JP5876704B2 - Field of view measurement method and field of view measurement apparatus - Google Patents

Description

  The present invention relates to a visual field measurement method and a visual field measurement device.

There have been proposed methods and apparatuses for measuring a subject's visual field width and visual field defect. The visual field is the size of the visual angle in the vertical and horizontal directions that can process the visual stimulus with reference to the fovea at the center of the macular portion of the retina. The field of view is considered as follows. The visual field of a healthy person is about 60 degrees on the upper side in the vertical direction and about 75 degrees on the lower side. In the horizontal (left-right) direction, in the case of a single eye, the nose side is about 60 degrees and the ear side is about 100 degrees. Therefore, with both eyes, a viewing angle of about 120 degrees is observed in the left and right with the front as the center. Of these, the central circular visual field of about 20 degrees from the fovea is called the central visual field, and the donut-shaped visual field up to about 120 degrees around it is called the peripheral visual field. In general, the central visual field has higher spatial resolution than the peripheral visual field, and color vision is lost in the peripheral visual field.
For example, if the subject develops visual impairment such as glaucoma, there may be a defect in a part of the visual field, so it is possible to examine the degree of visual impairment by measuring the visual field it can.

  In general, the visual field is obtained by measuring the maximum range of the region that can be recognized by the subject by moving the visual stimulus that is the target without moving the eye to be examined. The visual field measurement methods are roughly classified into static visual field measurement and dynamic visual field measurement. Static visual field measurement is a method for examining retinal sensitivity while gradually increasing the brightness of a fixed point. Dynamic visual field measurement is a method for examining visual function by moving a light emitting point. Of the dynamic visual field measurements, a method for quantitatively measuring the entire visual field is called dynamic quantitative visual field measurement. It is common to use a Goldman perimeter for dynamic visual field measurement.

  Patent Document 1 describes a visual field measurement method in which a subject displays a fixed viewpoint (a viewpoint requiring attention) and moves a light emitting point serving as a target in a spiral around the fixed viewpoint. In this method, the subject recognizes the presence of the light emitting point at the moment when the light emitting point enters the peripheral visual field from outside the visual field range. The test subject operates the operation unit such as a push button without delay and replies that the light emitting point is recognized. Thereby, the outermost position of the peripheral vision of the subject is acquired.

  Patent Document 2 describes a visual field inspection system that tracks a subject's eye movement with respect to a visual stimulus and determines that the subject has recognized the visual stimulus when the viewpoint moves in the moving direction of the visual stimulus. According to this system, it is supposed that the area where the subject can visually recognize the visual stimulus can be objectively determined without depending on the subject's subjective answering operation. In this system, the display position of the next visual stimulus is determined based on the distance between the current viewpoint position of the subject and the viewpoint requiring attention.

JP 2007-0775350 A Special table 2010-526623

  The breadth and narrowness of a person's visual field fluctuates significantly depending on psychological factors such as whether or not they are paying attention to a visual target. In particular, it is empirically known that when the object to be viewed is a human face, the human instinctively focuses on that eye. On the other hand, when performing various actions on a person as an observation target, such as beauty judgment of a person's face or analysis of a person's movement, the entire face or face is included instead of focusing on a part of the face locally It may be preferable to observe the whole body. This will be described in detail using a reference experiment described later. And it is extremely useful to measure a field of view (hereinafter, narrow field of view) narrowed by a psychological factor due to the fact that the object to be viewed is not the medical maximum range of the peripheral field of view of the subject.

  On the other hand, the measurement of the visual field in Patent Document 1 and Patent Document 2 described above is to measure the maximum range of the peripheral visual field of the subject from a medical viewpoint. Therefore, these methods are performed in a state in which the subject's psychological factors are eliminated, in which the visual stimulus is moved along a spiral or linear geometric shape on a plain background. Therefore, in the inventions of Patent Literature 1 and Patent Literature 2, it is impossible to measure a narrow visual field due to the psychological factor of the subject.

  The present invention has been made in view of the above-described problems, and provides a method and apparatus for quantitatively measuring a subject's visual field in a special situation in which the human visual field tends to become psychologically narrow. is there.

  The visual field measurement method of the present invention presents a face image including a human face to a subject, and provides visual stimulation at a plurality of positions at different distances from the point of interest determined on the face region in the face image. The visual field of the subject is measured based on the generation position of the visual stimulus recognized by the subject.

  In the visual field measurement device of the present invention, the distance between the face image presenting means for presenting a face image including a human face to the subject and the point of interest determined on the face area in the face image is different from each other. Stimulus generating means for generating visual stimuli at a plurality of positions.

  According to the said invention, a test subject is in the state which tends to become a narrow visual field instinctively by visually observing a face image. In this state, when the point of interest on the face area is viewed, the subject's field of view tends to be particularly narrow. Thus, by forming a special situation in which the subject's field of view is particularly likely to become psychologically narrow, by measuring the subject's field of view, it is possible to suitably determine the visual tendency of the subject with respect to the face and other objects. .

  ADVANTAGE OF THE INVENTION According to this invention, a test subject's visual field in the special condition where a human visual field tends to become psychologically narrow can be measured quantitatively.

It is a block diagram which shows an example of the visual field measuring apparatus concerning embodiment of this invention. It is a flowchart of the visual field measurement method concerning this embodiment. It is a figure which shows an example of a face image. It is a figure which shows a face image and a 1st visual stimulus. It is a figure which shows a face image and a 2nd visual stimulus. It is a flowchart which shows the modification of a visual field determination step. It is a figure which shows the landscape image which does not contain a human face. It is a figure which shows a plain image. It is the scatter diagram which plotted the measurement result of each subject's viewing angle about the subliminal moving picture and the fade-out moving picture concerning a face image. It is explanatory drawing of a test subject's viewing angle. It is the scatter diagram which plotted the measurement result of the viewing angle of each subject about the subliminal moving picture and the fade-out moving picture concerning the landscape image. It is the scatter diagram which plotted the measurement result of the viewing angle of each subject about the subliminal moving picture and the fade-out moving picture concerning the plain image. It is a figure which shows the face sample image used for the reference experiment 2. FIG. It is a graph which shows the illusion accuracy rate (%) of each group. A graph showing the change in the illusion accuracy rate when the distance between the subject's eyes and the face image is changed for a group of five people with a large viewing angle and a group of five people with a small viewing angle for the subliminal video of the face image. It is. A graph showing the change in the illusion accuracy rate when the distance between the subject's eyes and the face image is changed for a group of five people with a large viewing angle and a group of five people with a small viewing angle for a fade-out video of a face image It is. It is a graph which shows transition of the illusion accuracy rate when changing the stimulus presentation time for a group of five people with a large viewing angle and a group of five people with a small viewing angle for a subliminal video of a face image. It is a graph which shows transition of the illusion correctness rate when changing the stimulus presentation time for a group of five people with a large viewing angle and a group of five people with a small viewing angle for a fade-out moving image of a face image.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same reference numerals are given to the same components, and the description will be omitted as appropriate.

