Image analysis method, image analysis device, program, and recording medium

Abstract

The present invention provides an image analysis method, an image analysis device, a program, and a recording medium. The image analysis method can accurately estimate the image signal value of an object displayed with a specific color material using the image signal value of a color chart. In the present invention, the spectral characteristics of the object displayed with the specific color material are obtained, and the spectral characteristics of each of a plurality of color charts formed with a color material different from the specific color material are obtained. The spectral characteristics of the object are approximated using an approximate formula that includes the spectral characteristics of each of the plurality of color charts as variables. The spectral characteristics of the object are obtained by photographing each of the plurality of color charts using a photographic device. For each color chart, an image signal value corresponding to the color of the photographed image is obtained. The image signal value of the object when photographed by the photographic device is estimated based on the image signal values ​​obtained for each color chart and the approximate formula.

Description

Technical Field

[0001] The present invention relates to an image analysis method, an image analysis device, a program, and a recording medium, and more particularly to an image analysis method, an image analysis device, a program, and a recording medium for estimating the image signal value of the color of an object displayed by a specific color material. Background Art

[0002] An object colored by a specific color material is photographed by a camera or other imaging device, and the color of the object is identified from the image. In some cases, the energy imparted to the object is determined from information related to the identified color.

[0003] Meanwhile, information obtained from photographic images, such as the image signal values ​​for each RGB (Red, Green, and Blue) color, can vary depending on the spectral sensitivity of the imaging device. Therefore, image signal values ​​obtained from capturing a particular object require calibration, taking into account differences in spectral sensitivity between imaging devices.

[0004] In the device (color processing device) described in Patent Document 1, a reference spectral reflectance is obtained by measuring a patch image using a reference spectrometer. A spot image is generated using a spectrometer used as a calibration target to obtain the calibration target spectral reflectance. Furthermore, a correction coefficient is generated for each wavelength based on the reference spectral reflectance, and the calibration target spectral reflectance is corrected using this correction coefficient for each wavelength. This allows for high-precision correction of the spectral reflectance measurement results of the calibration target device to account for differences in spectral sensitivity between device models.

[0005] Previous technical literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Laid-Open No. 2012-142920 Summary of the Invention

[0008] Technical issues to be solved by the invention

[0009] When calibrating each device to be calibrated, as described in Patent Document 1, it is necessary to photograph the object for each device and obtain calibration signal values ​​for each device. However, considering the durability of specific colorants contained in the object and the time required to prepare the object for color development, it is difficult to photograph the object and calibrate each device.

[0010] For the reasons described above, a color chart, such as printing ink, that is formed to the same color as the object using a color material different from a specific color material can be used in place of the actual object. However, if the spectral characteristics of the color material contained in the object differ from the color material used to form the color chart, the image signal values ​​obtained from an image of the color chart may differ from the image signal values ​​of the actual object. In other words, even the image signal values ​​obtained by capturing the color chart as a substitute for the object may not accurately reproduce the image signal values ​​of the object.

[0011] The present invention has been made in light of the above circumstances, and its object is to achieve the following objectives. Specifically, the present invention aims to resolve the above-mentioned problems in the prior art and to provide an image analysis method, image analysis device, program, and recording medium that can accurately estimate the image signal value of an object displayed with a specific color material based on the image signal value of a color chart.

[0012] Means for solving technical problems

[0013] In order to achieve the above-mentioned purpose, the image analysis method of the present invention is characterized in that it has: a first acquisition step of acquiring the spectral characteristics of an object colored by a specific color material; a second acquisition step of acquiring the respective spectral characteristics of a plurality of color charts formed by a color material different from the specific color material and having different colors; an approximation step of approximating the spectral characteristics of the acquired object by using an approximate formula including the respective spectral characteristics of the acquired plurality of color charts as variables; a third acquisition step of acquiring by photographing each of the plurality of color charts with a photographic device, and acquiring an image signal value corresponding to the color of the photographed image for each color chart; and an estimation step of estimating the image signal value when the object is photographed by the photographic device based on the image signal value of each color chart acquired in the third acquisition step and the approximate formula.

[0014] According to the image analysis method of the present invention, the image signal value when an object is imaged by an imaging device can be estimated accurately using the image signal values ​​of each of a plurality of color charts without depending on the spectral sensitivity of the imaging device.

[0015] Furthermore, in the image analysis method of the present invention, in the approximation step, the spectral characteristics of the acquired object are approximated by an approximation formula, and the approximation formula has a plurality of terms consisting of the respective spectral characteristics of the plurality of color charts acquired and coefficients multiplied by the respective spectral characteristics. In the estimation step, the image signal values ​​corresponding to the respective terms in the image signal values ​​of each color chart acquired in the third acquisition step are substituted into each of the plurality of terms included in the approximation formula, thereby also being able to estimate the image signal value when the object is photographed by the photographic device.

[0016] In the above configuration, by substituting the image signal values ​​of the plurality of color charts into the approximate formula, it is possible to more easily estimate the image signal value when the object is photographed by the photographing device.

[0017] Furthermore, in the image analysis method of the present invention, in the approximation step, it is more preferable to approximate the acquired spectral characteristics of the object by an approximation formula represented by a linear sum of the same number of terms as the plurality of color charts.

[0018] With the above configuration, it is possible to more easily estimate the image signal value when an object is photographed by the photographing device.

[0019] Furthermore, in the image analysis method of the present invention, in the approximation step, it is more preferable to set the coefficients of the plurality of terms to be within a range of -0.5 to 0.5.

