimation in order to mitigate the influence of outliers.

[0115] Figure 16 is an example of a graph of the posterior probability distribution of the subjective equivalence point (μ). Figure 16 shows the posterior probability distribution of the subjective equivalence point (μ) obtained from response data of right eye change processing when the second direction is 30 degrees. In Figure 16, the horizontal axis (μ) shows the value of the subjective equivalence point (PSE), and the vertical axis shows the posterior probability density for each value of μ.

[0116] In the distribution shown in Figure 16, the information processing unit 650 obtains the posterior mode (14.95) of the subjective equivalence point (μ) by determining the mode (maximum value) of the posterior probability density distribution shown in Figure 16. Figure 16 also shows a 95% confidence interval [11.65 to 16.67], indicating a high probability that the subjective equivalence point lies within this range. The information processing unit 650 sets the feature quantities of the standard stimulus image in this embodiment as reference values ​​and calculates the RMS error (Root Mean Square Error) based on these values. By changing the experimental conditions to reduce this RMS error, highly accurate experiments can be achieved. Note that the formula for calculating the RMS error is publicly known and therefore its explanation is omitted.

[0117] Furthermore, the subjective equivalence point (PSE) and the judgment threshold (JND) are estimated from the posterior distribution using the following formulas.

[0118] P(μ|D)=ΣP(θ|D)…(15)

[0119] PSE(Mode)=Arg max P(μ|D)...(16)

[0120] JND(Mode)=z ０．７５ × Arg max P(σ|D)…(17)

[0121] z０．７５ For a random variable X that follows a standard normal distribution, P(X < z ０．７５ This is the value where ) = 0.75. JND is calculated as the value corresponding to a discrimination probability of 0.75. The reason for adopting the mode of the posterior distribution as an estimate is that the mode minimizes the RMS error. Furthermore, 95% HPD (Highest Posterior Density) corresponds to the 95% confidence interval in the maximum likelihood method, but if the original data is a discrete distribution, it can be obtained by smoothing using kernel density estimation (KDE).

[0122] The graph in Figure 16 visualizes the posterior distribution of μ obtained by equation (15). The peak of the graph indicates the maximum point of the posterior distribution and corresponds to the subjective equivalence point (PSE) obtained by equation (16).

[0123] Figure 17 is a graph showing the posterior probability distribution of the square root of the variance σ. Figure 17 shows the posterior probability distribution of the square root of the variance σ obtained from the response data of the right eye change processing when the second direction is 30 degrees. In Figure 17, the horizontal axis (σ) shows the value of the standard deviation (square root of the variance), and the vertical axis shows the posterior probability density for each value of σ. The information processing unit 650 can obtain the posterior mode (σ = 2.43) by calculating the maximum value of the posterior probability density. Note that Figure 17 shows a range of [1.42 to 14.25] as the 95% confidence interval. The information processing unit 650 can calculate the discrimination threshold (JND) by substituting the posterior mode obtained from this graph into equation (17). ０．７５ ≈ 0.6745. Therefore, JND = 0.6745 × 0.39 ≈ 1.64.

[0124] Figure 18 is a graph showing the posterior probability distribution of the decision boundary C. Figure 18 shows the posterior probability distribution of the decision boundary C obtained from the response data of the right eye change processing when the second direction is 30 degrees. The horizontal axis of Figure 18 shows the value of the decision boundary C, and the vertical axis shows the posterior probability density for each value of C. The information processing unit 650 finds the posterior mode (C = 2.24) in the point estimation of C by finding the maximum value of the posterior probability distribution.

[0125] Figure 19 is a graph plotting the psychometric function (PF) based on the posterior distribution of the three parameters (μ, σ, C) described above. This psychometric function shows the relationship between the feature quantities and the response probability of the subject TS when the second direction is 30 degrees. This psychometric function continuously models how the response probability of the subject TS changes as the feature quantity increases. The response data of the subject TS obtained by presenting a stimulus image with a certain feature quantity to the subject TS is originally represented as a discrete distribution, but by applying kernel density estimation (KDE) to this, it is smoothed and the psychometric function is shown as a nonparametric regression curve.