(Visual field measuring device)
FIG. 1 is a configuration diagram showing an example of a visual field measuring apparatus 100 according to an embodiment of the present invention.

First, an outline of the present embodiment will be described.
The field-of-view measuring apparatus 100 includes a plurality of different distances between the face image presentation unit 10 that presents the face image 50 including a human face to the subject S and the point of interest RP defined on the face area in the face image 50. And a stimulus generation unit 60 for generating a visual stimulus VS at the position.

  The visual field measuring apparatus 100 reads a computer program and executes a corresponding processing operation, so that a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), an I / F (Interface) unit, Hardware configured with general-purpose devices such as, a dedicated logic circuit configured to execute a predetermined processing operation, a combination thereof, and the like.

  The visual field measurement device 100 further includes a line-of-sight acquisition unit 20, a line-of-sight data storage unit 30, and an expert data storage unit 32. The face image presentation unit 10, the line-of-sight acquisition unit 20, the line-of-sight data storage unit 30, the expert data storage unit 32, and the stimulus generation unit 60 are connected to each other.

  The face image presentation unit 10 is an image display device such as a liquid crystal display or a screen. The face image presentation unit 10 may be planar or hemispherical. The face image presenting unit 10 is arranged facing the subject S. A face image 50 is displayed on the face image presentation unit 10. The subject S looks at the face image 50.

  The face image 50 is set with a point of interest RP. This point of interest RP may be a target displayed on the face image 50 or a virtual point that is not displayed on the face image 50. The point of interest RP is preferably set near the eyes in the face image 50. Since the subject S instinctively pays attention to the eyes in the face image 50, the viewpoint of the subject S is easily concentrated on the point of interest RP by setting the point of interest RP in the eye or in the vicinity thereof.

  While the subject S is viewing the face image 50, the stimulus generator 60 generates a visual stimulus VS on the face image 50 in a predetermined pattern. Specifically, the stimulus generator 60 is a CPU. The stimulus generator 60 generates a visual stimulus VS below the perceptual threshold in the peripheral visual field of the subject S, both in the peripheral visual field of the subject S and outside the visual field.

  Human central vision has high spatial and temporal resolution and low perceptual threshold. On the other hand, the peripheral visual field has a lower spatial resolution and temporal resolution than the central visual field, and the perception threshold is higher than that of the central visual field. In the present embodiment, since the visual stimulus VS lower than the perceptual threshold in the peripheral visual field is generated outside the visual field of the subject S or in the peripheral visual field, the visual stimulus VS generated in the peripheral visual field for the subject S is a sub-threshold stimulus. Become. Therefore, the subject S does not recognize the visual stimulus VS generated outside the visual field, and recognizes the visual stimulus VS generated in the peripheral visual field unconsciously.

  Note that the stimulus intensity of the visual stimulus VS may be an upper threshold (primary) stimulus in the central visual field or a lower threshold (subliminal) stimulus in the central visual field.

  The stimulus generator 60 may generate a visual stimulus VS in the central visual field of the subject in addition to the out-of-view and peripheral vision. However, when the visual stimulus VS of the upper threshold (primary) stimulus in the central visual field is generated in the central visual field, the subject S senses the visual stimulus VS with consciousness, and the psychology of paying attention to the point of interest RP is alleviated. The Therefore, when the visual stimulus VS is generated in the central visual field, the stimulation intensity of the visual stimulus VS is set to the threshold in the central visual field, or after the visual stimulation VS is generated a predetermined number of times in the out-of-view and peripheral visual fields, It is preferable to generate a visual stimulus VS in the central visual field. Thereby, the range of the peripheral visual field in the state where the subject S is paying attention to the point of interest RP can be measured without the subject S being unconscious.

  For the same reason, the generation position of the visual stimulus VS recognized by the subject S after the generation of the visual stimulus VS is generated intermittently or continuously without notifying the subject S over the predetermined stimulus time. It is preferable to obtain an answer regarding the subject from the subject S. Thereby, it is possible to measure the visual field of the subject S while maintaining the state where the subject S instinctively pays attention to the attention point RP of the face image 50 and has a narrow visual field. Specifically, based on the answer from the subject S, the face image 50 is visually observed by analyzing how far the visual stimulus VS is recognized by the subject S from the point of interest RP. The maximum range of the peripheral visual field of the subject S in the state can be determined.

  The line-of-sight acquisition unit 20 is means for acquiring line-of-sight data VD related to the line-of-sight direction of the subject over a predetermined stimulation time in which visual stimulation is generated intermittently or continuously. The line-of-sight data storage unit 30 is storage means for storing line-of-sight data VD acquired from the subject S. An eye detection device (eye camera) can be used for the line-of-sight acquisition unit 20. As an example, a gaze point tracking system (eye tracker) manufactured by Tobii can be suitably used.

  The expert data storage unit 32 is a storage means for storing the result of measuring the visual field as described above (expert line-of-sight data PD) with a beauty professional as a subject. Usage modes of the line-of-sight acquisition unit 20, the line-of-sight data storage unit 30, and the expert data storage unit 32 will be described later.

(Field of view measurement method)
Hereinafter, a visual field measurement method (hereinafter sometimes referred to as the present method) performed using the visual field measurement device 100 of the present embodiment will be described. FIG. 2 is a flowchart of the method.