[0020] By limiting the setting range of the coefficients of each term in the approximate expression, it is possible to suppress the influence of the coefficients that affect the estimation result of the image signal value of the object from becoming excessively large.

[0021] Furthermore, in the image analysis method of the present invention, in the third acquisition step, a camera having an image sensor built therein as an imaging device may be used to capture images of each of the plurality of color charts, and an image signal value may be acquired for each color chart.

[0022] As described above, by photographing a plurality of color charts using a camera having a built-in image sensor and acquiring an image signal value for each color chart, it is possible to easily acquire an image signal value for each color chart.

[0023] Furthermore, in the image analysis method of the present invention, the spectral reflectance of the object is acquired in the first acquisition step, and the spectral reflectance of each of the plurality of color charts may be acquired in the second acquisition step.

[0024] Alternatively, in the image analysis method of the present invention, the spectral transmittance of the object is acquired in the first acquisition step, and the spectral transmittance of each of a plurality of color charts may be acquired in the second acquisition step.

[0025] Furthermore, in the image analysis method of the present invention, the plurality of color cards may include a plurality of color charts having different colors.

[0026] According to the above configuration, the image signal value of the object can be estimated with high accuracy using the image signal values ​​of each of a plurality of color charts including color chips having different colors.

[0027] Furthermore, in the image analysis method of the present invention, the object may be a sheet that contains a specific coloring material and develops color according to the amount of external energy applied thereto.

[0028] According to the above configuration, the effect of the present invention of being able to estimate the image signal value of the object with high accuracy becomes more significant.

[0029] Furthermore, in the image analysis method of the present invention, the estimation step may estimate the image signal value of a first object to which a known amount of external energy has been applied. In this case, the image analysis method may further include: a fourth acquisition step of capturing an image of a second object to which an unknown amount of external energy has been applied using an imaging device, and acquiring the image signal value of the second object; and a prediction step of predicting the amount of external energy applied to the second object based on the acquired image signal value of the second object and the estimated image signal value of the first object.

[0030] With the above configuration, the image signal value of the second object can be estimated with high accuracy. Furthermore, based on the correspondence between the image signal value of the first object and the amount of external energy applied to the first object, the amount of external energy applied to the second object can be accurately predicted based on the estimation result of the image signal value of the second object.

[0031] Furthermore, in order to solve the aforementioned problems, the image analysis device of the present invention includes a processor, and the aforementioned image analysis device is characterized in that the processor performs the following processing: obtaining spectral characteristics of an object displayed in color by a specific color material, obtaining respective spectral characteristics of a plurality of colorimetric charts formed by color materials different from the specific color material and having different colors, approximating the spectral characteristics of the acquired object by using an approximate formula including the respective spectral characteristics of the acquired plurality of colorimetric charts as variables, obtaining the spectral characteristics by photographing each of the plurality of colorimetric charts by a photographic device, obtaining an image signal value corresponding to the color of the photographed image for each colorimetric chart, and estimating the image signal value when the object is photographed by the photographic device based on the image signal value of each acquired colorimetric chart and the approximate formula.

[0032] According to the image analysis device of the present invention, the image signal value when an object is imaged by an imaging device can be estimated with high accuracy and without depending on the spectral sensitivity of the imaging device, using the image signal values ​​of each of a plurality of color charts.

[0033] Furthermore, in order to solve the aforementioned problems, a program of the present invention is a program for causing a computer to execute each step of any of the above-described image analysis methods.

[0034] The program of the present invention enables the image analysis method of the present invention to be implemented on a computer. Specifically, by executing the program, the image signal values ​​of an object captured by a photographic device can be estimated with high accuracy and independently of the spectral sensitivity of the photographic device, using the image signal values ​​of each of a plurality of color charts.

[0035] Furthermore, a recording medium that can be read by a computer and on which a program for causing the computer to execute each step of any of the above-described image analysis methods is recorded can also be realized.

[0036] Effects of the Invention

[0037] According to the present invention, the image signal value when an object is imaged by an imaging device can be estimated with high accuracy and independently of the spectral sensitivity of the imaging device, using the image signal values ​​of each of a plurality of color charts.

[0038] Furthermore, according to the present invention, based on the correspondence between the image signal value of the object and the amount of external energy applied to the object, the amount of external energy applied to the object can be accurately predicted from the estimated image signal value of the object. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A diagram showing a color-developing object and a plurality of color charts.

[0040] Figure 2 This is a diagram showing the configuration of an image analysis device according to one embodiment of the present invention.

[0041] Figure 3 This is a diagram showing the functions of the image analysis device according to one embodiment of the present invention.

[0042] Figure 4 This is a diagram showing an example of the spectral reflectance of each of a plurality of color charts.

[0043] Figure 5 This is a diagram (Part 1) showing an example of the spectral reflectance of an object measured by a colorimeter.

[0044] Figure 6 This is a diagram (Part 1) showing an example of the spectral reflectance of an object approximated by the spectral reflectances of a plurality of color charts.

[0045] Figure 7 This is a diagram (part 2) showing an example of the spectral reflectance of an object measured by a colorimeter.

[0046] Figure 8 This is a diagram (part 2) showing an example of the spectral reflectance of an object approximated by the spectral reflectances of a plurality of color charts.

[0047] Figure 9 This is a diagram (part 3) showing an example of the spectral reflectance of an object measured by a colorimeter.

[0048] Figure 10This is a diagram (Part 3) showing an example of the spectral reflectance of an object approximated by the spectral reflectances of a plurality of color charts.

[0049] Figure 11 This is an equation for estimating the image signal value of an object.

