JP5761049B2 - Autism diagnosis support device and autism diagnosis support method - Google Patents

Description

  The present invention relates to an autism diagnosis support apparatus and an autism diagnosis support method for supporting diagnosis of autism.

  Autism is a type of developmental disorder, and the number of autistic patients has been increasing in recent years. However, it is said that social adaptability of autistic patients can be improved by detecting autism early and starting medical treatment. Therefore, in Japan, aiming for early detection of autism by conducting an interview with a doctor during a one-and-a-half year examination. However, diagnosis of autism requires skillful techniques, and there may be a difference in diagnosis results depending on doctors.

  Therefore, using the feature that autistic patients often do not see the eyes of others facing each other, one color camera that images the target person (for example, mother) and generates an image, and a color camera generate Display unit for displaying the image, two subject cameras for detecting the line of sight of the subject (infant) who sees the image displayed on the display unit, and two for the subject cameras for capturing the pupils of the subject A technology for assisting diagnosis of autism is disclosed based on how much the subject looks at the position of the eye of the subject in the image displayed on the display unit using the autism diagnosis support apparatus including (For example, Patent Document 1).

JP 2011-206542 A

  However, since the technique of Patent Document 1 uses five cameras, the cost is increased. In the technique of Patent Document 1, the position of the pupil of the subject imaged by the subject camera is coordinate-converted to the position in the image generated by the color camera, or based on the subject's gaze direction detected by the subject camera. Therefore, it is necessary to perform complicated calculation processing such as calculating which position on the image displayed on the display unit is being viewed.

  In view of such problems, the present invention provides an autism diagnosis support apparatus capable of deriving the position of a subject's line of sight without performing complicated calculation processing while reducing costs, and autism The purpose is to provide a diagnosis support method.

In order to solve the above problems, an autism diagnosis support apparatus according to the present invention includes a mirror, an infrared irradiation unit that irradiates infrared rays toward a subject located on the reflection surface side of the mirror, and both lateral sides of the mirror in the horizontal direction. Two imaging units that capture at least infrared reflected light reflected by the subject and generate two image data having horizontal parallax, and the pupil of the subject based on the image data generated by the imaging unit based position and pupil deriving unit that derives the, on the basis of the position of the pupil, which is derived by the pupil deriving unit, the line of sight deriving unit that derives the information of eye gaze of the subject, gaze information derived by the line-of-sight deriving unit Te, comprising: a determining viewpoint determining unit whether within a predetermined tracking range subjects viewpoint on the reflecting surface, the tracking range, position of the mirror image of the subject's pupil on the reflecting surface at the viewpoint of a subject Characterized in that the subject content is in the range-based.

  The horizontal width of the mirror may be less than the upper limit of the distance between the two imaging units that can generate image data from which the pupil deriving unit can derive the position of the pupil.

  A subject different from the subject is located with the subject on the reflecting surface side of the mirror, and the tracking range is a subject range that is a range based on the position of the mirror image of the pupil of the subject on the reflecting surface in the subject's view. You may do that.

  The pupil deriving unit counts the number of pupil images in the image data generated by the imaging unit, and determines the tracking range according to the number of pupil images counted by the pupil deriving unit, or the subject range or the subject range and the subject. A range determining unit that determines the range may be further provided.

  A range setting unit that sets the tracking range as one or both of the subject range and the subject range according to user input may be further provided.

  It is good also as providing the light emission part which irradiates light to the test subject side from the reflective surface side of a mirror, and the alerting | reporting part which promotes the visual recognition to a light emission part.

In order to solve the above problems, an autism diagnosis support method according to the present invention includes a mirror, an infrared irradiation unit that irradiates infrared rays toward a subject located on the reflection surface side of the mirror, and both lateral sides of the mirror in the horizontal direction. An autism diagnosis support method using two imaging units arranged to image at least reflected infrared light reflected by a subject and generate two image data having horizontal parallax, the imaging unit The position of the pupil of the subject is derived based on the image data generated by the above, the line-of-sight information indicating the line of sight of the subject is derived based on the derived position of the pupil, and the viewpoint of the subject is derived based on the derived line-of-sight information. Is determined to be within a predetermined tracking range on the reflecting surface, and the tracking range is a subject range that is a range based on the position of a mirror image of the pupil of the subject on the reflecting surface at the subject's viewpoint. When That.

  On the reflecting surface side of the mirror, a subject different from the subject is located together with the subject, count the number of pupil images in the image data generated by the imaging unit, and according to the number of counted pupil images, The tracking range is a subject range that is based on the position of the mirror image of the subject's pupil on the reflecting surface in the subject view, or a range that is based on the subject range and the position of the mirror image of the subject's pupil on the reflecting surface in the subject view. The target range may be determined.

  According to the present invention, it is possible to derive the position of the subject's line of sight without reducing complicated costs and performing complicated calculation processing.

It is a figure for demonstrating the arrangement | positioning relationship between the autism diagnosis assistance apparatus concerning 1st Embodiment, and a test subject. It is a block diagram for demonstrating the structure of the autism diagnosis assistance apparatus concerning 1st Embodiment. It is explanatory drawing for demonstrating the relationship between a cornea and an imaging part. It is a figure for demonstrating an infrared irradiation part. It is a figure for demonstrating generation | occurrence | production of ON / OFF of an infrared irradiation part, and the image data of an imaging part. It is a figure for demonstrating the coordinate derivation | leading-out process by a pupil derivation | leading-out part. It is a figure in the top view of a mirror for demonstrating the determination process of the tracking range by a range determination part. It is a flowchart which shows the flow of a process of the autism diagnosis assistance method using the autism diagnosis assistance apparatus concerning 1st Embodiment. It is a flowchart which shows the flow of a calibration process. It is a figure for demonstrating the arrangement | positioning relationship between a test subject and a subject, and the autism diagnosis assistance apparatus concerning 2nd Embodiment. It is a block diagram for demonstrating the structure of the autism diagnosis assistance apparatus concerning 2nd Embodiment. It is a figure in the side view of a mirror for demonstrating the determination process of the tracking range by a range determination part. It is a flowchart for demonstrating the flow of a process of the autism diagnosis assistance method using the autism diagnosis assistance device concerning 2nd Embodiment. It is a block diagram for demonstrating the structure of the autism diagnosis assistance apparatus concerning 3rd Embodiment.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The dimensions, materials, and other specific numerical values shown in the embodiments are merely examples for facilitating the understanding of the invention, and do not limit the present invention unless otherwise specified. In the present specification and drawings, elements having substantially the same function and configuration are denoted by the same reference numerals, and redundant description is omitted, and elements not directly related to the present invention are not illustrated. To do.