  An outline of this method will be described. In this method, a face image 50 including a human face is presented to the subject S (FIG. 2: step S10), and a plurality of distances from the point of interest RP determined on the face area in the face image 50 are different from each other. A visual stimulus VS is generated at the position (FIG. 2: step S30), and the visual field of the subject S is measured based on the generation position of the visual stimulus VS recognized by the subject S (FIG. 2: step S70).

  Next, this method will be described in detail. FIG. 3 is a diagram illustrating an example of the face image 50. The face image 50 is an image including an object that can be recognized as a human face as a subject. In addition to the human head, a doll image or computer graphics simulating the human head may be used.

  In the present method, the visual stimulus VS is generated on the region of the hair HR (the hair region R1) in the face image 50 and on the external region (the region outside the head R2) of the head HD including the face FC and the hair HR. A visual stimulus VS may be further generated on the area of the face FC.

  The point of interest RP is set between both eyes in the face image 50 or in the vicinity of the eyes EY. In the present method, the vicinity of the eye EY means the inside of the circumscribed circle OC of both eyes in the face image 50 or on the circumscribed circle OC. The point of interest RP of the present embodiment is set between both eyes in the face image 50. The point of interest RP of the present embodiment is a target displayed on the face image 50.

  As described above, the visual stimulus VS generated in the face image 50 by this method is a stimulus having a subthreshold intensity in the peripheral visual field of a general person (subject S) and a threshold intensity in the central visual field. . The specific aspect of the visual stimulus VS is not particularly limited. The visual stimulus VS may be a point, a minute region that can be regarded as a point, or a line or a surface region.

  The size of the face FC in the face image 50 is preferably approximately the same size as a general adult face. As a result, the field of view when the subject S views the face image 50 is substantially the same as the field of view when the subject S views an actual person.

  Here, generating the visual stimulus VS in the face image 50 means that the visual stimulus VS is visually recognized in a state of being superimposed on the face image 50 when the subject S views the face image 50. As a specific example, the pixel value of the pixel is set such that the ratio or inverse ratio of the brightness or color component of one or more pixels in the face image 50 to the brightness or color component of the surrounding pixels is equal to or greater than a predetermined value. Can be mentioned. In addition, the visual stimulus VS may be a visual target such as a light emitter arranged on the surface of the face image 50 or in a spatial gap between the face image 50 and the eye of the subject S.

  Hereinafter, this method exemplifies two types of visual stimulation VS. The first visual stimulus VS1 is instantaneous light emission of a light spot. By setting the luminous intensity and the light emission time of the light spot to be equal to or less than predetermined values, the stimulus intensity of the first visual stimulus VS1 can be suppressed below the perception threshold in the peripheral visual field of the subject S. The second visual stimulus VS2 is a change in the pixel value of the face image 50. By setting the change rate of the pixel value in each pixel to a predetermined value or less and the moving speed of the region in which the pixel value is changed to a predetermined value or less, the stimulus intensity of the second visual stimulus VS2 is below the perception threshold in the peripheral visual field of the subject S. Can be suppressed.

  In the visual field measurement device 100 of the present embodiment, the mode (mode) of the visual stimulus VS is selected by the operation of the operation unit (not shown) by the operator (FIG. 2: step S10).

  In step S10, the timing of presentation of the face image 50 to the subject S and the timing for selecting the stimulation mode are arbitrary. Furthermore, the start timing of the generation of the visual stimulus VS (step S30) and the timing before and after the timing of presenting the face image 50 to the subject S are also arbitrary. That is, in this method, generation of the visual stimulus VS may be started after the face image 50 is presented to the subject S, or the face image 50 in which the generation of the visual stimulus VS has already started is presented to the subject S. Also good.

(First visual stimulus)
FIG. 4 is a diagram illustrating the face image 50 and the first visual stimulus VS1. As described above, the first visual stimulus VS1 shown in FIG. 4 is light emission of light spots 1 to 8 (illustrated by white circles) that are turned on at the light intensity and light emission time below the perceptual threshold in the peripheral visual field of the subject S. More specifically, a plurality of light spots 1 to 8 that are different from each other in at least one of a distance or a direction from the point of interest RP are sequentially turned on, and one or a plurality of light spots 1 to 8 in which the subject S recognizes the light emission. And the field of view of the subject S is determined based on the maximum distance from the point of interest RP.

  In FIG. 4, a concentric circle CC centered on the point of interest RP is illustrated for convenience, but the concentric circle CC is not presented to the subject S. The light spots 1 to 8 are respectively arranged on concentric circles CC having different diameters. The smaller the number of the light spots 1 to 8, the larger the concentric circle CC is. Accordingly, by emitting light spots 1 to 8 in numerical order and intermittently, the light spots are visually recognized as if moving in a spiral shape with a diameter decreasing from the outer edge of the face image 50 toward the point of interest RP. A first visual stimulus VS1 is generated.

  The operator instructs the subject S to pay attention to the eye EY or the point of interest RP of the face image 50. While the subject S is looking at the eye EY or the point of interest RP, the light spot 1 is allowed to emit light for a short time (single light emission time). Such light emission may be on or flashing. The single light emission time is preferably 1 ms or more and less than 100 ms, more preferably 10 ms or more and 50 ms or less. If the light emission time is too short, the subject's S may underestimate the field of view, and if the light emission time is too long, the first visual stimulus VS1 may be the upper threshold stimulus in the peripheral visual field depending on the subject S. By setting the single light emission time within the above range, the light emission at the light spots 1 to 8 becomes the sub-threshold stimulus of the subject S. Hereinafter, such a face image 50 is referred to as a subliminal moving image.

  A part of the light spots 1 to 8 (light spots 1 to 4) is set in an external region (outside head region R2) of the head HD in the face image 50. The other part (light spots 5 to 7) of the light spots 1 to 8 is set on the region of the hair HR (hair region R1) in the face image 50. Still another part (light spot 8) of the light spots 1 to 8 is set on the area of the face FC in the face image 50. The light spot 8 is included in the central visual field of the general subject S.