[0050] Figure 12 This is a diagram showing the correspondence between the image signal value of an object and the amount of external energy applied to the object.

[0051] Figure 13 This is a diagram showing the flow of the image analysis process. DETAILED DESCRIPTION

[0052] With reference to the accompanying drawings, a specific embodiment of the present invention (hereinafter referred to as the present embodiment) will be described. However, the embodiment described below is merely an example for facilitating understanding of the present invention and does not limit the present invention. That is, the present invention can be modified or improved from the embodiment described below as long as it does not depart from its purpose. Furthermore, the present invention includes its equivalents.

[0053] Furthermore, in this specification, a numerical range expressed using “to” means a range including the numerical values ​​described before and after “to” as the lower limit and the upper limit.

[0054] In this specification, the term "color" refers to a concept that represents "hue," "saturation," and "lightness," and includes density (concentration) and hue.

[0055] [Regarding the Object of the Present Embodiment]

[0056] The object (hereinafter referred to as object S) of this embodiment is a sheet containing a specific color material, placed in a measurement environment, and exhibiting color depending on the amount of external energy applied thereto.

[0057] The “specific color material” is, for example, a color material composed of a color developer and a color developer microencapsulated in a support, and specifically, a color material composed of a color developer and a color developer described in Japanese Patent Application Laid-Open No. 2020-073907.

[0058] If the above-mentioned developer and color developer are coated on the object S and external energy is applied to the object S, the microcapsules are destroyed and the color developer is adsorbed on the color developer. Figure 1 As shown, the object S is colored in a predetermined color (strictly speaking, a predetermined hue). Furthermore, the number of microcapsules destroyed changes according to the external energy applied, thereby changing the color (strictly speaking, the concentration, hereinafter referred to as the color concentration) of the colored object S.

[0059] “External energy” refers to force, heat, magnetism, ultraviolet rays, infrared rays and other energy waves imparted to the object S in the measurement environment where the object S is placed. Strictly speaking, it is energy that causes color development of the object S (i.e., destruction of microcapsules in the object S) by imparting these energy.

[0060] Furthermore, the "amount of external energy" is, for example, the instantaneous magnitude of the external energy applied to the object S (specifically, force, heat, magnetism, energy waves, etc. acting on the object S). However, this is not limiting. If external energy is continuously applied to the object S, the cumulative amount applied over a predetermined period of time (i.e., the cumulative value of the amount of force, heat, magnetism, and energy waves acting on the object S) may be regarded as the external energy.

[0061] In order to measure the amount of external energy applied under a measurement environment, an object S is used. Specifically, external energy is applied to the object S to cause the object S to develop color, the colored object S is photographed by a photographic device, and the external energy can be estimated from the image signal value indicating the color (strictly speaking, the color density) of the photographed image.

[0062] Furthermore, when the external energy applied is non-uniform across various parts of the object S, each part of the object S develops color at a concentration corresponding to the external energy, resulting in a distribution of color density on the surface of the object S. Utilizing this phenomenon, a two-dimensional distribution of the external energy applied to the object S can be measured (predicted) from the distribution of color density on the surface of the object S.

[0063] The object S is preferably a sufficiently thin material that can be positioned well under the measurement environment, and can be composed of, for example, paper, film, or sheet. Furthermore, the application of the object S, in other words, the type of external energy to be measured (predicted) using the object S, is not particularly limited. For example, the object S can be a pressure-sensitive sheet that develops color when pressure is applied, a thermal sheet that develops color when heat is applied, or a photosensitive sheet that develops color when irradiated with light.

[0064] In addition, the following description will assume that the object S is a pressure-sensitive sheet and predict the magnitude or cumulative value of the pressure applied to the object S.

[0065] [Regarding the Image Processing Device of the Present Embodiment]

[0066] refer to Figure 2 and Figure 3 , the image analysis device of this embodiment (hereinafter referred to as image analysis device 10) will be described. Figure 2As shown, the image analysis device 10 is a computer equipped with a processor 11. The processor 11 is composed of dedicated circuits, such as general-purpose processors like CPUs (Central Processing Units) and FPGAs (Field Programmable Gate Arrays) whose circuit configuration can be modified after manufacture, such as programmable logic devices (PLDs), and circuit configurations specifically designed to perform specific processing, such as ASICs (Application Specific Integrated Circuits).

[0067] The processor 11 executes a series of processing for image analysis by executing the image analysis program. In other words, the processor 11 and the image analysis program work together to realize the image analysis. Figure 3 Specifically, the plurality of processing units shown are a spectral characteristic acquisition unit 21, a storage unit 22, an approximation unit 23, an image signal value acquisition unit 24, an estimation unit 25, and a prediction unit 26. These processing units will be described in detail later.

[0068] in addition, Figure 3 The multiple processing units shown may be composed of one of the multiple processors described above, or may be composed of a combination of two or more processors of the same or different types, such as a combination of multiple FPGAs or a combination of an FPGA and a CPU. Figure 3 The plurality of processing units shown may be constituted by one of the plurality of processors described above, or two or more processing units may be integrated into one processor.

[0069] Furthermore, for example, in computers such as servers and clients, it is considered that a processor is composed of a combination of one or more CPUs and software. Figure 3 Furthermore, a system on a chip (SoC) or the like may be considered, in which a processor realizes the functions of the entire system including the multiple processing units on a single IC (Integrated Circuit) chip.

[0070] Furthermore, the hardware configuration of the various processors described above may be a circuit (Circuitry) formed by combining circuit elements such as semiconductor elements.