(First embodiment: autism diagnosis support apparatus 100)
FIG. 1 is a diagram for explaining the positional relationship between the autism diagnosis support apparatus 100 and the subject 10, and FIG. 2 illustrates the configuration of the autism diagnosis support apparatus 100 according to the first embodiment. It is a block diagram for.

  As shown in FIG. 1, a subject 10 (for example, an infant of about one and a half years) is positioned on the reflecting surface 110 a side of the mirror 110 constituting the autism diagnosis support apparatus 100, and is placed on both sides of the mirror 110. The subject 10 is imaged using the two arranged imaging units 114a and 114b (only the imaging unit 114a is shown in FIG. 1) to generate image data. The autism diagnosis support apparatus 100 derives line-of-sight information indicating the line of sight of the subject 10 based on the image data generated by the imaging units 114a and 114b.

  Autistic patients are characterized in that they often do not see the eyes of others who face each other as described above. In addition, regardless of whether or not the child has autism, infants about one and a half years old cannot distinguish themselves from others and often do not see their own eyes in the mirror 110. Therefore, the autism diagnosis support apparatus 100 determines whether or not the subject 10 is looking at the pupil (subject 10) and the vicinity of the pupil reflected in the mirror 110 based on the derived line-of-sight information. By referring to such a determination result of the autism diagnosis support apparatus 100, it can be used as an objective index when diagnosing whether or not the subject 10 has autism. Hereinafter, the autism diagnosis support apparatus 100 will be specifically described.

  As shown in FIG. 2, the autism diagnosis support apparatus 100 includes a mirror 110, an infrared irradiation unit 112, imaging units 114a and 114b, a central control unit 116, a light emitting unit 118, a notification unit 120, and a display. Part 122.

  In the present embodiment, the horizontal width of the mirror 110 is less than the upper limit value of the distance between the two imaging units 114a and 114b that can generate image data from which a pupil deriving unit 150 (to be described later) can derive the position of the pupil. As will be described later, the two imaging units 114 a and 114 b are arranged in the vicinity of both end portions 110 b of the mirror 110 with the same horizontal position.

  FIG. 3 is an explanatory diagram for explaining the relationship between the cornea and the imaging unit 114. As illustrated in FIG. 3A, when the distance between the image capturing units 114 is too large, the image capturing unit 114 may be able to capture human eyes depending on the viewing direction (the direction of the line of sight) of the human eye. In some cases, the cornea cannot be imaged. On the other hand, as shown in FIG. 3 (b), when the distance between the imaging units 114 is made smaller than in the case shown in FIG. 3 (a), there is a possibility that the imaging unit 114 can capture the cornea. It becomes higher than the case of 3 (a).

  The distance (baseline length) between the imaging units 114a and 114b and the angle with the eye will be described. When the eye is positioned at the center of the imaging unit 114a and the imaging unit 114b as shown in FIG. When the imaging units 114a and 114b are arranged at a position within the angle θ (θ = about 46 degrees) from the eye (center of the cornea), the imaging units 114a and 114b can generate image data including a pupil image. . That is, about 0.85 times the distance from the mirror 110 to the eye of the subject 10 is the upper limit value of the distance between the imaging units 114a and 114b.

  Therefore, by setting the horizontal width of the mirror 110 to be less than 0.85 times the distance from the mirror 110 to the eye of the subject 10, the imaging units 114a and 114b are arranged in the vicinity of both side end portions 110b of the mirror 110. However, image data including a pupil image can be suitably generated. At this time, the distance from the mirror 110 to the eyes of the subject 10 may be fixed using a chair or the like.

  The infrared irradiation unit 112 is disposed on both sides of the mirror 110 and irradiates infrared rays toward the subject 10. FIG. 4 is a diagram for explaining the infrared irradiation unit 112. As shown in FIG. 4, in this embodiment, the infrared irradiation unit 112 includes a first irradiation unit 112a (shown by hatching in FIG. 4) and a second irradiation unit 112b (shown by cross hatching in FIG. 4). Consists of. The first irradiation unit 112a is arranged at equal intervals in the circumferential direction on the outer peripheral position of the lens 160 of the imaging units 114a and 114b, and the second irradiation unit 112b is arranged on the side facing the lens 160 in the first irradiation unit 112a. Arranged at equal intervals in the direction. The 1st irradiation part 112a is comprised by LED (Light Emitting Diode) etc., and irradiates infrared rays whose center wavelength is 780 nm-less than 900 nm. The 2nd irradiation part 112b is comprised by LED etc. and irradiates infrared rays with a center wavelength of 900 nm-970 nm. Moreover, the 1st irradiation part 112a and the 2nd irradiation part 112b irradiate infrared rays so that it may become in parallel with the optical axis of the lens of the imaging parts 114a and 114b. Image data generated when the imaging units 114a and 114b capture the reflected light when the subject 10 reflects the infrared light irradiated by the infrared irradiation unit 112 will be described in detail later.

  The imaging units 114 a and 114 b are arranged on both sides of the mirror 110 with the same horizontal position. The imaging units 114a and 114b capture at least infrared reflected light reflected by the subject 10 in accordance with a frame synchronization signal from the central control unit 116, which will be described later, and generate two image data having horizontal parallax at, for example, 60 fps. To do. In the present embodiment, the imaging units 114a and 114b generate image data in both the odd field and the even field in the frame synchronization signal.

  Here, the nature of the human pupil will be described. When the first irradiation unit 112a that irradiates infrared rays having a wavelength of less than 900 nm irradiates the human with infrared rays and images the reflected light, the infrared rays having a wavelength of 900 nm or more are emitted. A bright pupil image (hereinafter referred to as a bright pupil image) is generated as compared with the case. Further, when the second irradiation unit 112b that irradiates infrared rays having a wavelength of 900 nm or more irradiates humans with infrared rays and images the reflected light, a dark pupil image that is a darker pupil image than the bright pupil image is generated. On the other hand, a substantially uniform image is generated in portions other than the human pupil, regardless of the wavelength of infrared rays.

  Therefore, when the difference between the generated image data including the bright pupil image and the image data including the dark pupil image is taken, the luminance is canceled out in the portion other than the pupil, and only the portion of the pupil image becomes clear. . Note that the bright pupil image in the image data is not necessarily brighter (higher brightness) than the surroundings depending on conditions such as ambient brightness, and the dark pupil image in the image data is necessarily darker (lower brightness) than the surroundings. However, the bright pupil image is a relatively bright image compared to the dark pupil image.