[0126] Figure 20 shows the results of measuring the perceptual characteristics of the right eye with the second direction set to 30 degrees, and the results of measuring the perceptual characteristics of the right and left eyes with the second direction set to 0 degrees. Figure 20 shows the subjective equivalence point (PSE), RMS error, discrimination threshold (JND = 0.6745σ), and judgment boundary C as results of the perceptual characteristic measurement.

[0127] As shown in Figure 20, the error between the PSE and the standard image features was calculated using the following formula, resulting in a 1.6% error when the second direction was 0 degrees in the right eye measurement (hereinafter referred to as "right eye 0 degrees"), an 8.9% error when the second direction was 0 degrees in the left eye measurement (hereinafter referred to as "left eye 0 degrees"), and a 5.1% error when the second direction was 30 degrees in the right eye measurement (hereinafter referred to as "right eye 30 degrees").

[0128] Error (%) = (|PSE - standard image feature value| / standard image feature value) × 100 ... (18)

[0129] Furthermore, the discrimination threshold (JND) was defined as the minimum value at which the subject's TS perceived a difference in blurredness, and was measured as 0.26 for the right eye at 0 degrees, 0.66 for the left eye at 0 degrees, and 1.64 for the right eye at 30 degrees. In addition, the judgment boundary (C) represents the distance between the subjective equivalence point (PSE) and the judgment criterion value, and was 0.93 for the right eye at 0 degrees, 1.98 for the left eye at 0 degrees, and 2.26 for the right eye at 30 degrees. In this measurement, image features were calculated using a mathematical method that mimics the function of visual system neurons, and these values ​​could be quantified by performing Bayesian estimation based on the response data of the subject's TS in left and right eyes and in lateral gaze.

[0130] In this way, the information processing unit 650 obtains evaluation values ​​based on binocular single-view of the subject TS, obtained by presenting a left-eye image GL to the left eye and a right-eye image GR to the right eye. The evaluation values ​​may be one or more of the subjective equivalence score (PSE), discrimination threshold (JND = 0.6745σ), judgment boundary C, and psychometric function. The evaluation values ​​may also be response data. That is, the evaluation values ​​may be response data, one or more of the subjective equivalence score (PSE), discrimination threshold (JND = 0.6745σ), judgment boundary C, and psychometric function calculated from the response data, or both. In the left-eye change processing, the information processing unit 650 may obtain the above evaluation values ​​in association with the features of the left-eye image, and in the right-eye change processing, it may obtain the above evaluation values ​​based on binocular single-view of the subject TS, in association with the features of the right-eye image.

[0131] The information processing unit 650 may control the display positions of the left-eye image GL and the right-eye image GR on the display screen based on the inter-eye distance, such that the line of sight of the left eye passes through the geometric center of the left-eye image GL and the line of sight of the right eye passes through the geometric center of the right-eye image GR. The inter-eye distance is the distance between the right-eye monocular 211R and the left-eye monocular 211L.

[0132] The information processing unit 650 may determine, based on the image captured by the imaging unit 250 in the left eye gaze guidance unit, whether the center position of the pupil of the subject TS's left eye, the center of the left eye image, and the optical center of the visual acuity correction lens 212 in the left eye gaze guidance unit coincide. Similarly, the information processing unit 650 may determine, based on the image captured by the imaging unit 250 in the right eye gaze guidance unit, whether the center position of the pupil of the subject TS's right eye, the center of the right eye image, and the optical center of the visual acuity correction lens 212 in the right eye gaze guidance unit coincide. The information processing device 400 may output these results to the display device 300, another display device, or an external communication terminal.

[0133] Note that a contrast image may be used as the stimulus image instead of a blurred image. A contrast image is a stimulus image in which the feature quantity is the contrast ratio. For example, the image control unit 600 generates a square image of 600 pixels vertically and 600 pixels horizontally. The image control unit 600 creates a stimulus image by creating characters from the letter A onwards from left to right using a specific font with a side length of 104 pixels and arranging them on the square image. This stimulus image is set to have a contrast ratio as illustrated in Figure 21, based on the difference between the background pixel value and the character pixel value.

