Patent Application
20230380678
The present disclosure relates to an information processing device, an information processing method, and a computer program.
An optical member such as a lens for a pair of glasses has optical characteristics such as spectral transmittance. The spectral transmittance of a lens for a pair of glasses changes a spectral distribution of light entering the eye, and so when a visual object is looked at by a person through a lens for a pair of glasses, the luminance and color thereof is changed. Color is, for example, expressed by tristimulus values in CIE XYZ color space. An optical member can be designed to match each individual by calculating the optical characteristics for a vision function of each individual based on objective quantitative evaluation. There is accordingly a proposal for technology capable of performing objective quantitative evaluation for each individual.
A vision function examination system disclosed in Patent Document 1 includes an analysis means to find at least two boundaries by finding a boundary between a luminance contrast region where a visual target is visible to a test subject and a luminance contrast region where the visual target is not visible to the test subject for each combination of stimuli values of color stimuli of a visual target in a coordinate system indicating a correlation between a luminance contrast value between a luminance of an attention portion and a luminance of a background portion, and a luminance average value of the visual target including the attention portion and the background portion.
Patent Document 1: International Publication (WO) No. 2018/012334
In the technology disclosed in Patent Document 1, although a visual acuity examination is performed on a test subject and a vision function of the test subject can be quantified thereby, objective quantitative evaluation is not performed in consideration of a combination of information about the visual acuity of the test subject and information about a visual object.
In consideration of the above circumstances, an object of the present disclosure is to provide an information processing device, an information processing method, and a computer program that enable an optimal optical member to be designed for each test subject in consideration of a combination of information about the visual acuity of the test subject and information about a visual object.
In order to achieve the above object, an information processing device according to the present disclosure includes a calculation unit that calculates an evaluation function using a visibility threshold of a test subject, obtained by performing a visual acuity examination on the test subject, and using a parameter related to a visual object, and a spectral transmittance optimization unit that optimizes a spectral transmittance of a filter having a specific spectral transmittance to light from the visual object using the evaluation function when passed through the filter.
The spectral transmittance optimization unit may be configured to determine the spectral transmittance of the filter such that a value of the evaluation function is maximized.
The spectral transmittance optimization unit may be configured to determine the spectral transmittance such that the value of the evaluation function is maximized from among spectral transmittances generated by combining pre-prepared dyes.
The spectral transmittance optimization unit may be configured to determine a spectral transmittance such that a value of the evaluation function is maximized using a specific numerical optimization algorithm.
The calculation unit may be configured to calculate the evaluation function for a specified part region of the visual object.
The calculation unit may be configured to calculate the evaluation function for a part region of the visual object as specified by a result of performing edge detection on the visual object.
The spectral transmittance optimization unit may be configured to optimize the spectral transmittance of the filter using the evaluation function when passed through the filter at a point in the region where a value of the evaluation function when not passed through the filter is maximized.
The spectral transmittance optimization unit may be configured to, in a case in which a first point that is a point in the region where the value of the evaluation function when not passed through the filter is maximized differs from a second point that is a point in the region where a value of the evaluation function when passed through the filter is maximized, optimize the spectral transmittance of the filter using the evaluation function at the first point.
The visibility threshold of the test subject may be a luminance contrast visibility threshold.
The luminance contrast visibility threshold may be generated by finding a boundary between a luminance contrast region where a visual target is visible to the test subject and a luminance contrast region where the visual target is not visible to the test subject in a coordinate system indicating a correlation between a luminance contrast value between a luminance of an attention portion and a luminance of a background portion, and a luminance average value of the visual target including the attention portion and the background portion.
The boundary may be generated by finding a visibility function that approximates to the one which takes visibility changes into account due to superposition of color of visual targets in the coordinate system.
The parameter related to the visual object may be a luminance contrast of the visual object calculated from spectral radiance of the visual object.
Moreover, in order to achieve the above object, an information processing method according to the present disclosure is a method that is executed by a computer. The method includes calculating an evaluation function using a visibility threshold of a test subject obtained by performing a visual acuity examination on the test subject and using a parameter related to a visual object, and optimizing a spectral transmittance of a filter having a spectral transmittance to light from the visual object using the evaluation function when passed through the filter.
Moreover, in order to achieve the above object, a computer program according to the present disclosure causes a computer to execute processing. The processing includes calculating an evaluation function using a luminance contrast visibility threshold of a test subject obtained by performing a visual acuity examination on the test subject and using a parameter related to a visual object, and optimizing a spectral transmittance of a filter having a spectral transmittance to light from the visual object using the evaluation function when passed through the filter.