  Therefore, the first irradiating unit 112a and the second irradiating unit 112b irradiate infrared rays exclusively and alternately according to the frame synchronization signal from the central control unit 116. Specifically, in the odd field in the frame synchronization signal, the first irradiation unit 112a emits infrared light, and in the even field, the second irradiation unit 112b emits infrared light.

  FIG. 5 is a diagram for explaining ON / OFF of the infrared irradiation unit 112 and generation of image data of the imaging units 114a and 114b. As shown in FIG. 5, when the first irradiation unit 112a is turned on (irradiates infrared rays), the image data generated by the imaging units 114a and 114b includes a bright pupil image. When the second irradiation unit 112b is turned on (irradiates infrared rays), the image data generated by the imaging units 114a and 114b includes a dark pupil image. By doing so, it is possible to capture a bright pupil image and a dark pupil image with the imaging units 114a and 114b, respectively.

  Referring back to FIG. 2, the central control unit 116 is composed of a semiconductor integrated circuit including a CPU (Central Processing Unit), reads out programs and parameters for operating the CPU itself from the ROM, and serves as a RAM as a work area. In addition, the entire autism diagnosis support apparatus 100 is managed and controlled in cooperation with other electronic circuits. Further, the central control unit 116 functions as a pupil deriving unit 150, a line-of-sight deriving unit 152, a range determining unit 154, and a viewpoint determining unit 156.

  The pupil deriving unit 150 obtains a difference in luminance between the image data including the bright pupil image generated by the imaging unit 114a and the image data including the dark pupil image, and thereby the image data in which the pupil image is clarified (hereinafter, referred to as the pupil data). The image data is generated by obtaining a difference in luminance between the image data including the bright pupil image generated by the imaging unit 114b and the image data including the dark pupil image. Here, since the image data generated by the imaging unit 114a and the image data generated by the imaging unit 114b have horizontal parallax, the pupil deriving unit 150 determines whether the imaging unit 114a has a predetermined position (coordinates) of the imaging unit 114a. The position (coordinates) of the pupil is derived from the generated image data, the predetermined position (coordinates) of the imaging unit 114b, and the image data generated by the imaging unit 114b. In this embodiment, there is one coordinate system, and all coordinates can be represented by this one coordinate system.

  Here, the case of deriving the right eye pupil position will be described as an example, and the left eye pupil position deriving process is the same as the right eye pupil position deriving process, so a duplicate description will be given. Omitted.

  FIG. 6 is a diagram for explaining the coordinate deriving process by the pupil deriving unit 150. As shown in FIG. 6, in this embodiment, the pupil deriving unit 150 uses the center point of the connection between the principal point (optical center) of the lens 160 of the imaging unit 114a and the principal point of the lens 160 of the imaging unit 114b as the origin. P (0, 0, 0), the axis passing through the connection between the principal point of the lens 160 of the imaging unit 114a and the principal point of the lens 160 of the imaging unit 114b is the X axis (the imaging unit 114a side is + X), and the origin P The axis extending in the vertical direction is defined as the Y axis (vertically upward is + Y), and the axis passing through the origin P and perpendicular to the XY plane is the Z axis (reflecting surface 110a side is + Z). The coordinates of the reflection of the infrared ray irradiated by 112 on the cornea are derived.

  Specifically, the pupil deriving unit 150 obtains the imaging pupil vector A from the pupil image of the right eye in the image data generated by the imaging unit 114a and the predetermined coordinates of the imaging unit 114a, and is generated by the imaging unit 114b. An imaging pupil vector B is obtained from the pupil image of the right eye in the image data and predetermined coordinates of the imaging unit 114b. The pupil deriving unit 150 sets the coordinates of the intersection of the imaging pupil vector A and the imaging pupil vector B as the coordinates of the right-eye pupil.

  Returning to FIG. 2, the line-of-sight deriving unit 152 derives line-of-sight information indicating the line of sight of the subject 10 based on the position of the pupil of the subject 10 derived by the pupil deriving unit 150. Various existing technologies can be used to derive the line-of-sight information. For example, the pupil deriving unit 150 derives the position of corneal reflection in the vicinity of the pupil from the image data generated by the imaging units 114a and 114b, and the line-of-sight deriving unit 152 uses the technique described in Japanese Patent Laid-Open No. 2005-185431. Utilizing this, from the pupil position and the corneal reflection position of the subject 10 derived by the pupil deriving unit 150, a line-of-sight vector indicating the direction of the line of sight starting from the pupil position is derived as line-of-sight information.

  The range determination unit 154 determines a tracking range used by a viewpoint determination unit 156 described later based on the pupil position of the subject 10 derived by the pupil deriving unit 150. In the present embodiment, the tracking range is a subject range that is a range based on the position of the mirror image of the pupil of the subject 10 on the reflecting surface 110a when viewed by the subject 10.

  FIG. 7 is a top view of the mirror 110 for explaining the tracking range determination processing by the range determination unit 154.

  As shown in FIG. 7, when the subject 10 is looking at a mirror image of the pupil of the subject 10 on the reflecting surface 110 a of the mirror 110 when viewed by the subject 10, the position of the pupil of the subject 10 and the reflecting surface 110 a when viewed by the subject 10 are viewed. The position of the mirror image of the pupil of the upper subject 10 is perpendicular to the reflecting surface 110a. Therefore, when the subject 10 is looking at the mirror image of the pupil of the subject 10 on the reflecting surface 110a in the subject 10 view, the distance from the back surface 110c of the mirror 110 is the subject 10 and the reflecting surface 110a. It can be considered that the image 12 located at a distance equal to the distance to is seen. Therefore, when the coordinates of the pupil RE1 of the right eye of the subject 10 derived by the pupil deriving unit 150 are (xR1, yR1, zR1) and the coordinates of the pupil LE1 of the left eye are (xL1, yL1, zL1), the image The coordinates of the right eye pupil RM1 of 12 are (xR1, yR1, -zR1), and the coordinates of the left eye pupil LM1 of the image 12 are (xL1, yL1, -zL1).

  Moreover, the intersection of the perpendicular | vertical to the reflective surface 110a from the position of the pupil of the test subject 10 and the reflective surface 110a becomes the position of the mirror image of the pupil of the test subject 10 on the reflective surface 110a in the test subject 10 view. Therefore, the coordinates of the mirror image RRM1 of the pupil of the right eye of the subject 10 on the reflecting surface 110a in the view of the subject 10 are (xR1, yR1,0), and the coordinates of the mirror image LLM1 of the pupil of the left eye are (xL1, yL1,0). It becomes.