  When the light emission of the light spots 1 to 8 is completed (FIG. 2: step S40), the operator notifies the subject S of the fact that one or more light spots are emitted. The subject S answers the position of the luminescence recognized under consciousness by pointing to the oral or facial image 50 (FIG. 2: step S60). In this method, the positions of all the light spots 1 to 8 are in different directions around the point of interest RP. In other words, individual vectors (not shown) having the point of interest RP as the base and light spots 1 to 8 as the tips do not coincide with each other. For this reason, the subject S may answer not only the position of the light spot where the light emission has been recognized, but also the direction of the light spot based on the point of interest RP. Thereby, the light spots 1-8 recognized by the subject S can be easily identified.

  In addition, the light spots 1 to 8 are all different in distance from the point of interest RP. In the visual field determination step S70, among the light spots recognized by the subject S (for example, the light spots 6 to 8), the one having the maximum distance from the point of interest RP (light spot 6) is selected as the peripheral visual field of the subject S. Determined as the outer edge. Then, the radius of the concentric circle CC on which the light spot (light spot 6) is placed is determined as the range of the narrow field of view of the subject S when the face image 50 is viewed.

(Second visual stimulus)
FIG. 5 is a diagram illustrating the face image 50 and the second visual stimulus VS2. The second visual stimulus VS2 shown in FIG. 5 is a moving image given by changing the pixel value of the face image 50. Specifically, a plurality of annular areas CA having different diameters surrounding the point of interest RP are provided in the face image 50. The pixel values of the pixels included in the annular area CA are changed with time under a perceptual threshold in the peripheral visual field. Then, the visual field is determined based on the maximum radius of the one or more annular regions CA recognized by the subject S that the pixel value has changed.

  Although the concentric circle CC centering on the attention point RP is also shown in FIG. 5 for convenience, the concentric circle CC is not presented to the subject S. The annular area CA is an area separated by concentric circles CC and includes a large number of pixels. The numbers (1) to (8) in the figure are centered on the point of interest RP, the annular region CA (1) located at the distal end to the annular region CA (7) located proximally, and the circular region at the center ( Reference numerals are given in order toward 8).

  The operator instructs the subject S to pay attention to the eye EY or the point of interest RP of the face image 50. While the subject S is looking at the eye EY or the point of interest RP, the pixel values of the pixels included in the annular area CA (1) are changed. In this method, the brightness before the change is increased by a predetermined ratio (change ratio). If the change ratio is too small, the subject S cannot know the change in the pixel value, and the field of view is evaluated too low. In addition, if the change ratio is excessive, the second visual stimulus VS2 may be an upper threshold stimulus in the peripheral visual field depending on the subject S. In this method, it is preferable to increase the brightness or luminance of the pixels included in the annular area CA at an increase rate of 10% or more and less than 50%. Hereinafter, the face image 50 is referred to as a fade-out moving image.

  The annular areas CA (1) to CA (3) are set in the external area (outside the head area R2) of the head HD in the face image 50. The annular areas CA (4) and CA (5) are set on the area of the hair HR (hair area R1) in the face image 50. The annular areas CA (6) and CA (7) include part or all of the face FC.

  After changing the brightness of the pixels included in the annular area CA (1) at a predetermined ratio (change ratio), the brightness of the pixels included in the annular area CA (2) adjacent to the inside of the annular area CA (1). Let the change begin. The brightness of the pixels included in each annular area is changed to the annular area CA (7) in the following order. Furthermore, when the brightness of the circular area (8) including the point of interest RP inside the annular area CA (7) is changed, the generation of the second visual stimulus VS2 is finished (FIG. 2: Step S40).

  Hereinafter, the operator notifies the subject S of the fact that the pixel value of the face image 50 is changed. The subject S answers the position at which the change in the pixel value is consciously recognized by pointing to the oral or facial image 50 (FIG. 2: step S60).

  Here, the number of the annular areas CA is not particularly limited, but by setting the number to 5 or more, the visual field of the subject S can be measured with a practical resolution. The number of annular areas CA is preferably 20 or less. If the number of annular areas CA is excessive, the width dimension of each annular area CA becomes minute, so that pixels whose pixel values change from the outer edge of the face image 50 toward the point of interest RP are substantially continuous. Will progress. This makes it difficult for the subject S to notice the change in the pixel value, and the field of view is underestimated. On the other hand, if the number of annular areas CA is within the above range, each annular area CA has a predetermined width dimension, so that the area where the pixel value changes progresses stepwise toward the point of interest RP. Will be. As a result, the subject S can appropriately recognize the change in the pixel value of the pixel inside the peripheral visual field with consciousness.

  In the visual field determination step S70, the annular area CA in which the subject S first recognizes the change of the pixel value is determined as the outer edge of the peripheral visual field of the subject S. Specifically, the radius of the center of the width of the annular area CA or the radius of the inner periphery of the annular area CA is determined as the range of the narrow field of view of the subject S when viewing the face image 50.

(Other processes)
In the visual field presentation step S80, the visual field determined in the visual field determination step S70 is presented to the subject S. Here, presenting the visual field includes not only presenting the range of the visual field but also presenting the positional relationship between the face image 50 and the subject S and the viewing angle calculated based on the visual field.

  In this method, a visual field is measured from one subject S and another subject who is a beauty specialist using the visual field measurement method described above. Expert line-of-sight data PD representing the visual field of a beauty expert is stored in the expert data storage unit 32 (see FIG. 1). The subject S is assumed to be a beauty amateur (beauty non-expert). Then, the visual field of one subject S (amateur visual field) and the visual field (expert field of view) of another subject (beauty specialist) are presented to one subject S in comparison. Accordingly, it is possible to quantitatively indicate to the subject S that the visual field range is different between a beauty expert and a beauty non-expert when viewing a human face. Specifically, it is preferable to compare the visual field of the subject S and the visual field of the beauty specialist by superimposing and displaying two circles centered on the point of interest RP on the face image 50.

(Gaze tracking)
In this method, the line-of-sight data VD of the subject S who views the visual stimulus VS is acquired using the line-of-sight acquisition unit 20 and stored in the line-of-sight data storage unit 30 (see FIG. 1). Specifically, the line-of-sight acquisition unit 20 emits a light beam such as infrared rays toward the head of the subject S and receives the reflected light. The line-of-sight acquisition unit 20 images the received light and calculates the pupil position of the subject S based on the pixel contrast, thereby detecting the position of the eye of the subject S and calculating the viewpoint position (line-of-sight direction) as numerical data or Acquired as image data.