The present disclosure thereby enables provision of an information processing device, an information processing method, and a computer program that enable an optimal optical member to be designed for each test subject in consideration of a combination of information about the visual acuity of the test subject and information about a visual object by using an evaluation function derivable from a visibility threshold of the test subject and a parameter related to a visual object.
Following description is regarding an example of an exemplary embodiment of the present disclosure, with reference to the drawings. Note that configuration element and parts that are the same or equivalent in the drawings are appended with the same reference numerals. Moreover, sometimes dimensional proportions have been exaggerated in the drawings for ease of explanation, and differ from actual proportions.
The information processing device 10 is a device for designing an optimal optical member for each test subject by calculating an evaluation value based on a vision function of the test subject as quantitatively evaluated by a visual acuity examination of the test subject, and based on information about a visual object. The visual acuity examination of the test subject may, for example, be performed using the technology disclosed in International Publication (WO) No. 2018/012334. The information processing device 10 may hold information about a visual acuity examination of the test subject as performed using the technology disclosed in WO No. 2018/012334, and may hold information about a visual acuity examination of the test subject as performed using such technology with a different device to the information processing device 10.
The input device 20 includes a pointing device such as a mouse and a keyboard, and is employed to perform various inputs. The input device 20 is employed in optical member design processing by the information processing device 10, and is employed to input spectral radiance information of an image. The spectral radiance information of images is obtained by measuring a given visual object using a spectral radiance meter. A two-dimensional spectral radiance meter may be employed as the spectral radiance meter. Imaging the visual object using a two-dimensional spectral radiance meter enables spectral radiance information for each pixel of a specific image size to be obtained as an image. Moreover, point measurement values for a given visual object may be collected using a spectral radiance meter to obtain luminance distribution information of the visual object.
The display device 30 is, for example, a liquid crystal display or an organic EL display, and displays various information. The display device 30 may employ a touch panel so as to also function as the input device 20. The display device 30 is employed during a vision function examination of the test subject, and is moreover also employed to display information when designing an optimal optical member for each test subject.
As illustrated in
The CPU 11 is a central processing unit that executes various programs and controls each section. Namely, the CPU 11 reads a program from the ROM 12 or the storage 14, and executes the program using the RAM 13 as workspace. The CPU 11 controls each of the above configuration and performs various calculation processing according to the program recorded on the ROM 12 of the storage 14. In the present exemplary embodiment there is an optical member design program for designing an optimal optical member for each test subject stored on the ROM 12 or the storage 14.
The ROM 12 is stored with various programs and various data. The RAM 13 serves as workspace to temporarily store a program or data. The storage 14 is configured by a storage device such as a hard disk drive (HDD), solid state drive (SSD), or flash memory, and is stored with various programs including an operating system, and various data.
The interface 15 is an interface for exchanging signals from other devices, such as the input device 20 or the display device 30, and employs, for example, a standard such as universal serial bus (USB), DVI-D, DVI-I, High Definition Multimedia Interface (HDMI, registered trademark), Ethernet (registered trademark), FDDI, Wi-Fi (registered trademark), or the like.
The information processing device 10 implements various functions when executing the above design program using the hardware resources described above. Explanation follows regarding functional configuration implemented by the information processing device 10.
As illustrated in
The calculation unit 101 calculates an evaluation value from an evaluation function based on visibility threshold data of a test subject obtained by performing a visual acuity examination on the test subject and based on two-dimensional spectral radiance image data of a visual object. The visual object may be pixels in an image. The calculation unit 101 calculates a value of an evaluation function FVIS (hereafter referred to as an evaluation value) as defined by the following Equation (1), wherein LCth is a luminance contrast visibility threshold calculated from visibility threshold data of a test subject, and LCobj is a luminance contrast calculated from two-dimensional spectral radiance image data of a visual object. The evaluation value is an index in which the larger the value, the more easily the visual object is seen by the test subject. Namely, the evaluation value becomes larger by the denominator LCth being smaller and the numerator LCobj being larger. In particular, an evaluation value of larger than 1 means that the visual object is visible the test subject, and an evaluation value of 1 or lower means that the visual object is not visible to the test subject. An objective of the optical member design program is to calculate a spectral transmittance of a lens for a pair of glasses (hereafter referred to as a filter) so as to raise the ease with which the visual object is seen by the test subject, by calculation using the evaluation function. Note that Equation (1) is disclosed as Visibility Level in “Study on visibility evaluation of circular visual targets for sighted subjects—research on visibility evaluation method using a luminance contrast evaluation graph assuming subjects with poor vision”, Part 2 by Yoko KATO, Yoshiki NAKAMURA and Michico IWATA in Journal of Japanese Environmental Engineering Vol. 83, No. 749, pp. 565-572, July 2018.