  Then, the range determining unit 154 determines a range based on the position of the mirror image RRM1 and the position of the mirror image LLM1 derived as described above as a subject range (tracking range). For example, the position of the mirror image RRM1 and the position of the mirror image LLM1 are set as the focal points, and the elliptical range in which the sum of the distances from the two focal points is the first predetermined distance is set as the subject range. Alternatively, an egg-shaped range including the position of the mirror image RRM1 and the position of the mirror image LLM1 is set as a subject range, a circular range that is a first predetermined distance from the position of the mirror image RRM1, and a second predetermined distance from the position of the mirror image LLM1. Or the subject range is the circular range.

  Returning to FIG. 2, the viewpoint determination unit 156 determines that the coordinates of the intersection (hereinafter simply referred to as the viewpoint) between the line-of-sight information (line-of-sight vector) derived by the line-of-sight deriving unit 152 and the reflecting surface 110a are within the subject range. It is determined whether or not.

  Then, the viewpoint determination unit 156 counts the number of image data in which the coordinates of the viewpoint are not within the subject range with respect to the image data generated within a predetermined time, and sets the time for two frames (1/30 sec) to the number of image data. The integrated value is derived by integration. Subsequently, the viewpoint determination unit 156 derives a value obtained by dividing the integrated value by a predetermined time, and determines that there is a suspicion of autism if the divided value is equal to or greater than a predetermined threshold.

  The light emitting unit 118 and the notification unit 120 are used for calibration processing to be described later. The light emitting unit 118 is configured by an LED or the like, and is provided at a total of four positions on the four corners of the mirror 110. The light emitting unit 118 irradiates light from the reflecting surface 110a side of the mirror 110 to the subject 10 side in accordance with a control command from the central control unit 116. The notification unit 120 is configured by an audio output device (speaker), and outputs sounds and sounds that prompt the light emitting unit 118 to be visually recognized. The calibration process will be described in detail later.

  The display unit 122 displays an image showing all the viewpoints within the predetermined time and the subject range, and an image showing the trajectories of all the viewpoints within the predetermined time and the subject range. For example, the central controller 116 regards the position (coordinates) of the pupil derived by the pupil deriving unit 150 as the position of the pupil image in the image data generated by the imaging units 114a and 114b, and the viewpoint derived by the viewpoint determination unit 156. The display unit 122 superimposes and displays the coordinate-converted viewpoint on the image data generated by the imaging units 114a and 114b. The display unit 122 is used when a doctor diagnoses autism or when the doctor explains the diagnosis result of autism to the parent of the subject 10. By providing the display unit 122, a doctor or a parent can visually recognize how much the subject 10 sees a mirror image of his / her own eye.

(Autism diagnosis support method)
FIG. 8 is a flowchart showing the flow of processing of the autism diagnosis support method using the autism diagnosis support apparatus 100 described above. First, the subject 10 is positioned on the reflecting surface 110a side of the mirror 110. Then, the central control unit 116 performs calibration (S200) related to the derivation of the viewpoint position (coordinates). The calibration process S200 will be described in detail later.

  When the calibration process S200 is completed, the infrared irradiation unit 112 emits infrared rays based on the frame synchronization signal transmitted by the central control unit 116, and the imaging units 114a and 114b capture the subject 10 and generate image data. (S202) The central control unit 116 starts viewpoint determination processing.

  When the generation of the image data is completed, the central control unit 116 sets the processing target to the first two frames (two frames each for the imaging units 114a and 114b, a total of four frames) (S204).

  The pupil deriving unit 150 derives the position of the pupil of the subject 10 from the image data generated in the data generation process S202 (S206), and derives the position of the corneal reflection of the subject 10 from the image data (S208).

  The line-of-sight deriving unit 152 has a line-of-sight vector (line-of-sight information) indicating the line of sight of the subject 10 based on the position of the pupil of the subject 10 and the position of corneal reflection derived in the pupil position derivation process S206 and the corneal reflection position derivation process S208. ) Is derived (S210).

  The range determining unit 154 determines a subject range (tracking range) based on the pupil position of the subject 10 derived in the pupil position deriving process S206 (S212). Then, the viewpoint determination unit 156 derives the coordinates of the intersection (viewpoint) between the line-of-sight vector derived in the line-of-sight derivation process S210 and the reflecting surface 110a (S214), and whether or not the derived viewpoint position is within the subject range. Is determined (S216).

  When the viewpoint determination unit 156 determines that there is no viewpoint of the subject 10 within the subject range (NO in S216), the time when the subject 10 does not have the viewpoint within the subject range (hereinafter simply referred to as out-of-range integration time) is derived. Therefore, the time for 2 frames (for example, 1/30 second) is integrated (S218). Then, the central control unit 116 integrates the time for two frames in order to derive the total time of the image data generated in the data generation process S202 (S220). When the viewpoint determination unit 156 determines that the viewpoint of the subject 10 is within the subject range (YES in S216), the total time integration processing S220 is performed without performing the out-of-range time integration processing S218.

  Then, the central control unit 116 determines whether or not the set frame is the last two frames of the image data generated in the data generation process S202 (S222). If it is not the last two frames (NO in S222), the central control unit 116 sets the processing target to the next two frames (S224). And the process after frame determination process S206 is repeated.

  If it is determined in the frame determination process S222 that it is the last two frames (YES in S222), the central control unit 116 determines the out-of-range integrated time derived in the out-of-range time integration process S218 in the total time integration process S220. A value divided by the total time of the derived image data is derived (S230), and the derived result is displayed on the display unit 122.

  As described above, according to the autism diagnosis support apparatus 100 and the autism diagnosis support method according to the present embodiment, the eye (face) to be viewed by the subject 10 is imaged by using the mirror 110. Even if there is no imaging unit for displaying the image and no display unit for displaying the image, the target eye (the eye of the subject 10) can be projected onto the mirror 110, and the cost can be reduced. In particular, when the subject 10 is an infant of about one and a half years old, it is difficult to distinguish the face between himself and the other person, so there is a high possibility of grasping his own eye projected on the mirror 110 as the other person's eye. . Therefore, even if there is no other person such as a guardian (mother, etc.) of the subject 10, it can be determined whether or not the subject 10 has seen his / her eyes projected on the mirror 110 by only the subject 10. It can be used as an index when diagnosing whether or not it is autism.