  After the face image 50 is presented to the subject S (FIG. 2: step S10), acquisition of the line-of-sight data VD of the subject S is started (FIG. 2: step S20). However, the timing of starting the acquisition of the line-of-sight data VD is not limited to that in FIG. 2, and may be after the generation of the visual stimulus VS is started (FIG. 2: step S30). Then, after the generation of the visual stimulus VS is stopped (FIG. 2: step S40), the acquisition of the line-of-sight data VD is ended (FIG. 2: step S50).

  By acquiring the line-of-sight data VD of the subject S, it is possible to grasp whether or not the subject S is paying attention to the point of interest RP during the generation of the visual stimulus VS. Further, by using the line-of-sight data VD, the visual field of the subject S can be measured with high accuracy. This will be described below.

  FIG. 6 is a flowchart showing a modification of the visual field determination step S70. When the line-of-sight acquisition unit 20 acquires the line-of-sight data VD, the CPU of the visual field measurement device 100 determines whether or not the viewpoint position of the subject S is included within a predetermined normal range from the point of interest RP (FIG. 6: Step). S71). When it is determined that the viewpoint position of the subject S is not within the normal range (FIG. 6: step S71 = NO), a notification output indicating the determination result may be performed and the visual field measurement may be stopped.

  If it is determined that the viewpoint position is normal (FIG. 6: step S71 = YES), the stimulation history data indicating the generation history of the visual stimulation VS and the line-of-sight data VD are synchronized based on the time information (FIG. 6). 6: Step S72). The stimulus history data and the line-of-sight data VD include both time information that can be associated with each other. Thereby, the deviation distance between the generation position of the visual stimulus VS and the eye viewpoint of the subject S at that moment is calculated (FIG. 6: Step S73).

  When the operator receives an answer regarding the visual stimulus VS (light spots 1 to 8 in FIG. 4 or the annular area CA in FIG. 5) recognized by the subject S under consciousness (FIG. 2: step S60), Based on the maximum value, the maximum distance from the eye viewpoint is calculated (FIG. 6: Step S74). The maximum distance is set as the maximum range of the peripheral vision of the subject S.

  That is, in this method, based on the line-of-sight data VD and position information indicating the generation position of the visual stimulus VS, the maximum distance between the generation position of the one or more visual stimulations VS recognized by the subject S and the eye viewpoint of the subject S Is calculated. And the visual field of the test subject S is determined from this maximum distance.

  Thereby, even when the viewpoint of the subject S is unconsciously separated from the point of interest RP, the narrow field of view of the subject S can be suitably measured.

  Hereinafter, the result of a reference experiment showing that it is effective to measure the visual field range when a human face is visually observed as in the visual field measurement method of the present embodiment will be described.

<Reference experiment 1>
In Reference Experiment 1, it was verified that the wide and narrow difference between the field of view of a beauty professional and the field of view of a non-beauty expert varies depending on the object to be visually observed, and particularly when this is a human face.
FIG. 7 is a diagram illustrating a landscape image 51 that does not include a human face. The landscape image 51 is a color image. FIG. 8 is a diagram showing the plain image 52. The landscape image 51 and the plain image 52 are internally set with a point of interest RP and a concentric circle CC centered on the point of interest RP. The point of interest RP is set near the center of each of the landscape image 51 and the plain image 52. 7 and 8 show the point of interest RP and the concentric circle CC for convenience, but these are not presented to the subject S. On each concentric circle CC in FIGS. 7 and 8, light spots 1 to 8 are arranged in the same manner as the face image 50 (subliminal moving image) in FIG. That is, FIG. 7 is a subliminal video of the landscape image 51, and FIG. 8 is a subliminal video of the plain image 52.

  Further, in addition to the landscape image 51 of FIG. 7, an annular area CA centering on the point of interest RP is defined in the same manner as the face image 50 (fade-out movie) of FIG. A fade-out moving image (not shown) for changing the pixel values of the pixels included in each annular area CA was prepared toward the annular area CA located at the center. Similarly, a fade-out video was prepared for the plain image 52 in FIG.

  For the subliminal video and the fade-out video of the face image 50, the landscape image 51, and the plain image 52, the field of view of 20 subjects S was measured. Of the 20 subjects S, 5 were beauty specialists and 15 were non-beauty specialists.

  In the subliminal moving image shown in FIG. 4, the light spot 1 is caused to emit light after a predetermined time has elapsed since the face image 50 was presented to the subject. Then, the time (light emission interval) until light spot 1, light spot 2,... And the next light spot emit light is 2 seconds, and light is emitted sequentially and intermittently once to light spot 8. It was. Therefore, the time from light emission at the light spot 1 to light emission at the light spot 8 is 14 seconds. Each lighting time of the light spot 1 to the light spot 8 was 30 milliseconds. That is, this subliminal moving image is a moving image that repeats light emission for 0.03 seconds and non-light emission for 1.97 seconds. Hereinafter, the time from the light emission start of the light spot 1 to the light emission end of the light spot 8 is referred to as a stimulus presentation time in the subliminal moving image.

  Regarding the fade-out moving image shown in FIG. 5, from the start of presentation of the face image 50, the circular area CA (1), the circular area CA (7), and the circular area (8) are sequentially and continuously included in each area. The brightness of the pixel was increased by 20%. The brightness of the pixels in each region was increased in 3 to 4 seconds, and the overall brightness of the face image 50 was increased in about 30 seconds in total. Hereinafter, the time from the start of change of the pixel value of the circular area CA (1) to the end of change of the pixel value of the circular area (8) is referred to as a stimulus presentation time in the fade-out moving image.

  Regarding the landscape image 51 in FIG. 7 and the plain image 52 in FIG. 8, the visual stimulation in the subliminal moving image and the fade-out moving image is the same as that in the face image 50.

  FIG. 9 is a scatter diagram in which the viewing angle measurement results of each subject S are plotted for the subliminal video and the fade-out video related to the face image 50. Measurement results of beauty professionals are circled. The viewing angle is an apex angle θ (see FIG. 10) representing the maximum range FOV in which the subject S has recognized the visual stimulus. The vertical dimension of the face image 50 shown in FIGS. 4 and 5 is 28.5 cm, the horizontal dimension is 38 cm, and the distance from the subject S to the facial image 50 is 45 cm. The same applies to the landscape image 51 and the plain image 52.