F
VIS=LCobj/LCth (1)
The visual acuity examination on the test subject may be performed using, for example, the technology disclosed in WO No. 2018/012334. The visibility threshold data of the test subject is obtained as a borderline luminance between a luminance of a visual target that was visible and a luminance of a visual target that was not visible, for example as disclosed in WO No. 2018/012334. Moreover, the two-dimensional spectral radiance image data of the visual object is obtained by measuring a given visual object at a specific measurement angle (for example 1°) using a spectral radiance meter.
A basis under which a visibility threshold curve L is defined is research disclosed in “Study on visibility evaluation of circular visual targets for sighted subjects—research on visibility evaluation method using a luminance contrast evaluation graph assuming subjects with poor vision, Part 1” by Yoko KATO, Yoshiki NAKAMURA and Michico IWATA in Journal of Japanese Environmental Engineering No. 743, p. 21-28, January 2018”, and “Visibility evaluation method considering disability glare using luminance contrast evaluation graph” by Yoko KATO, and Yoshiki NAKAMURA in Illuminating Engineering Institute of Japan, Volume 103, No. 2, pp. 67-74, February 2019”. In
The visibility threshold curve L illustrated in
LCth=LCth0+ΔCY+ΔCX+ΔCZ (2)
In above Equation (2), LCth0 is a term for the luminance contrast visibility threshold for cases in which a body equivalent to a background luminance of a visual target has been placed at a periphery of the visual target having a spectral distribution of tristimulus values imparted to the test subject equivalent to the spectral distribution of standard illuminant D65 of the International Commission on Illumination (CIE) (hereafter referred to as D65). ΔCY is a term to reflect an incremental threshold from superposition of the tristimulus values of D65 for cases in which a body having a different luminance to the background luminance of the visual target is placed at the periphery of the visual target. ΔCX and ΔCZ are visibility functions to approximate to points (visibility threshold data) that take into account a change in visual acuity due to color superposition, and are functions that express that visual acuity is different depending on the spectral distribution a visual target. More specifically, they are terms to reflect incremental thresholds from superposition in cases in which a body of spectral distribution that imparts tristimulus values different to the spectral distribution of D65 is placed at the periphery of the visual target.
The calculation unit 101 calculates the tristimulus values X, Y, Z for each pixel from the two-dimensional spectral radiance image data of the visual object. The spectral distribution is taken into account when calculating the evaluation function FVIS via the tristimulus values. The term LCth0 on the right hand side of Equation (2) is a function of variable Y and parameter α. The term ΔCY is a function of variable Y and parameter aY. The term ΔCX+ΔCZ are functions of variables X and Z, and parameters bX and bZ. Note that parameters α, aY, bX, bZ are parameters expressing the visual acuity of the test subject, and are found by fitting each function term to the luminance contrast visibility threshold obtained under the respective examination conditions described above. Namely, the parameters α, aY, bX, bZ have unique values peculiar to each test subject.
The evaluation value may be calculated for a specified part region of a visual object. When doing so, the region for calculation may be a part region of the visual object specified using the results of edge detection on the visual object, and may be specified by an examiner or the test subject.
When calculating the evaluation value, the calculation unit 101 uses the spectral distribution data resulting from application (multiplication) of a filter having a given spectral transmittance to the two-dimensional spectral radiance image data of the visual object, and calculates a Y tristimulus value for each pixel and calculates a luminance contrast LCobj of the visual object. Employing the spectral distribution data applied with the filter enables design of an optical member optimized for each test subject, namely, design of a filter having a spectral transmittance optimized for each test subject.
When calculating the evaluation value, the calculation unit 101 calculates the evaluation value for a given region within the two-dimensional spectral radiance image. For example, a region may be set for the body within the two-dimensional spectral radiance image, and a region may be set so as to include a boundary between bodies.