  In addition, since the imaging units 114a and 114b directly image the subject 10 without passing through a filter or a half mirror, the amount of light incident on the imaging units 114a and 114b is larger than when imaging through the filter or half mirror. Therefore, the SNR (Signal to Noise Ratio) of the image data generated by the imaging units 114a and 114b is good, and the pupil position and line-of-sight information can be derived with high accuracy.

  Furthermore, conventionally, since an imaging unit for imaging the eye (face) of the subject viewed by the subject 10 and two imaging units for deriving the viewpoint of the subject 10 are required, the coordinates of the target eye, A complicated calculation process related to coordinate conversion for matching the coordinates of the viewpoint of the subject 10 is required. However, since the autism diagnosis support apparatus 100 according to the present embodiment derives the coordinates of the target eye (pupil) and the coordinates of the viewpoint of the subject 10 using only the two imaging units 114a and 114b, the complexities relating to coordinate conversion are involved. Therefore, it is not necessary to perform a calculation process and the processing load can be reduced.

(Calibration process)
FIG. 9 is a flowchart showing the flow of the calibration process S200. Note that, in the calibration process S200, the processes S204, S210, S214, S222, and S224 in the above-described autism diagnosis support method are substantially the same, and thus the same reference numerals are given and redundant description is omitted.

  As shown in FIG. 9, in response to an operation input by a user (for example, a doctor or a laboratory technician), among the five light emitting units 118, the light emitting unit 118 located in the center when viewed from the subject 10 is caused to emit light (S240). And the alerting | reporting part 120 outputs the audio | voice which promotes the visual recognition to the light emission part 118 (S242). For example, when the subject 10 is an infant about one and a half years old, an animal call or a bell sound may be output. By doing so, it is possible to promote the visual recognition of the subject 10 to the mirror 110 and further to the visual recognition of the light emitting unit 118 by the light emitting unit 118 emitting light.

  Subsequently, for the calibration process, based on the frame synchronization signal transmitted by the central control unit 116, the infrared irradiation unit 112 emits infrared rays, and the imaging units 114a and 114b capture the subject 10 and image data. Is generated (S244). Then, a frame setting process S204, a pupil position deriving process S206, a corneal reflection position deriving process S208, a line of sight deriving process S210, and a viewpoint deriving process S214 are performed, and the central control unit 116 determines the coordinates of the viewpoint derived in the viewpoint deriving process S214. Is held (S246). Then, the frame determination process S222 and the frame setting process S224 are performed, and when the derivation of the coordinates of the viewpoint of the subject 10 is completed in all the image data generated in the data generation process S244 (YES in S222), the central control unit 116 Then, the average value of the coordinates held in the coordinate holding process S246 is derived (S248). Then, the central control unit 116 derives a deviation amount between the predetermined coordinates of the light emitting unit 118 and the average value of the derived coordinates as a correction value (S250). Thus, the derived correction value is used in the autism diagnosis support method described above.

  Since the optical axis of the eye and the line of sight differ depending on the person, it is possible to calibrate the deviation of the coordinates of the viewpoint for each subject 10 by performing calibration using the light emitting unit 118 and the notification unit 120 in this way. It becomes.

(Second Embodiment: Autism Diagnosis Support Device 300)
In the first embodiment described above, the autism diagnosis support apparatus 100 that determines whether or not the subject 10 has autism by only the subject 10 facing the mirror 110 has been described. In the second embodiment, the autism diagnosis for the subject 10 is supported even when the subject 20 (the partner that the subject 10 wants to look at, for example, a guardian) faces the mirror 110 together with the subject 10. The autism diagnosis support apparatus 300 will be described.

  FIG. 10 is a diagram for explaining the arrangement relationship between the subject 10 and the subject 20 and the autism diagnosis support apparatus 300, and FIG. 11 shows the autism diagnosis support apparatus 300 according to the second embodiment. It is a block diagram for demonstrating the structure of these.

  As shown in FIG. 10, a subject 10 (for example, an infant of about one and a half years) and a subject 20 (for example, a guardian of the subject 10) are placed on the reflecting surface 110 a side of the mirror 110 that constitutes the autism diagnosis support apparatus 300. ) Is positioned, and the subject 10 is imaged using the two imaging units 114a and 114b (only the imaging unit 114a is shown in FIG. 10) arranged on the back surface 110c side of the mirror 110 to generate an image. . The autism diagnosis support apparatus 300 derives line-of-sight information indicating the line of sight of the subject 10 based on the images generated by the imaging units 114a and 114b.

  As shown in FIG. 11, the autism diagnosis support apparatus 300 includes a mirror 110, an infrared irradiation unit 112, imaging units 114a and 114b, a central control unit 316, a light emitting unit 118, a notification unit 120, and a display. Part 122. The mirror 110, the infrared irradiation unit 112, the imaging units 114a and 114b, the light emitting unit 118, the notification unit 120, and the display unit 122 in the first embodiment described above have substantially the same functions, and thus have the same reference numerals. The central control unit 316 having a different function will be described in detail by omitting redundant description.

  The central control unit 316 is composed of a semiconductor integrated circuit including a CPU (Central Processing Unit), reads programs and parameters for operating the CPU itself from the ROM, and cooperates with the RAM as a work area and other electronic circuits. Thus, the entire autism diagnosis support apparatus 300 is managed and controlled. In addition, the central control unit 316 functions as a pupil deriving unit 350, a line-of-sight deriving unit 352, a range determining unit 354, and a viewpoint determining unit 356.

  The pupil deriving unit 350 obtains image data in which the pupil image is clarified by obtaining a difference in luminance between the image data including the bright pupil image generated by the imaging units 114a and 114b and the image data including the dark pupil image. Generate. Then, the pupil deriving unit 350 measures the number of pupil images in the image data in which the pupil image is clarified.

  Subsequently, when the number of pupil images is three or more, the pupil deriving unit 350 captures both eyes when the number of pupils is relatively large (the number of pupils is 1 (facing the side or the like). (When it is not possible), it is divided into a group) and a pupil group having a relatively small Y coordinate value. For example, the pupil deriving unit 350 derives the Y coordinate position of each pupil and derives the difference of the Y coordinates of the pupils. Then, the pupil deriving unit 350 groups the pupils whose Y coordinate difference is less than a predetermined threshold, and compares the maximum Y coordinate values in the grouped pupil groups, so that the Y coordinate value is relatively determined. It is divided into a large pupil group and a pupil group having a relatively small Y coordinate value.