  From the result of FIG. 9, in the case of the face image 50 including a human face, it is understood that the visual field of the beauty specialist is remarkably wider than that of the non-beauty specialist regardless of the difference between the subliminal video and the fade-out video. It was.

  FIG. 11 is a scatter diagram in which the viewing angle measurement results of each subject S are plotted for the subliminal video and the fade-out video regarding the landscape image 51. FIG. 12 is a scatter diagram in which the viewing angle measurement results of each subject S are plotted for the subliminal moving image and the fade-out moving image related to the plain image 52.

  From the results of FIG. 11 and FIG. 12, no significant difference was found in the breadth of the visual field between the beauty specialist and the non-beauty specialist. That is, it can be seen that, among the subjects S, the field of view of the beauty specialist was not originally medically wider than that of the non-beauty specialist. Therefore, the tendency of a beauty professional to view a person's face with a wider field of view than a non-beauty expert is considered to be an ability acquired by a beauty professional through training or experience.

<Reference experiment 2>
In Reference Experiment 2, it was verified that beauty professionals have higher sensitivity with respect to changes in the appearance of skin color than non-beauty professionals. FIG. 13 is a diagram showing a face sample image 53 used in Reference Experiment 2.

  According to the study of the present inventor, it is clear that the appearance of the skin color changes when the hair color is changed due to visual effects such as color assimilation and contrast. More specifically, it has been experimentally shown that when the lightness of the hair color is changed, a contrast effect is produced and the apparent lightness of the skin color changes in the opposite direction. When the saturation or hue of the hair color is changed, it has been experimentally revealed that an assimilation effect occurs and the apparent saturation or hue of the skin color changes in the same direction. Therefore, when changing hair color using hair coloring etc., consider the illusion effect that the appearance of the skin color changes due to the change of the hair color, and determine the suitability of the hair color and the skin color Is preferred.

(Illusion test)
Therefore, when the saturation, brightness, and hue of the hair HR shown in FIG. 13 are changed, the subject S accurately determines that an assimilation effect or a contrast effect (hereinafter also referred to as an illusion effect) occurs in the skin color of the face FC. The accuracy of human face beauty judgment was measured based on whether or not it was possible. Specifically, first, the face sample image 53 of FIG. 13 and another image (high saturation image: not shown) in which the saturation of the hair HR of the face sample image 53 is increased by 7% (= 7/100). Were observed by the subject S. And, which skin color is vivid, that is, the saturation looks high, the answer was made by forced alternative method. The answer is that the skin color of the high saturation image looks vivid.
Similarly, a high brightness image (not shown) in which the brightness of the hair HR shown in FIG. 13 is increased by 7% (= 7/100), and conversely, the brightness is decreased by 7% (= 7/100). A low brightness image (not shown) was prepared, and each image and the face sample image 53 of FIG. Then, he asked which skin color lightness seemed higher by a forced alternative method. Due to the brightness contrast effect between the hair color and the skin color, the skin color of the high brightness image looks darker (lower brightness) than the skin color of the original face sample image 53, and conversely, the skin color of the low brightness image is the original face sample image 53. The correct answer is that it looks brighter (higher brightness) than the skin color of.
Further, with respect to the hue of the hair HR shown in FIG. 13, a high hue image (not shown) that is yellowed by about 2% (= 7/360) (hereinafter, the hue is increased) is about 2 % (= 7/360) of a red hue image (not shown) that is reddish (hereinafter, the hue is lowered) is prepared, and each image and the face sample image 53 of FIG. Comparison was made individually. Then, he asked which skin tone the color looks higher by a forced alternative method. Answer that the skin color hue of the high hue image looks high (yellowish) and the skin color hue of the low hue image looks low (reddish) due to the assimilation effect of the hair color and the skin color It is correct.

Regarding the subtle changes in the appearance of skin color due to these optical illusion effects, the accuracy rate of the following groups was calculated.
(1) Facial wide-field group: A group of five subjects S having a large viewing angle with respect to the face image 50 plotted on the upper right in FIG.
(2) Narrow face field group: A group of five subjects S having a small viewing angle with respect to the face image 50 plotted on the lower left in FIG.
(3) Landscape wide-field group: A group of five subjects S having a large viewing angle with respect to the landscape image 51 plotted on the upper right in FIG.
(4) Narrow landscape group: A group of five subjects S having a small viewing angle with respect to the landscape image 51 plotted at the lower left in FIG.
(5) Solid wide-field group: A group of five subjects S having a large viewing angle with respect to the plain image 52 plotted on the upper right in FIG.
(6) Solid narrow-field group: A group of five subjects S having a small viewing angle with respect to the plain image 52 plotted on the lower left in FIG.

  FIG. 14 is a graph showing the illusion accuracy rate (%) of each group of the above (1) to (6). The illusion accuracy rate of each group is as follows: (1) Facial wide field group = 92%, (2) Facial narrow field group = 48%, (3) Landscape wide field group = 72%, (4) Landscape narrow field group = 64 %, (5) plain wide-field group = 64%, (6) plain narrow-field group = 68%. In addition, since it is a compulsory alternative method, the illusion accuracy rate of answers based on guesswork is 50%.

  From this result, the wide-angle visual field group with a large viewing angle with respect to the face image 50 has a remarkably high illusion accuracy rate, which is about twice the illusion accuracy rate of the narrow-face field group. On the other hand, there was no significant difference in the correct illusion rate between the landscape wide-field group and the landscape narrow-field group. Similarly, there was no significant difference in the illusion accuracy rate between the plain wide field group and the plain narrow field group. From this result, it was found that it is effective in observing a person's head with a wide field of view like a beauty specialist in beauty judgment.