[0134] The image control unit 600 creates 10 stimulus images, each with a different contrast ratio as a feature quantity. Figure 22 shows an example of 10 contrast images. Each of the 10 contrast images shown in Figure 22 is labeled with an identification number from "01" to "10". This label indicates the order in which the contrast ratio of the images changes, with the image with the smallest feature quantity being labeled "01", and the numbers increasing step by step as the feature quantity increases, being labeled "02", "03", and so on. This makes it easy to identify which stage of feature quantity each presented image has, and allows for correspondence with the order of the images in experimental procedures and result analysis.

[0135] The following shows the results of measuring the perceptual characteristics of subjects (TS) using contrast images instead of blurred images as stimulus images. The method for measuring perceptual characteristics is the same as that used with blurred images. The feature quantity in these contrast images is the contrast ratio. When the contrast of the image that the subject TS is fixating on gradually changes, the way it appears also changes. In this case, when the contrast ratio is high, the image is usually easy to see and appears clear. On the other hand, when the contrast ratio is low, the image becomes difficult to see. Also, if the contrast ratios of the front and back images are equal or very close, it is expected that the subject TS will not be able to recognize the difference and will judge them to be the same.

[0136] First, Figures 23 to 26 show the results of measuring the perceptual characteristics of the right eye when the stimulus image is used as a contrast image and the second direction is set to 30 degrees. Figure 23 is an example of a graph of the posterior probability distribution of the subjective equivalence point (μ). The posterior probability distribution shown in Figure 23 was obtained from 3-point response data based on a cumulative normal distribution function. In this example, the information processing unit 650 obtains the posterior mode (3.52) of the subjective equivalence point (μ) by finding the maximum value of the posterior probability distribution shown in Figure 23. The 95% confidence interval for μ is in the range of 3.04 to 7.17. Based on this posterior probability distribution, it is possible to accurately evaluate the subjective equivalence point at which the subject TS judges that they cannot distinguish between the standard image and the comparison image.

[0137] Figure 24 is a graph showing the posterior probability distribution of the square root of the variance σ in an experiment using a cumulative normal distribution function. This posterior probability distribution is based on the response data of subjects' TS to a contrast image presented to the right eye when the gaze direction was 30 degrees. From the posterior probability distribution shown in Figure 24, the posterior mode of σ is 0.71. The 95% confidence interval is 0.40 to 5.95. Based on these results, it is possible to quantitatively evaluate the subjective variability of TS.

[0138] Figure 25 is a graph showing the posterior probability distribution of the judgment boundary C in a psychometric function using a cumulative normal distribution function. This posterior probability distribution is based on measurements performed using contrast images under the condition that the gaze direction to the right eye is 30 degrees. From the posterior probability distribution shown in Figure 25, the mode of the judgment boundary C is 1.14. The 95% confidence interval is 0.71 to 4.24. This result makes it possible to quantitatively evaluate how subjects TS judge differences in stimuli.

[0139] Figure 26 shows the results of psychophysical measurements performed using contrast images under the condition that the gaze direction to the right eye was 30 degrees. The graph shows the psychometric function that models the response probability using the cumulative normal distribution function. The results of this measurement were a subjective equivalence score (PSE) of 3.52, a judgment threshold (JND) of 0.48, and a decision boundary (C) of 1.14. Based on this psychometric function, it is possible to quantitatively evaluate how the subject's response (TS) changes in response to stimulus values.

[0140] Figure 27 shows the results of measuring the perceptual characteristics of the right eye using contrast images with the second direction set to 30 degrees, and the results of measuring the perceptual characteristics of the right eye with the second direction set to 0 degrees. Figure 27 shows the subjective equivalence point (PSE), RMS error, discrimination threshold (JND = 0.6745σ), and judgment boundary C as results of the perceptual characteristics measurement. Figure 27 also shows the psychometric function based on the results of measuring the perceptual characteristics of the right eye when the second direction is 0 degrees using contrast images.

[0141] As shown in Figure 27, the error between the PSE and the features of the standard image was 10.8% at 0 degrees in the right eye and 33.8% at 30 degrees in the right eye. The discrimination threshold (JND) was defined as the minimum value at which the subject TS perceives a difference in contrast ratio, and was measured at 0.59 at 0 degrees in the right eye and 0.48 at 30 degrees in the right eye. Furthermore, the judgment boundary (C) was 0.32 at 0 degrees in the right eye and 1.14 at 30 degrees in the right eye.