The calculation unit 101 first calculates the evaluation value for the spectral distribution of the visual object in a state in which a filter having given spectral transmittance is not applied, and extracts a point from the visual object where the evaluation value is maximized (first point). Reference here to a point means a position on the two-dimensional spectral radiance image. The calculation unit 101 then calculates the evaluation value for the spectral distribution of the visual object in a state in which the filter having the given spectral transmittance has been applied, and extracts a point from the visual object where the evaluation value is maximized (second point). In cases in which the first point and the second point are different, the calculation unit 101 calculates the evaluation value for the first point.
The spectral transmittance optimization unit 102 uses the evaluation value calculated by the calculation unit 101 to optimize the spectral transmittance of the optical member for applicability to the test subject having the luminance contrast visibility threshold LCth as expressed by the denominator of the evaluation function FVIS. For example, the spectral transmittance optimization unit 102 decides a filter having a spectral transmittance so as to maximize the evaluation value. For example, the spectral transmittance optimization unit 102 may decide the spectral transmittance for which the evaluation value is the maximum by employing a specific numerical optimization algorithm. For example, a solver function of Excel (registered trademark) software from Microsoft Corporation may be employed as the specific numerical optimization algorithm. Moreover, for example, the spectral transmittance optimization unit 102 may decide the spectral transmittance so as to maximize the evaluation value based on results of machine learning.
Note that even when the spectral transmittance can be decided so as to maximize the evaluation value, this does not necessarily mean that this spectral transmittance can be realized as this because of the combination of dyes actually available. The spectral transmittance optimization unit 102 may accordingly decide the spectral transmittance so as to maximize the evaluation value from out of spectral transmittances generated by finite combinations of pre-prepared dyes.
Following description is regarding a specific example of processing by the calculation unit 101 and the spectral transmittance optimization unit 102. For example, in cases in which 10 types of filter have been prepared that each have different respective spectral transmittances, the calculation unit 101 may apply the respective spectral transmittances to the two-dimensional spectral radiance image data of the visual object to obtain corresponding pairs of the 10 individual LCobj and LCth. The spectral transmittance optimization unit 102 employs the respective pairs of the 10 individual LCobj and LCth calculated by the calculation unit 101 to calculate the evaluation value, and selects the pair of LCobj and LCth so as to maximize the evaluation value. By deciding the filter having the spectral transmittance corresponding to the pair of LCobj and LCth that maximizes the evaluation value, the spectral transmittance optimization unit 102 optimizes the spectral transmittance of the filter for applicability to the test subject.
The information processing device 10 according to the present exemplary embodiment has such a configuration, and so is able to design an optimal optical member for each test subject using an evaluation function derivable from the luminance contrast visibility threshold of the test subject and luminance contrast of the visual object.
Next, following description is regarding operation of the information processing device 10.
The CPU 11 first inputs examination results for the test subject, namely visibility threshold data for the test subject, calculates the luminance contrast visibility threshold from this data, and calculates parameters α, aY, bX, bZ to express the visual acuity of the test subject by fitting the function LCth to the luminance contrast visibility thresholds (step S101). In addition to input of the visibility threshold data, the CPU 11 acquires the two-dimensional spectral radiance image of spectral radiance data measured for each pixel (step S102). Acquiring the two-dimensional spectral radiance image enables output of a RGB image, a log-average luminance image, and a luminance contrast image.
Continuing from step S102, the CPU 11 decides a region to calculate the evaluation value in the two-dimensional spectral radiance image data based on the examination result of the test subject (step S103). The CPU 11 may decide the region based on instruction from the examiner, and may decide the region based on image processing performed on the two-dimensional spectral radiance image.
Continuing from step S103, the CPU 11 extracts the point with the maximum evaluation value in the region decided at step S103 (step S104). Namely, the CPU 11 first performs calculation of the evaluation value for a case in which no filter is applied, and extracts the point with the maximum evaluation value therein. The CPU 11 calculates the evaluation value based on the above Equation (1) and Equation (2).