[0071] The image analysis program executed by the processor 11 corresponds to the program of the present invention and is a program for executing each step of the image analysis flow described later in the processor 11 (specifically, Figure 13The image analysis program is recorded on a recording medium. The recording medium may be the memory 12 and storage 13 provided in the image analysis device 10, or a computer-readable medium such as a CD-ROM (Compact Disc Read Only Memory). Furthermore, a storage device included in an external device (e.g., a server computer) capable of communicating with the image analysis device 10 may be used as the recording medium, and the image analysis program may be recorded on the storage device of the external device.

[0072] The image analysis device 10 also includes an input device 14 and a communication interface 15. The image analysis device 10 receives user input via the input device 14 or communicates with other devices via the communication interface 15 to acquire various information. The information acquired by the image analysis device 10 includes information required for image analysis, and more specifically, information required for pressure measurement (pressure prediction) using the object S. This information includes, for example, the spectral characteristics of the object S and a plurality of color charts C (described later), as well as photographic images of the plurality of color charts C.

[0073] In this embodiment, the spectral characteristic is spectral reflectance, which can be measured by a known colorimeter 101 (eg, "eXact" and "i1PRO" from X-Rite, "FD-7" and "FD-5" from Konica Minolta, Inc.).

[0074] A photographic image is an image captured by a known photographic device 102 such as a digital camera, smartphone, or tablet computer, or a scanner. In this embodiment, a photographic image is obtained by digitizing a video signal output from an image sensor included in the photographic device 102 and compressing it in a predetermined format to form data. The photographic image data (hereinafter referred to as image data) represents the image signal value of each pixel. The image signal value corresponds to the color of the photographic image, specifically, the grayscale value of each pixel in the photographic image within a predetermined numerical range (e.g., 0 to 255 for 8-bit data).

[0075] In this embodiment, the imaging device 102 is a camera with a built-in imaging lens and image sensor (imaging element). Specifically, it produces a color image as a photographic image. Specifically, in this embodiment, the image sensor is a three-color RGB sensor, and the image signal values ​​represented by the image data are grayscale values ​​for each RGB color. However, this is not limiting, and the image signal values ​​may also be grayscale values ​​for a monochrome image (specifically, a grayscale image).

[0076] In addition, Figure 2In the illustrated embodiment, the image analyzing device 10, the colorimeter 101, and the imaging device 102 are separate, but this is not limiting. The image analyzing device 10 may have either or both the colorimeter function (i.e., the function of measuring spectral reflectance) and the imaging device function (i.e., the function of capturing images).

[0077] Furthermore, the image analyzing device 10 includes an output device 16 such as a display, and can output the image analysis results (for example, the prediction results of the pressure value described later) to the output device 16 to notify the user.

[0078] [Regarding conventional pressure measurement using an object]

[0079] Next, conventional pressure measurement using the object S will be described.

[0080] By applying pressure, the object S develops color at a color density corresponding to the pressure value. The pressure value corresponds to external energy and is the magnitude of instantaneous pressure or the cumulative amount of pressure when it is continuously applied over a predetermined time.

[0081] By utilizing this property, the pressure value applied to the object S can be measured from the color density of the object S. Specifically, the relationship between color density and pressure value, specifically the correspondence between image signal values ​​corresponding to each color density and pressure value, is determined in advance. The pressure value can then be predicted based on this correspondence from the image signal values ​​obtained by imaging the object S.

[0082] On the other hand, the image signal value depends on, for example, the spectral sensitivity of the imaging device. Therefore, when an object S is captured by an imaging device, the image signal value of the captured image must be corrected to a predetermined image signal value, specifically, a reference image signal value corresponding to the color density of the object S. To accurately perform this correction, it is necessary to produce a calibration sheet using, for example, a color material having the same spectral characteristics as the color material used for the object S (specifically, the color developer and color development agent described above).

[0083] However, considering the production cost, production time, durability, and discoloration (fading) of the colorant, it is difficult to produce correction sheets using the colorants described above. Therefore, substitutes such as printed materials that reproduce the same color (color density) as the color produced by the colorants are usually used. These substitutes are usually produced using a colorant different from the colorants described above, such as printing ink or paint. However, when the colorants are different, the spectral reflectance may be different. Therefore, when using substitutes such as printed materials for correction, it is necessary to consider the impact of the difference in spectral reflectance of the colorant between the substitute and the object S on the image signal value.

[0084] Conventionally, as described above, it was necessary to perform a correction taking into account the influence of differences in the spectral reflectance of the color material, and this correction had to be performed each time the imaging device used was changed. Consequently, conventional pressure measurement using the object S was time-consuming and could potentially reduce operational efficiency.

[0085] In contrast, the image analysis method of this embodiment, performed using the image analysis device 10, solves the aforementioned problems associated with conventional pressure measurement using the object S, and can accurately and efficiently perform pressure measurement using the object S. The functions of the image analysis device 10 and the image analysis method of this embodiment are described in detail below.

[0086] [Functions of the Image Analysis Device of This Embodiment]

[0087] The image analyzing device 10 includes a spectral characteristic acquiring unit 21, a storage unit 22, an approximation unit 23, an image signal value acquiring unit 24, an estimation unit 25, and a prediction unit 26 (see FIG. Figure 3 ).

[0088] The spectral characteristics acquisition unit 21 acquires the spectral characteristics of the colored object S, more specifically, the spectral reflectance of the colored portion of the object S. In this embodiment, when the colorimeter 101 measures the spectral reflectance of the object S, the spectral characteristics acquisition unit 21 receives the measurement result from the colorimeter 101 and thereby acquires the spectral reflectance of the object S. However, this is not limiting. For example, the spectral characteristics acquisition unit 21 may acquire the spectral reflectance of the object S by inputting the spectral reflectance of the object S measured by the colorimeter 101 through the input device 14. Alternatively, if information on the spectral reflectance of the object S is stored in an external computer, the image analysis device 10 may communicate with the external computer via the communication interface 15 to acquire the spectral reflectance of the object S.