  When the subject 10 and the subject 20 are opposed to the mirror 110, the subject 20 is more likely to be vertically above the subject 10. Therefore, the pupil deriving unit 350 defines the pupil group of the subject 20 as a pupil group having a relatively large Y coordinate (vertical direction) value and the pupil group of the subject 10 as a pupil group having a relatively small Y coordinate value. To do. In addition, even if either one of the subject 10 or the subject 20 is facing obliquely or one eye is collapsed, the subject 10 and the subject 20 are considered to have different vertical positions. There is no problem even if the pupil deriving unit 350 groups the pupils depending on the position in the Y-axis direction.

  The pupil deriving unit 350 derives the position of each pupil from the image data generated by the imaging units 114a and 114b, and derives the position of corneal reflection in the vicinity of the pupil. Since the coordinate deriving process by the pupil deriving unit 350 is substantially the same as the coordinate deriving process by the pupil deriving unit 150 described above, a duplicate description is omitted.

  The gaze deriving unit 352 derives gaze information (gaze vector) indicating the gaze of the subject 10 based on the position of the pupil of the subject 10 derived by the pupil deriving unit 350 and the position of corneal reflection. The line-of-sight information deriving process performed by the line-of-sight deriving unit 352 is substantially the same as the line-of-sight information deriving process performed by the line-of-sight deriving unit 152 described above, and thus redundant description is omitted.

  The range determination unit 354 determines a tracking range used by a viewpoint determination unit 356 described later based on the position of the pupil of the subject 10 derived by the pupil deriving unit 350 and the position of the pupil of the subject 20.

  FIG. 12 is a side view of the mirror 110 for explaining the tracking range determination processing by the range determination unit 354. In FIG. 12, it is assumed that the left eye exists behind the right eye, and reference numerals relating to the left eye are shown in parentheses. The range determination unit 354 determines the positions of the mirror images RRM1 and LLM1 of the pupil of the subject 10 on the reflection surface 110a in the view of the subject 10 and the positions of the mirror images RRM2 and LLM2 of the pupil of the subject 20 on the reflection surface 110a in the view of the subject 10. To derive. In FIG. 12, a mirror image RRM1 of the right pupil of the subject 10 and a mirror image RRM2 of the right pupil of the subject 20 are shown. In addition, the position of the mirror image RRM1 and LLM1 of the pupil of the subject 10 on the reflecting surface 110a in the view of the subject 10 is derived from the mirror image RRM1 of the pupil of the subject 10 on the reflecting surface 110a in the view of the subject 10 by the range determination unit 154 described above. Since the process is substantially the same as the derivation of the position of the LLM1, duplicate description will be omitted.

  As shown in FIG. 12, the position of the pupil of the subject 20 and the position of the mirror image of the pupil of the subject 20 on the reflecting surface 110a in the view of the subject 20 are perpendicular to the reflecting surface 110a. Therefore, when the subject 20 is looking at the mirror image of the pupil of the subject 20 on the reflecting surface 110a as viewed by the subject 20, the distance from the back 110c is the back 110c side of the mirror 110. It can be considered that the image 22 positioned at a distance equal to the distance between the reflecting surface 110a and the reflecting surface 110a is viewed. Therefore, when the coordinates of the pupil RE2 of the right eye of the subject 20 derived by the pupil deriving unit 350 are (xR2, yR2, zR2) and the coordinates of the pupil LE2 of the left eye are (xL2, yL2, zL2), The coordinates of the pupil RM2 of the right eye of the image 22 are (xR2, yR2, -zR2), and the coordinates of the pupil LM2 of the left eye of the image 22 are (xL2, yL2, -zL2). When the subject 10 is looking at the mirror image of the pupil of the subject 20 on the reflecting surface 110a, the subject 10 can be regarded as looking at the pupils RM2 and LM2 of the image 22 of the subject 20.

  Therefore, the range determination unit 354 determines the coordinates of the intersection of the line-of-sight vector C and the reflecting surface 110a when the subject 10 views the right eye pupil RM2 of the image 22 and the right eye of the subject 20 in the subject 10 view. This is a mirror image RRM2 of the pupil. That is, the X coordinate of the mirror image RRM2 is xR1 + (xR2-xR1) × zR1 / (zR1 + zR2), the Y coordinate is yR1 + (yR2-yR1) × zR1 / (zR1 + zR2), and the Z coordinate is 0 (zero). Similarly, the coordinate of the intersection of the line-of-sight vector and the reflecting surface 110a when the subject 10 views the left-eye pupil LM2 of the image 22 is the mirror image LLM2 of the left-eye pupil of the subject 20 in the subject 10 view. To do. That is, the X coordinate of the mirror image LLM2 is xL1 + (xL2-xL1) × zL1 / (zL1 + zL2), the Y coordinate is yL1 + (yL2-yL1) × zL1 / (zL1 + zL2), and the Z coordinate is 0 (zero).

  Therefore, the range determination unit 354 sets the range based on the position of the mirror image of the pupil of the subject 20 on the reflecting surface 110a in the subject 10 view as the subject range. For example, the position of the mirror image RRM2 and the position of the mirror image LLM2 are set as the focal points, and the elliptical range in which the sum of the distances from the two focal points is the third predetermined distance is set as the target person range. Alternatively, the egg-shaped range including the position of the mirror image RRM2 and the position of the mirror image LLM2 is set as the subject range, the circular range that is the fourth predetermined distance from the position of the mirror image RRM2, and the fourth predetermined distance from the position of the mirror image LLM2. Or the subject range is the circular range.

  Then, the range determination unit 354 determines the tracking range as a subject range, or a subject range and a subject range (hereinafter simply referred to as both ranges), according to the number of pupil images counted by the pupil deriving unit 350. . For example, the range determination unit 354 sets the tracking range as both ranges when the number of pupil images is 3 or more, and sets the subject range when the number of pupil images is less than 3.

  The range determination unit 354 can appropriately set the tracking range according to the number of persons facing the mirror 110 by changing the tracking range according to the number of pupil images counted by the pupil deriving unit 350. . When the number of pupil images is estimated to be 3 or more, together with the subject 10 on the reflective surface 110a side, the subject 10 appears on the reflective surface 110a by setting the tracking range to both ranges. Even if it is possible to determine the mirror image of itself and not to see the mirror image of its own pupil, it is possible to prompt the other person (subject 20) to see the mirror image, and to increase the probability of misdiagnosis of autism. It becomes possible to reduce. Moreover, the subject 20 often does not want to admit that there is a possibility of autism when the subject 10 is a relative such as his / her child. Therefore, the range determination unit 354 can notify the subject 20 that the subject 10 does not even see the subject 20's eyes by adding the subject range to the tracking range, and the subject 20 is automatically closed to the subject 10. It is possible to convince that there is a possibility of illness.