<Reference Experiment 3>
We verified that the accuracy of the illusion is improved by widening the field of view. FIG. 15 shows the distance between the eye of the subject S and the face image 50 for a group (1a) having a large viewing angle and a group (2a) having a small viewing angle with respect to the subliminal video of the face image 50. It is a graph which shows the change of the illusion accuracy rate when changing to 45 cm, 90 cm, and 135 cm.
FIG. 16 shows the distance between the eyes of the subject S and the face image 50 for a group of five people (1b) with a large viewing angle and a group of two people with a small viewing angle (2b) for the fade-out video of the face image 50. It is a graph which shows the change of the illusion accuracy rate when changing to 45 cm, 90 cm, and 135 cm.

  From the results of FIGS. 15 and 16, even the subject S with a small viewing angle can observe the entire head HD by observing the face image 50 from a distant distance, and as a result, the occurrence of the illusion effect can be accurately determined. I found out that I could judge.

  As a result, it has been clarified that it is effective to make a general observation of the entire head HD rather than locally observing the hair HR and the face FC in making a beauty decision. Therefore, it is effective to measure the visual field of the subject S with respect to the face image 50 by the visual field measurement method performed using the visual field measurement device 100 of the present embodiment. Then, when the visual field of the subject S is not sufficient, it is effective to present the measurement result in comparison with the expert line-of-sight data PD.

<Reference experiment 4>
It was clarified that there is a suitable numerical range for the application time of the visual stimulus VS. In the same manner as in Reference Experiment 1, the subliminal moving image (FIG. 4) of the face image 50 was visually observed and the viewing angle was measured. In this experiment, the face image 50 and the first visual stimulus VS1 were applied to 15 subjects over a stimulus presentation time of 5 seconds (the time from the start of light emission at light spot 1 to the end of light emission at light spot 8). presentation. The individual light emission times from the light spot 1 to the light spot 8 were set to 30 milliseconds as in the reference experiment 1, and the light emission intervals of the respective light spots were made uniform. After the light emission at the light spot 8, the light spot at which the subject recognized the light emission was answered verbally, and the viewing angle was calculated in the same manner as in Reference Experiment 1.

From the result, a group of five people (1a) with a large viewing angle with respect to the subliminal moving image of the face image 50 and a group of two people with the same small viewing angle (2a) were extracted.
For this group (1a) and (2a), an illusion test using the face sample image 53 of FIG. 13 and its high saturation image was performed in the same manner as in Reference Experiment 2 and Reference Experiment 3, and the correct illusion rate was obtained. .

  The same viewing angle measurement and optical illusion test as described above were performed for 15 different subjects with the stimulus presentation time changed from 5 seconds to 10 seconds. Furthermore, 12 sets of 15 different subjects were prepared, and the stimulus presentation time was 15 seconds, 20 seconds, 25 seconds, 30 seconds, 35 seconds, 40 seconds, 45 seconds, 50 seconds, 60 seconds, 75 seconds, 90 seconds. The viewing angle measurement and the optical illusion test were performed in the same manner at 120 seconds. FIG. 17 shows the relationship between the stimulus presentation time and the illusion accuracy rate.

  Next, for 15 different subjects, the fade-out moving image (FIG. 5) of the face image 50 was visually observed, and the viewing angle was measured. The stimulus presentation time at this time (the time from the start of change of the pixel value of the annular area CA (1) to the end of change of the pixel value of the circular area (8)) was also 5 seconds. A group of five people (1b) with a large viewing angle and a group of two people (2b) with a small viewing angle were extracted. An illusion test was performed on the groups (1b) and (2b) in the same manner as described above to obtain the illusion accuracy rate.

  Furthermore, 13 sets of 15 different subjects were prepared, and the stimulus presentation time was 10 seconds, 15 seconds, 20 seconds, 25 seconds, 30 seconds, 35 seconds, 40 seconds, 45 seconds, 50 seconds, 60 seconds, 75 seconds. , 90 seconds, and 120 seconds were changed, and the viewing angle measurement and the illusion test were similarly performed on the fade-out moving image. FIG. 18 shows the relationship between the stimulus presentation time and the illusion accuracy rate.

  From the results of FIGS. 17 and 18, it was found that the groups (1a) and (1b) having a large viewing angle have a particularly high illusion accuracy when the stimulus presentation time is 15 seconds or longer and 45 seconds or shorter. This is presumably because the face image 50 could be visually observed with sufficient attention because the stimulus presentation time was 15 seconds or longer. In addition, because the stimulus presentation time is 45 seconds or less, the generation of the stimulus ends before the brain of the subject S starts spontaneous activity, and thus the entire observation of the face image 50 centered on the point of interest RP has a concentration. It is inferred that it was carried out in a sustained manner.

  On the other hand, when the stimulus presentation time was 50 seconds or more, the illusion correctness rate of the groups (2a) and (2b) with a small viewing angle became substantially equal to the groups (1a) and (1b). This is presumed to be because, after receiving a stimulus presentation of more than 45 seconds, the subjects started to move spontaneously for the first time in their brains, and as a result, the attention to the point of interest RP was lost. Is done. That is, when the stimulus presentation time becomes longer, the subject's viewpoint starts to move unconsciously, so that the measurement accuracy of the viewing angle is reduced, the entire face image 50 can be observed and the illusion accuracy rate is high, and the viewing angle is large. This is probably because the group no longer matches.

  Therefore, in measuring the visual field of the subject S, it is preferable that the stimulation time for generating the visual stimulation VS intermittently or continuously is 15 seconds to 45 seconds. When the time is 15 seconds or longer, the viewpoint of the subject S is determined as the point of interest RP, and a suitable overall observation of the face image 50 becomes possible. Moreover, since it is 45 seconds or less, it is possible to continue gazing at the point of interest RP of the face image 50, and it is possible to accurately recognize the visual stimulus VS generated inside the peripheral visual field.