[0089] The spectral characteristic acquisition unit 21 acquires the spectral reflectance for each color of the object S (specifically, each color density) and also acquires the spectral reflectance for each type of the object S (in other words, each type of color material used in the object S).

[0090] Furthermore, the spectral characteristic acquisition unit 21 acquires the spectral characteristics of each of the plurality of color charts C, specifically, the spectral reflectance. Figure 1As shown, the multiple color charts C include multiple color chips with different colors (hue, chroma, and lightness) and are formed from a color material different from the specific color material used in the object S, such as a universal color material such as printing ink or paint. Examples of the multiple color charts C include Macbeth Chart and X-rite's "ColorChecker." The number of color charts C used to obtain spectral reflectance is not particularly limited, but is preferably 10 or more, more preferably 15 or more, and even more preferably 20 or more.

[0091] Regarding the spectral reflectance of each of the plurality of color charts C acquired by the spectral characteristic acquisition unit 21, Figure 4 An example is shown in . Note that the procedure for acquiring the spectral reflectance of the color chart C is the same as that for the object S.

[0092] The storage unit 22 stores various information required for image analysis, specifically, information required for pressure measurement using the object S. The information stored in the storage unit 22 includes the spectral reflectance of the object S and the plurality of color charts C obtained by the spectral characteristic acquisition unit 21, and the approximation formula obtained by the approximation unit 23.

[0093] The approximation unit 23 approximates the spectral reflectance of the object S acquired by the spectral characteristics acquisition unit 21 using an approximation formula that includes the spectral reflectance of each of the plurality of color charts C acquired by the spectral characteristics acquisition unit 21 as a variable. In this embodiment, the approximation formula has a plurality of terms consisting of the spectral reflectance of each of the plurality of acquired color charts C and coefficients multiplied by each spectral reflectance. More specifically, the approximation formula is a polynomial expressed by the linear sum of the same number of terms as the plurality of color charts C, as shown in the following equation (1).

[0094] P=c1×R1+c2×R2+···+ci×Ri (1)

[0095] In the above formula (1), P on the left represents the spectral reflectance of the object S (strictly speaking, the approximate spectral reflectance), R1 to Ri on the right represent the spectral reflectance of the plurality of color charts C (strictly speaking, the acquired spectral reflectance), and c1 to ci on the right represent coefficients. i is a natural number greater than or equal to 3 and is the number of the plurality of color charts C.

[0096] The coefficients of each term are set so that the spectral reflectance of the object S approximated by the above-mentioned approximation formula is closest to the actual spectral reflectance, that is, the spectral reflectance acquired by the spectral characteristic acquisition unit 21 .

[0097] The specific method of setting the coefficients of each of the plurality of terms in the above-mentioned approximate expression is not particularly limited, and a known optimization method (eg, the least squares method) used to obtain the approximate expression can be used.

[0098] Furthermore, the coefficients of each term can be set within any range, but it is preferable to limit the setting range of the coefficients, and particularly preferably set within the range of -0.5 to 0.5.

[0099] The approximation unit 23 then sets a set of coefficients c1 to c1 in the above-mentioned approximation formula for each color (specifically, each color density) of the object S. Furthermore, the approximation unit 23 sets a set of coefficients c1 to c1 for each type of the object S (in other words, each type of color material used in the object S). Thus, an approximation formula, in other words, a set of coefficients c1 to c1, is set for each spectral reflectance of the object S acquired by the spectral characteristic acquisition unit 21.

[0100] in addition, Figure 5 、 Figure 7 and Figure 9 An example of the spectral reflectance of the object S acquired by the spectral characteristic acquisition unit 21 is shown in FIG. Figure 6 、 Figure 8 and Figure 10 The spectral reflectances of the same object S through a plurality of color charts C are shown in FIG. Figure 4 Spectral reflectance of the 18 color charts shown in C) Approximate spectral reflectance.

[0101] The image signal value acquisition unit 24 acquires, for each color chart, an image signal value (strictly speaking, an RGB image signal value) obtained by imaging each of the plurality of color charts C using the imaging device 102. In this embodiment, the plurality of color charts C are imaged using a camera serving as the imaging device 102, and the image signal value acquisition unit 24 receives image data of each color chart C from the camera, thereby acquiring an image signal value for each color chart. However, this is not limiting. For example, the image signal value acquisition unit 24 may acquire an image signal value by a user transferring image data of each color chart C from the camera to the image analysis device 10, or more specifically, by removing a recording medium from the camera and attaching it to the image analysis device 10.

[0102] Furthermore, the image signal value acquisition unit 24 acquires image signal values ​​obtained by imaging an object S (hereinafter referred to as a predicted object) placed in the measurement environment and whose color develops when pressure is applied, using the imaging device 102. The predicted object is the object S for which the pressure value of the applied pressure is unknown. The procedure for acquiring the image signal values ​​of the predicted object is the same as that for the color chart C.

[0103] The estimation unit 25 estimates the image signal value of the object S based on the image signal values ​​of each of the plurality of color charts C acquired by the image signal value acquisition unit 24 and the approximate formula set by the approximation unit 23. The image signal value of the object S estimated by the estimation unit 25 is assumed to be an image signal value obtained when the object S is imaged by the imaging device 102.