  The viewpoint determination unit 356 determines whether or not the viewpoint of the subject 10 is within a predetermined tracking range on the reflecting surface 110a based on the line-of-sight information (line-of-sight vector) derived by the line-of-sight deriving unit 352. Since the determination process by the viewpoint determination unit 356 is substantially the same as the determination process by the viewpoint determination unit 156 described above, a duplicate description is omitted.

(Autism diagnosis support method)
FIG. 13 is a flowchart for explaining the flow of processing of the autism diagnosis support method using the autism diagnosis support apparatus 300 described above. In the autism diagnosis support method according to the second embodiment, the processes S200 to S210 and S214 to S230 in the autism diagnosis support method of the first embodiment described above are substantially the same, and therefore the same. The duplicated explanation is omitted.

  As shown in FIG. 13, first, based on the frame synchronization signal transmitted by the central control unit 316, the infrared irradiation unit 112 irradiates infrared rays, and the imaging units 114a and 114b generate the image data by imaging the subject 10. (S300). Then, the pupil deriving unit 350 measures the number of pupil images in the image data (S302).

  Then, the range determination unit 354 determines whether or not the number of pupil images is 3 or more (S304), and if it is less than 3 (NO in S304), the range determination unit 354 determines Y from the maximum value of the Y coordinate value in each pupil. The minimum value of the coordinate values is subtracted, and it is determined whether the subtraction value is equal to or greater than a threshold value (S306). Then, if the subtraction value is less than the threshold value (NO in S306), range determination unit 354 determines that only subject 10 is opposite to reflection surface 110a, and range determination unit 354 sets the tracking range to the subject range. Determine (S308).

  If YES in the pupil number determination process S304 or YES in the subtraction value determination process S306, it is determined that the subject 10 and the subject 20 are opposed to the reflecting surface 110a, and the range determination unit 354 determines the tracking range. Are determined in both ranges (S310).

  When the range determination process is performed by the range determination unit 354 (S308, S310), the central control unit 316 performs the calibration process S200 and performs the processes S202 to S230.

  As described above, according to the autism diagnosis support apparatus 300 and the autism diagnosis support method according to the present embodiment, the pupil of the subject 10 and the target can be detected based on the difference in the coordinate values regarding the vertical direction in the coordinates of the derived pupil. Therefore, the two imaging units for deriving the position of the pupil of the subject 10 and the two imaging units for deriving the position of the pupil of the subject 20 as in the past are provided. It becomes possible to aggregate.

  Further, conventionally, an imaging unit for imaging the eye (face) to be viewed by the subject 10, two imaging units for deriving the viewpoint of the subject 10, and the position of the pupil of the subject 20 are derived. Therefore, a complicated calculation process related to coordinate conversion for matching the coordinates of the pupil of the subject 20 and the coordinates of the viewpoint of the subject 10 is necessary. However, the autism diagnosis support apparatus 300 according to the present embodiment derives the coordinates of the pupil of the subject 10, the coordinates of the pupil of the subject 20, and the coordinates of the viewpoint of the subject 10 using only the two imaging units 114a and 114b. Therefore, complicated calculation processing relating to coordinate conversion is not required, and the processing load can be reduced.

  In the present embodiment, the imaging units 114a and 114b need to detect the pupil and corneal reflection of the subject 10 with high accuracy (high resolution). On the other hand, for the pupil and corneal reflection of the subject 20, As long as the position of the pupil can be derived, it need not be more accurate than the subject 10. Therefore, the imaging units 114 a and 114 b are preferably installed at positions closer to the eye position (height) of the subject 10 than to the subject 20. In addition, when the imaging units 114a and 114b employ short-focus and fixed-focus lenses, a close target (subject 10) is captured large in the image data, and a distant target (subject 20) is captured small. . As described above, since it is sufficient that the position of the pupil can be derived for the target person 20, it may not be more accurate than the subject 10, so even if a short focus, fixed focus lens is employed, the target person With regard to the vicinity image of the 20 eyes, it is possible to secure the number of pixels necessary for deriving the pupil position.

(Third embodiment: autism diagnosis support apparatus 500)
In the second embodiment described above, the range determination unit 354 determines the tracking range as the subject range or both ranges according to the number of pupil images. In the third embodiment, an autism diagnosis support apparatus 500 that can reliably set a measurement target by setting a tracking range according to an operation input by a user (for example, a doctor or a laboratory technician) will be described.

  FIG. 14 is a block diagram for explaining a configuration of an autism diagnosis support apparatus 500 according to the third embodiment. As shown in FIG. 14, the autism diagnosis support apparatus 500 according to the third embodiment includes a mirror 110, an infrared irradiation unit 112, imaging units 114a and 114b, an operation unit 502, a central control unit 516, and the like. The light emitting unit 118, the notification unit 120, and the display unit 122 are included. Since the mirror 110, the imaging units 114a and 114b, the infrared irradiation unit 112, the light emitting unit 118, the notification unit 120, and the display unit 122 described above have substantially the same functions, the same reference numerals are given and redundant description is given. The operation unit 502 and the central control unit 516, which are omitted and have different functions, will be described in detail.

  The operation unit 502 includes an operation key, a cross key, a joystick, a touch panel superimposed on the display surface of the display unit 122, and the like, and receives an operation input from a user (for example, a doctor or a laboratory technician). When a remote controller for remote operation is provided, the remote controller also functions as the operation unit 502.

  The central control unit 516 is composed of a semiconductor integrated circuit including a CPU (central processing unit), reads programs and parameters for operating the CPU itself from the ROM, and cooperates with the RAM as a work area and other electronic circuits. Thus, the entire autism diagnosis support apparatus 500 is managed and controlled. The central control unit 516 functions as a pupil deriving unit 350, a line-of-sight deriving unit 352, a range setting unit 554, and a viewpoint determining unit 356. Note that the pupil deriving unit 350, the line-of-sight deriving unit 352, and the viewpoint determining unit 356 described above have substantially the same functions, and thus the same reference numerals are given and redundant description is omitted.

  The range setting unit 554 sets the tracking range to any one of the subject range, both ranges, and the subject range in accordance with an operation input to the operation unit 502 by the user. Thereby, since the tracking range can be set in advance according to the user input, it becomes possible to set the measurement target with certainty.

  As mentioned above, although preferred embodiment of this invention was described referring an accompanying drawing, it cannot be overemphasized that this invention is not limited to this embodiment. It will be apparent to those skilled in the art that various changes and modifications can be made within the scope of the claims, and these are naturally within the technical scope of the present invention. Understood.