[0142] The execution order of operations, procedures, steps, and stages in the devices, systems, programs, and methods shown in the claims, specification, and drawings is not explicitly stated as "before," "prior to," etc. Furthermore, it should be noted that the execution order of each process can be implemented in any order, unless the output of a previous process is used in a later process. Even if the operation flow in the claims, specification, and drawings is described using phrases such as "first," "next," etc. for convenience, this does not mean that it is essential to perform the operations in that order. Also, to the extent permitted by law, the disclosures of Japanese Patent Application No. 2024-226337 and all documents cited in the embodiments described above are incorporated by reference as part of the text.

[0143] 100...Binocular vision function system, 200...Eye-tracking unit, 300...Display device, 400...Information processing device

Claims

1. A binocular vision function system for evaluating binocular monocular vision, comprising: a gaze guidance unit that fixes the subject's head in a first direction while directing the gaze directions of the subject's left and right eyes to a second direction different from the first direction; and a processing unit that spatially presents images separately to the subject's left and right eyes, whose gaze directions have been guided to the second direction by the gaze guidance unit, wherein the images have feature quantities as indicators of visual characteristics, and the feature quantities of the left-eye image, which is the image presented to the left eye, are different from the feature quantities of the right-eye image, which is the image presented to the right eye.

2. The binocular vision system according to claim 1, wherein the processing unit is capable of performing a left-eye change processing, which is a process of changing the feature quantities of the left-eye image in steps or continuously, and a right-eye change processing, which is a process of changing the feature quantities of the right-eye image in steps or continuously.

3. The binocular vision system according to claim 2, wherein the left eye change processing is a process of gradually or continuously changing the feature quantities of the left eye image while keeping the feature quantities of the right eye image fixed at a constant value, and the right eye change processing is a process of gradually or continuously changing the feature quantities of the right eye image while keeping the feature quantities of the left eye image fixed at a constant value.

4. The binocular vision function system according to claim 1, wherein the processing unit obtains an evaluation value based on the subject's binocular single-vision, obtained by presenting the left-eye image with the left eye and the right-eye image with the right eye.

5. The binocular vision function system according to claim 3, wherein the processing unit acquires an evaluation value based on the subject's binocular single-vision in the left eye change processing, and acquires an evaluation value based on the subject's binocular single-vision in the right eye change processing, and associates it with the feature quantities of the right eye image.

6. The binocular vision system according to claim 1, wherein the gaze guidance unit comprises a left-eye gaze guidance unit having a vision correction lens for the left eye and guiding the gaze direction of the left eye to the second direction, and a right-eye gaze guidance unit having a vision correction lens for the right eye and guiding the gaze direction of the right eye to the second direction, and in each of the left-eye gaze guidance unit and the right-eye gaze guidance unit, the gaze direction guided to the second direction is set to pass through the optical center of the vision correction lens.

7. A binocular vision system according to claim 6, comprising a display screen for displaying the left eye image and the right eye image, wherein the inter-eye distance, which is the distance between the left eye gaze guidance unit and the right eye gaze guidance unit, is adjustable to any distance, and the processing unit controls the display positions of the left eye image and the right eye image on the display screen based on the inter-eye distance such that the line of sight of the left eye passes through the geometric center of the left eye image and the line of sight of the right eye passes through the geometric center of the right eye image.

8. The binocular vision system according to claim 6, wherein each of the left eye gaze guidance unit and the right eye gaze guidance unit comprises an imaging unit for imaging the subject's eyeball, and the processing unit determines, based on the image captured by the imaging unit in the left eye gaze guidance unit, whether the center position of the pupil of the subject's left eye, the center of the left eye image, and the optical center of the vision correction lens in the left eye gaze guidance unit coincide, and determines, based on the image captured by the imaging unit in the right eye gaze guidance unit, whether the center position of the pupil of the subject's right eye, the center of the right eye image, and the optical center of the vision correction lens in the right eye gaze guidance unit coincide.

9. The binocular vision system according to claim 6, wherein each of the left eye gaze guidance unit and the right eye gaze guidance unit comprises a plurality of optical elements that reflect light at a certain angle, and the plurality of optical elements guide the gaze direction in the second direction.

Images