[0104] In this embodiment, the estimation unit 25 estimates the image signal value of the object S by substituting the image signal value corresponding to each term in the image signal values ​​of each color chart obtained by the image signal value acquisition unit 24 into each of the multiple terms included in the approximate equation. This allows the image signal value of the object S to be accurately estimated at any spectral sensitivity, regardless of the spectral sensitivity of the imaging device 102.

[0105] For simplicity, the image signal value of object S is determined by multiplying the spectral sensitivity (spectral sensitivity of the camera) at the time the image signal value is acquired by the spectral reflectance of object S. Here, the spectral reflectance P of object S is approximated by the spectral reflectances R1, R2, ..., Ri of multiple color charts C using the above-mentioned approximation. However, as can be seen from the following equation (2), the image signal value when object S is captured by a particular camera can be expressed as the linear sum of the products of the image signal values ​​when the camera is used to capture multiple color charts C and coefficients c1 to c1.

[0106] The image signal value of the object = [spectral sensitivity of the camera] × [spectral reflectivity of the object]

[0107] =[spectral sensitivity of the camera] × [c1 × R1 + c2 × R2 + ... + ci × Ri]

[0108] =c1×[image signal value of color material 1]+c2×[image signal value of color material 2]+···+ci×[image signal value of color material i] (2)

[0109] Through the above, such as Figure 11 As shown, the image signal value when the object S is imaged by a camera having any spectral sensitivity can be estimated by imaging a plurality of color charts C with each camera and obtaining the image signal value for each color chart.

[0110] If the Figure 11The estimation formula shown in FIG. 1 is used to explain this. In this formula, P1, P2, ..., Pn on the left are estimated values ​​of image signal values ​​when the object S is imaged by cameras (cameras #1 to #n) with different spectral sensitivities. Furthermore, R11, R21, R31, ..., Ri1 on the right are estimated values ​​of image signal values ​​for each of the multiple color charts C obtained by imaging them with camera #1. Similarly, R12, R22, R32, ..., Ri2 on the right are estimated values ​​of image signal values ​​for each of the multiple color charts C obtained by imaging them with camera #2. Finally, R1n, R2n, R3n, ..., Rin on the right are estimated values ​​of image signal values ​​for each of the multiple color charts C obtained by imaging them with camera #n.

[0111] In addition, n is a natural number greater than or equal to 2.

[0112] As described above, in this embodiment, it is not necessary to capture the object S each time the spectral sensitivity of the imaging device 102 is changed, nor is it necessary to prepare a calibration sheet, such as a printed material, in the same color as the object S. Therefore, even without adjusting the color (color, density, etc.) of the calibration sheet for each spectral sensitivity of the imaging device 102, it is possible to easily determine (estimate) the image signal value when capturing the object S at each spectral sensitivity.

[0113] Furthermore, when photographing with a camera, the spectral distribution of the illumination can change as the photographing environment changes. However, this embodiment can accommodate various spectral distributions of illumination. Specifically, by photographing multiple color charts C under a certain illumination to obtain image signal values, and substituting the image signal values ​​obtained for each color chart into the above approximate equation, it is possible to estimate the image signal value when the object S is photographed under the same illumination.

[0114] In this embodiment, the estimating unit 25 estimates the applied pressure value as the image signal value of a known object S (hereinafter referred to as the estimated object). The estimated object corresponds to the first object to which a known amount of external energy is applied. On the other hand, the predicted object corresponds to the second object to which an unknown amount of external energy is applied.

[0115] The prediction unit 26 predicts the pressure value applied to the prediction target object. In this embodiment, the prediction unit 26 estimates the pressure value applied to the prediction target object based on the image signal value of the prediction target object acquired by the image signal value acquisition unit 24 and the image signal value of the estimation target object estimated by the estimation unit 25.

[0116] Specifically, the estimated object is an object S to which the pressure value of the pressure applied as described above is known. Furthermore, there is a correlation between the image signal value of the estimated object and the pressure value of the pressure applied to the estimated object. That is, the pressure value is set to a plurality of values ​​to prepare a plurality of estimated objects (i.e., a plurality of objects S with the same color material but different color concentrations), and the image signal value of each estimated object is estimated in the above order. Thus, it is possible to determine the value of the image signal of the estimated object. Figure 12 The corresponding relationship between the image signal value and the pressure value is shown.

[0117] By determining the corresponding relationship, such as Figure 12 As shown, the pressure value (denoted by symbol Fa in the figure) of the pressure applied to the prediction object can be predicted from the image signal value (denoted by symbol Va in the figure) of the prediction object.

[0118] The pressure value predicted by the prediction unit 26, that is, the prediction result of the unknown pressure value to be applied to the prediction object, is output to the output device 16. This allows the user to confirm the predicted pressure value.

[0119] [Regarding the image analysis process of this embodiment]

[0120] Below, reference Figure 13 The image analysis process performed using the image analysis device 10 will be described. Figure 13 The image analysis flow shown is implemented using the image analysis method of the present invention. In other words, each step in the image analysis flow corresponds to each step constituting the image analysis method of the present invention.

[0121] In the image analysis process, the first acquisition step S001 and the second acquisition step S002 are first performed. In the first acquisition step S001, the spectral characteristics of the colored object S are acquired, specifically, the spectral reflectance measured by the colorimeter 101 is acquired. In this embodiment, in the first acquisition step S001, the spectral reflectance of the object S (i.e., the estimated object) is acquired for which the pressure value of the applied pressure is known.