  For example, the present invention is not limited to the above-described pupil position derivation method and line-of-sight derivation method, and other methods may be adopted as long as the pupil position, line-of-sight derivation, and viewpoint position can be derived.

  In the above-described embodiment, the horizontal width of the mirror 110 is set to be less than the upper limit value of the distance between the two imaging units 114a and 114b that can generate image data from which the pupil deriving units 150 and 350 can derive the position of the pupil. Therefore, the imaging units 114 a and 114 b are arranged in the vicinity of the both side end portions 110 b of the mirror 110. However, even if the horizontal width of the mirror is greater than or equal to the upper limit of the distance between the two imaging units 114a and 114b that can generate image data from which the pupil deriving units 150 and 350 can derive the position of the pupil, The imaging units 114a and 114b may be arranged below, or a mirror may be cut out to form a hole, and the imaging units 114a and 114b may be arranged in the hole. Even in this case, the base line length between the imaging units 114a and 114b is the upper limit value of the distance between the two imaging units 114a and 114b that can generate image data from which the pupil deriving units 150 and 350 can derive the position of the pupil. Less than.

  In the above-described embodiment, the case where both the first irradiation unit 112a and the second irradiation unit 112b are provided in the imaging units 114a and 114b has been described as an example. However, the first irradiation unit 112a and the second irradiation unit 112b are described as examples. There is no limitation in the installation position of the irradiation part 112b. For example, the first irradiation unit 112a may be provided in the imaging unit 114a, the second irradiation unit 112b may be provided in the imaging unit 114b, the second irradiation unit 112b may be provided in the imaging unit 114a, and the first irradiation unit 112a may be provided in the imaging unit 114b. May be provided.

  In the above-described calibration process, an example in which calibration is performed by turning on one light emitting unit 118 has been described. However, calibration may be performed by lighting any number of light emitting units 118 equal to or greater than two. . As the number of light emitting units 118 used for calibration is reduced, the time required for calibration can be shortened, and as the number of light emitting units 118 used for calibration is increased, the calibration accuracy can be improved. .

  In the above-described embodiment, the case where the central control unit 316 performs the calibration process on the subject 10 has been described as an example. However, the central control unit 316 also performs the calibration process on the subject 20. May be performed. By doing so, it is possible to accurately derive not only the position of the pupil of the subject 10 but also the position of the pupil of the subject 20.

  Note that each step in the autism diagnosis support method and calibration process of the present specification does not necessarily have to be processed in time series in the order described in the flowchart, and may include processes in parallel or by a subroutine. .

  The present invention can be used for an autism diagnosis support apparatus and an autism diagnosis support method for supporting diagnosis of autism.

100, 300, 500 ... Autism diagnosis support device 110 ... Mirror 110a ... Reflecting surface 112 ... Infrared irradiation part 114a ... Imaging part 114b ... Imaging part 118 ... Light emitting part 120 ... Notification part 122 ... Display part 150, 350 ... Pupil derivation Units 152 and 352 ... line-of-sight derivation units 154 and 354 ... range determination units 156 and 356 ... viewpoint determination unit 554 ... range setting unit

Claims (8)

With a mirror,
An infrared irradiation unit that irradiates infrared rays toward the subject located on the reflecting surface side of the mirror;
Two imaging units that are arranged on both sides of the mirror in the horizontal direction and that capture at least the reflected infrared light reflected by the subject to generate two image data having a horizontal parallax;
A pupil deriving unit for deriving a position of the pupil of the subject based on the image data generated by the imaging unit;
A line-of-sight deriving unit for deriving line-of-sight information indicating the line of sight of the subject based on the position of the pupil derived by the pupil deriving unit;
A viewpoint determination unit that determines whether or not the viewpoint of the subject is within a predetermined tracking range on the reflecting surface, based on the line-of-sight information derived by the line-of-sight deriving unit;
Equipped with a,
The autism diagnosis support apparatus , wherein the tracking range is a subject range that is a range based on a position of a mirror image of the subject's pupil on the reflecting surface at the subject's viewpoint .   The horizontal width of the mirror is less than an upper limit value of a distance between the two imaging units capable of generating image data from which the pupil deriving unit can derive the position of the pupil. Autism diagnosis support device. On the reflective surface side of the mirror, a subject different from the subject is located together with the subject,
The tracking range, autism according to claim 1 or 2, characterized in that as a range based on the position of the mirror image of the subject's pupil on the reflecting surface in visual point of the subject subject range Diagnosis support device. The pupil deriving unit counts the number of pupil images in the image data generated by the imaging unit,
The apparatus further comprises a range determining unit that determines the tracking range as the subject range or the subject range and the subject range according to the number of pupil images counted by the pupil deriving unit. Item 4. The autism diagnosis support apparatus according to Item 3 . 5. The apparatus according to claim 3 , further comprising a range setting unit configured to set the tracking range as one or both of the subject range and the target range according to a user input. Autism diagnosis support device. A light emitting unit that emits light from the reflecting surface side of the mirror to the subject side;
An informing unit for prompting the light emitting unit to be visually recognized;
Autism diagnosis support apparatus according to claim 1, any one of 5, characterized in that it comprises a. A mirror, an infrared irradiator for irradiating infrared rays toward a subject located on the reflecting surface side of the mirror, and an image of the infrared reflected light reflected by at least the subject, arranged on both sides of the mirror in the horizontal direction An autism diagnosis support method using two imaging units that generate two image data having horizontal parallax,
Based on the image data generated by the imaging unit, the position of the pupil of the subject is derived,
Based on the derived position of the pupil, deriving gaze information indicating the gaze of the subject ,
Based on the derived line-of-sight information, it is determined whether or not the subject's viewpoint is within a predetermined tracking range on the reflecting surface;
The autism diagnosis support method , wherein the tracking range is a subject range that is a range based on a position of a mirror image of the subject's pupil on the reflecting surface at the subject's viewpoint . On the reflective surface side of the mirror, a subject different from the subject is located together with the subject,
Count the number of pupil images in the image data generated by the imaging unit,
Depending on the number of the pupil image obtained by counting the tracking range, the subject content is in the range based on the position of the mirror image of the subject's pupil on the reflecting surface of the Viewpoint of the subject, or the subject range and the autism diagnosis support method according to claim 7, characterized in that determining the subject content is in the range based on the position of the mirror image of the subject's pupil on the reflecting surface of the Viewpoint of the subject.

Images