Although the evaluation value may be calculated for all of the points of the two-dimensional spectral radiance image 110, the efficiency of processing can be raised by limiting the calculation target. Reference numeral 115 indicates an example of a region where the evaluation value is calculated. The region 115 may be decided by being specified by the examiner or the test subject, and may be decided by performing edge detection in the two-dimensional spectral radiance image 110. For example, in cases in which the body A111A is a red body and the body B111B is a yellow body, the test subject may set the boundary between the body A111A and the body B111B when the test subject thinks they would like to see the boundary between the body A111A and the body B111B clearly. The evaluation value may then be calculated with the boundary portion set by the test subject as the calculation target. In the example of the two-dimensional spectral radiance image 110 illustrated in
Note that the evaluation value region 115 is not limited to the example illustrated in
Moreover similarly, although a luminance contrast LCobj may be calculated for all regions of the two-dimensional spectral radiance image 110, the region of the luminance contrast LCobj may also be restricted. Restricting the region of the luminance contrast LCobj enables more efficient processing.
Continuing from step S104, the CPU 11 reads spectral transmittance data of a filter for application to the two-dimensional spectral radiance image 110 (step S105). In cases in which spectral transmittance data are prepared for plural filters, the CPU 11 may read in any of the spectral transmittance data.
Continuing from step S105, the CPU 11 applies the spectral transmittance data read at step S105 to the two-dimensional spectral radiance image 110 (step S106). Applying the spectral transmittance data to the two-dimensional spectral radiance image 110 enables a filter-applied state to be output for a RGB image, a log-average luminance image, and a luminance contrast image.
Continuing from step S106, the CPU 11 extracts a point with the maximum evaluation value in the region decided at step S103 from the two-dimensional spectral radiance image 110 applied with the spectral transmittance data (step S107). The CPU 11 calculates the evaluation value based on the above Equation (1) and Equation (2).
Continuing from step S107, the CPU 11 compares the point with the maximum evaluation value extracted at step S107 against the evaluation value for different spectral transmittance data, and determines whether or not the optimum filter can be decided (step S108). The CPU 11 determines whether or not the optimum filter can be decided by determining whether or not the evaluation value is the maximum. For example, the CPU 11 sequentially reads in the spectral transmittance data for all filters, and takes the filter with the spectral transmittance resulting in the maximum evaluation value as the optimum filter. Alternatively, for example, the CPU 11 takes the filter with the spectral transmittance resulting in the maximum evaluation value from out of spectral transmittances generated by combining pre-prepared dyes as the optimum filter.
Note that in cases in which the point extracted at step S104 is different from the point extracted at step S107, the evaluation value at the same point as the point extracted at step S104 is employed for comparison.
The CPU 11 repeatedly performs the processing from step S105 to step S108 until the optimum filter can be decided. Then when determined that the optimum filter has been decided (step S108: Yes), the CPU 11 takes the decided filter as the optimum filter for the test subject (step S109).
By executing this cycle of processing, the information processing device 10 according to the present exemplary embodiment is able to use an evaluation function derivable from the luminance contrast visibility threshold of the test subject and from the luminance contrast of the visual object to design the optimal optical member function for each of the test subjects.
Although in the exemplary embodiment described above the information processing device 10 employs the luminance contrast visibility threshold as the visibility threshold for the test subject, and employs the luminance contrast of the visual object calculated from the spectral radiance of the visual object as a parameter related to the visual object, the present disclosure is not limited so such an example. For example, the information processing device 10 may employ a visibility threshold in color space as the visibility threshold of the test subject, and may employ color information of the visual object as the parameter related to the visual object. The color information of the visual object may, for example, be derived from the spectral radiance of the visual object.
Note that the optical member design processing executed in the exemplary embodiment described above by the CPU reading in software (a program) may be executed by various processors other than the CPU. Examples of such processors include programmable logic devices (PLD) that allow circuit configuration to be modified post-manufacture, such as a field-programmable gate array (FPGA), and dedicated electric circuits, these being processors including a circuit configuration custom-designed to execute specific processing, such as an application specific integrated circuit (ASIC). The optical member design processing may be executed by any one of these various types of processors, or may be executed by a combination of two or more of the same type or different types of processors (such as plural FPGAs, or a combination of a CPU and an FPGA). The hardware structure of these various types of processors is more specifically an electric circuit combining circuit elements such as semiconductor elements.
Moreover, although in the above exemplary embodiment an embodiment was described in which a program for the optical member design processing was pre-stored (installed) on in storage or ROM, there is no limitation thereto. The program may be provided in a format stored on a non-transitory recording medium such as a compact disk read only memory (CD-ROM), digital versatile disk read only memory (DVD-ROM), universal serial bus (USB) memory, or the like. Moreover, the program may be provided in a format downloadable from an external device over a network.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2020-177324 | Oct 2020 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2021/038456 | 10/18/2021 | WO |