[0122] In the second acquisition step S002, the spectral characteristics of each of the plurality of color charts C are acquired, specifically, the spectral reflectance measured by the colorimeter 101 is acquired. Figure 13 In the embodiment, the second acquisition step S002 is performed after the first acquisition step S001 is performed, but the present invention is not limited thereto. The second acquisition step S002 may be performed before the first acquisition step S001 or in parallel with the first acquisition step S001.

[0123] Next, an approximation step S003 is performed. In the approximation step S003, the spectral reflectance of the object S (strictly speaking, the estimated object) acquired in the first acquisition step S001 is approximated using an approximation formula that includes the spectral reflectance of each of the multiple color charts C acquired in the second acquisition step S002 as a variable. Specifically, as in the above-mentioned approximation formula (1), an approximation formula represented by a linear sum is obtained by adding multiple terms consisting of the product of the spectral reflectance of each color chart and a coefficient. Specifically, the coefficients of each term are set using an optimization method such as the least squares method, thereby approximating the spectral reflectance of the acquired object S (estimated object).

[0124] Furthermore, in the approximation step S003, the coefficients of each term in the approximation equation (i.e., a set of coefficients c1 to c1i) can be set within a predetermined numerical range, specifically preferably within the range of -0.5 to 0.5. In this case, in the subsequent estimation step S005, the influence of each coefficient on the estimated image signal value of the object S can be suppressed. If the setting range of the coefficients is unlimited and the coefficients are set to relatively large values, if the image signal values ​​of each color chart C substituted into the approximation equation contain errors, the influence of these errors on the estimation result will increase depending on the size of the coefficients. In contrast, if the coefficients of each term are set within the range of -0.5 to 0.5, the aforementioned influence caused by the errors can be further reduced.

[0125] Thereafter, a third acquisition step S004 is performed. In the third acquisition step S004, each of the plurality of color charts C is imaged by a camera having a built-in image sensor (strictly speaking, a color sensor), and an image signal value is acquired for each color chart.

[0126] After the third acquisition step S004 is performed, the estimation step S005 is performed. In the estimation step S005, the image signal value of the object S is estimated, more precisely, the image signal value of the estimated object, based on the image signal values ​​for each color chart obtained in the third acquisition step S004 and the approximation formula set in the approximation step S003. Specifically, the image signal value corresponding to each term among the image signal values ​​obtained for each color chart is substituted into each of the multiple terms included in the approximation formula. In this way, the image signal value of the estimated object is estimated.

[0127] Furthermore, in the image analysis process, the fourth acquisition step S006 is implemented. In the fourth acquisition step S006, the pressure value of the pressure applied to the unknown object S (i.e., the predicted object) is photographed by the same camera as that used in the third acquisition step S004, and its image signal value is acquired. Figure 13In the embodiment, the fourth acquisition process S006 is implemented after the estimation process S005 is implemented, but it is not limited to this. The fourth acquisition process S006 can be implemented before the estimation process S005 is implemented, for example, it can be implemented after the third acquisition process S004 is implemented or in parallel with the third acquisition process S004.

[0128] After the fourth acquisition step S006 is performed, the prediction step S007 is performed. In the prediction step S007, the pressure value of the pressure applied to the prediction target object is predicted based on the image signal value of the prediction target object acquired in the fourth acquisition step S006 and the image signal value of the estimation target object estimated in the estimation step S005. Specifically, based on the correspondence between the estimation result of the image signal value of the estimation target object and the pressure value applied to the estimation target object, the pressure value applied to the prediction target object is predicted from the image signal value of the prediction target object.

[0129] When the series of steps S001 to S007 described above are completed, the image analysis flow ends.

[0130] [Other embodiments]

[0131] The embodiment described above is a specific example given to illustrate the image analysis method, image analysis device, program, and recording medium of the present invention for easy understanding, but is merely an example, and other embodiments are also conceivable.

[0132] In the above embodiment, a series of steps of measuring the pressure of the object S, namely Figure 13 All steps S001 to S007 of the image analysis flow shown are executed by a computer constituting the image analysis device 10. However, this is not limiting. For example, the prediction step S007 of predicting the pressure value applied to the object S (strictly speaking, the prediction object) may be performed manually by a user rather than a computer.

[0133] Furthermore, in the above embodiment, spectral reflectance is acquired as the spectral characteristic of each of the object S and the plurality of color charts C. However, spectral characteristics other than spectral reflectance, such as spectral transmittance, may also be acquired. Specifically, in the first acquisition step S001 of the image analysis process, the spectral transmittance of the object S is acquired, and in the second acquisition step S002, the spectral transmittance of each of the plurality of color charts C may be acquired. In this case, in the approximation step S003, the acquired spectral transmittance of the object S is approximated using an approximation formula that includes the acquired spectral transmittance of each color chart C as a variable.

[0134] The spectral transmittance can be measured by a known transmittance measuring instrument (for example, “TLN-110 / 110v” manufactured by TOKAI OPTICAL CO., LTD. and “TLV-304-BP” manufactured by Asahi Spectra Co., Ltd.).

[0135] Furthermore, in the above embodiment, a plurality of terms consisting of the spectral reflectance and coefficient of each color chart are included. More specifically, the spectral reflectance of the object S is approximated by an approximation represented by a linear sum of the same number of terms as the plurality of color charts C. However, the present invention is not limited to this. For example, the spectral reflectance of the object S may be approximated by an approximation in which a single term includes the spectral reflectances of two or more color charts, or by an approximation represented in a form other than a linear sum.