The above embodiment includes the following technical idea.
(1) Presenting a face image including a human face to a subject, generating visual stimuli at a plurality of positions at different distances from a point of interest determined on the face area in the face image, Measuring the visual field of the subject based on the generation position of the visual stimulus recognized by the visual field;
(2) The visual field measurement method according to claim 1, wherein the visual stimulation is generated on a region of hair in the face image and on an external region of the head including the face and the hair.
(3) The visual field measurement method according to (1) or (2), wherein the point of interest is between the eyes or in the vicinity of the eyes in the face image;
(4) The visual field measurement method according to any one of (1) to (3), wherein the visual stimulus below a perceptual threshold in the peripheral visual field of the subject is generated in the peripheral visual field and out of the visual field of the subject;
(5) The visual stimulus is light emission of a light spot that is lit at a perceptual threshold light intensity and a light emission time in the peripheral visual field, and a plurality of the light spots that are different from each other in at least one of a distance or a direction from the point of interest The visual field measurement method according to (4), wherein the visual field is sequentially turned on, and the visual field is determined based on a maximum distance between one or a plurality of the light spots at which the subject has recognized the light emission and the point of interest.
(6) The visual stimulus is given by changing a pixel value of the face image, and a plurality of annular regions having different diameters surrounding the point of interest are provided in the face image, and pixels included in the annular region are provided. The pixel value is changed with time under a perceptual threshold in the peripheral visual field, and the visual field is determined based on a maximum radius of one or a plurality of the annular regions recognized by the subject that the pixel value has changed ( 4) Field of view measurement method according to
(7) The visual stimulus generation position recognized by the subject after the visual stimulus is generated intermittently or continuously over a predetermined stimulus time without notifying the subject of the occurrence of the visual stimulus. The visual field measurement method according to any one of (1) to (6), wherein an answer is obtained from the subject;
(8) The visual field according to any one of (1) to (7), wherein line-of-sight data regarding the line-of-sight direction of the subject is acquired over a predetermined stimulation time in which the visual stimulation is generated intermittently or continuously. Measuring method;
(9) Based on the line-of-sight data and position information indicating the generation position of the visual stimulus, the maximum distance between the one or more generation positions of the visual stimulus recognized by the subject and the eye viewpoint of the subject is calculated. The visual field measurement method according to (8), wherein the visual field is determined from the maximum distance;
(10) The visual field measurement method according to any one of (1) to (9), wherein a stimulation time in which the visual stimulation is generated intermittently or continuously is 15 seconds to 45 seconds;
(11) From one subject and another subject who is a beauty specialist, the visual field is measured using the visual field measurement method according to any one of (1) to (10) above, and A visual field measurement method, wherein the visual field and the visual field of the other subject are compared and presented to the one subject;
(12) Visual stimulation is performed at a plurality of positions at different distances between a face image presenting means for presenting a face image including a human face to a subject and a point of interest determined on the face area in the face image. A visual field measuring device comprising: stimulus generating means for generating the visual field.

Although the visual field measurement method of the present invention may be described using a plurality of steps described in order, the described order does not limit the order or timing of executing the plurality of steps. Therefore, when carrying out the visual field measurement method of the present invention, the order of the plurality of steps can be changed within a range that does not hinder the content, and some or all of the execution timings of the plurality of steps overlap each other. You may do it.
Further, the various components of the visual field measuring device of the present invention only have to be formed so as to realize the function. For example, dedicated hardware that exhibits a predetermined function, a predetermined function is given by a computer program The data processing apparatus or the image processing apparatus can be realized as a predetermined function realized in the data processing apparatus or the image processing apparatus by a computer program, any combination thereof, or the like. The various components of the field-of-view measuring apparatus of the present invention do not have to be independent of each other, but a plurality of components are formed as a single member, and a single component is formed of a plurality of members. That a certain component is a part of another component, a part of a certain component overlaps a part of another component, and the like.

1-8: Light spot, 10: Face image presentation unit, 20: Gaze acquisition unit, 30: Gaze data storage unit, 32: Expert data storage unit, 50: Face image, 51: Landscape image, 52: Plain image, 53: Face sample image, 60: Stimulus generator, 100: Visual field measuring device, θ: Apex angle, CA: Annular region, CC: Concentric circle, EY: Eye, FC: Face, FOV: Maximum range, HD: Head, HR: head hair, OC: circumscribed circle, PD: expert gaze data, R1: head hair area, R2: head outside area, RP: point of interest, S: subject, VD: gaze data, VS: visual stimulus, VS1: First visual stimulus, VS2: second visual stimulus

Claims (7)

Present a face image including a human face to the subject,
Generating visual stimuli at a plurality of positions at different distances from the point of interest determined on the face area in the face image;
Measuring the visual field of the subject based on the occurrence position of the visual stimulus recognized by the subject ,
Including
A visual field measuring method , wherein the visual stimulus below a perceptual threshold in the peripheral visual field of the subject is generated in the peripheral visual field and out of the visual field of the subject . The visual stimulus is light emission of a light spot that is lit at a subthreshold light intensity and light emission time in the peripheral visual field;
A plurality of the light spots that are different from each other in at least one of the distance or direction from the point of interest are sequentially turned on,
The visual field measurement method according to claim 1 , wherein the visual field is determined based on a maximum distance between one or a plurality of the light spots at which the subject has recognized the light emission and the point of interest. The visual stimulus is provided by changing a pixel value of the face image;
A plurality of annular regions having different diameters surrounding the point of interest are provided in the face image, and the pixel values of the pixels included in the annular region are changed with time under a perceptual threshold in the peripheral visual field, and the pixel values are changed. The visual field measurement method according to claim 1 , wherein the visual field is determined based on a maximum radius of one or a plurality of the annular regions recognized by the subject. The visual field measurement method according to any one of claims 1 to 3 , wherein the visual line data regarding the visual line direction of the subject is acquired over a predetermined stimulation time in which the visual stimulation is generated intermittently or continuously. Based on the line-of-sight data and position information indicating the generation position of the visual stimulus, the maximum distance between the one or more generation positions of the visual stimulus recognized by the subject and the eye viewpoint of the subject is calculated, and the maximum The visual field measurement method according to claim 4 , wherein the visual field is determined from a distance. Measuring one visual field using the visual field measuring method according to any one of claims 1 to 5 from one test subject and another test subject who is a beauty specialist,
A visual field measuring method, wherein the visual field of the one subject is compared with the visual field of the other subject and presented to the one subject. Face image presenting means for presenting a face image including a human face to a subject;
Stimulus generating means for generating visual stimuli at a plurality of positions at different distances from the point of interest determined on the face area in the face image;
Equipped with a,
The stimulus generating means, said visual stimulus perception subliminal in peripheral vision of the subject, Ru is generated in the peripheral vision and the view out of the subject,
Visual field measuring device.