[0136] Furthermore, in the above-described embodiment, the photographic device used to obtain the image signal values ​​of each of the object S (more specifically, the predicted object) and the plurality of color charts C is a camera with a built-in image sensor. However, this is not limiting, and a scanner may be used as the photographic device instead of the camera. When a scanner is used as the photographic device, changing the model of the scanner can change the distribution of spectral sensitivity and the spectral distribution of the illumination during photographing. In this case, if the image analysis method of the present invention is applied, it can correspond to the model of the changed scanner. That is, by photographing (reading) each of the plurality of color charts C with the scanner of the changed model to obtain the image signal value, and substituting the image signal value of each color chart obtained into the approximate formula, it is possible to estimate the image signal value when the object S is photographed (read) by the same scanner.

[0137] Explanation of symbols

[0138] 10-Image analysis device, 11-Processor, 12-Memory, 13-Storage, 14-Input device, 15-Communication interface, 16-Output device, 21-Spectral characteristic acquisition unit, 22-Storage unit, 23-Approximation unit, 24-Image signal value acquisition unit, 25-Estimation unit, 26-Prediction unit, 101-Colorimeter, 102-Photographic device, C-Colorimetric chart, S-Object.

Claims

1. An image analysis method, comprising: A first acquisition step of acquiring spectral characteristics of an object colored by a specific color material; a second acquisition step of acquiring spectral characteristics of each of a plurality of color charts formed with a color material different from the specific color material and having different colors; an approximation step of approximating the acquired spectral characteristic of the object using an approximation formula including the spectral characteristics of each of the plurality of acquired color charts as variables; a third acquisition step of acquiring each of the plurality of color charts by photographing the color charts with a photographing device, and acquiring an image signal value corresponding to a color of the photographed image for each color chart; and an estimating step of estimating the image signal value when the object is photographed by the imaging device based on the image signal value of each color chart acquired in the third acquiring step and the approximate expression; The object is a sheet containing the specific color material and developing color according to the amount of external energy when external energy is applied thereto. In the estimating step, the image signal value of the first object to which the external energy of a known amount is applied is estimated. The image analysis method has the following features: a fourth acquisition step of photographing a second object to which an unknown amount of external energy is applied by the imaging device to acquire the image signal value of the second object; and The prediction step predicts the amount of the external energy applied to the second object based on the acquired image signal value of the second object and the estimated image signal value of the first object.

2. The image analysis method according to claim 1, wherein: In the approximation step, the acquired spectral characteristic of the object is approximated by the approximation formula, the approximation formula having a plurality of terms consisting of the respective spectral characteristics of the plurality of acquired color charts and coefficients multiplied by the respective spectral characteristics. In the estimation process, the image signal value corresponding to each of the items in the image signal value of each color chart obtained in the third acquisition process is substituted into each of the multiple items included in the approximate formula, thereby estimating the image signal value when the object is photographed by the photographic device.

3. The image analysis method according to claim 2, wherein: In the approximation step, the acquired spectral characteristic of the object is approximated by the approximation formula represented by the linear sum of the same number of terms as the plurality of color charts.

4. The image analysis method according to claim 2 or 3, wherein: In the approximation step, the coefficient of each of the plurality of terms is set within a range of not less than -0.5 and not more than 0.

5.

5. The image analysis method according to any one of claims 1 to 3, wherein: In the third acquisition step, each of the plurality of color charts is imaged by a camera having a built-in image sensor as the imaging device, and the image signal value is acquired for each color chart.

6. The image analysis method according to any one of claims 1 to 3, wherein In the first acquisition step, the spectral reflectance of the object is acquired. In the second acquisition step, the spectral reflectance of each of the plurality of color charts is acquired.

7. The image analysis method according to any one of claims 1 to 3, wherein In the first acquisition step, the spectral transmittance of the object is acquired. In the second acquisition step, the spectral transmittance of each of the plurality of color charts is acquired.

8. The image analysis method according to any one of claims 1 to 3, wherein: The plurality of color charts include a plurality of color chips having different colors.

9. The image analysis method according to claim 1, wherein: In the prediction step, the amount of external energy applied to the second object is predicted based on the correspondence between the acquired image signal value of the second object and the estimated image signal value of the first object and the amount of external energy applied to the first object.

10. An image analysis device comprising a processor, wherein: The processor performs the following processing: Acquire the spectral characteristics of the object displayed by a specific color material, acquiring spectral characteristics of a plurality of color charts formed with color materials different from the specific color material and having different colors from each other, approximating the acquired spectral characteristics of the object using an approximation formula including the acquired spectral characteristics of each of the plurality of color charts as variables, The color charts are each photographed by a photographing device, and an image signal value corresponding to the color of the photographed image is obtained for each color chart. estimating the image signal value when the object is photographed by the imaging device based on the acquired image signal value for each color chart and the approximate expression; The object is a sheet containing the specific color material and developing color according to the amount of external energy when external energy is applied thereto. The processor estimates the image signal value of the first object to which the known amount of external energy is applied as the image signal value of the object, The processor further acquires the image signal value of the second object when the second object to which the unknown amount of external energy is applied is photographed by the imaging device. The amount of the external energy applied to the second object is estimated based on the acquired image signal value of the second object and the estimated image signal value of the first object.

11. The image analysis device according to claim 10, wherein: The processor predicts the amount of external energy applied to the second object based on a correspondence between the acquired image signal value of the second object and the estimated image signal value of the first object and the amount of external energy applied to the first object. 12 . A program product comprising a program for causing a computer to execute each step of the image analysis method according to claim 1 . 13 . A recording medium that can be read by a computer and records a program for causing the computer to execute each step of the image analysis method according to claim 1 .

Images