Queue detection device, white cane comprising same, queue detection method, queue detection program, passing space detection device, passing space detection method, and passing space detection program

Abstract

This queue detection device (10) comprises an image acquisition unit (11), a human body detection unit (12), a number-of-people detection unit (13), a face detection unit (14), a face direction detection unit (15), and a queue detection unit (16). The image acquisition unit (11) acquires an image that is captured in a predetermined direction. A human body detection unit (12) detects the presence or absence of a human body from the image that is acquired by the image acquisition unit (11). The number-of-people detection unit (13) detects the number of human bodies detected by the human body detection unit (12). The face detection unit (14) detects the faces of the human bodies detected by the human body detection unit (12). The face direction detection unit (15) detects the directions of the faces detected by the face detection unit (14). The queue detection unit (16) detects the presence or absence of a queue and the direction thereof on the basis of the number of human bodies detected by the number-of-people detection unit (13) and the direction of the faces of the human bodies detected by the face direction detection unit (15).

Description

Queue detection device and white cane equipped therewith, queue detection method, queue detection program, passage space detection device, passage space detection method and passage space detection program

[0001] The present invention relates, for example, to a queue detection device attached to a white cane used by visually impaired people, the elderly, etc., a white cane equipped with the same, a queue detection method, a queue detection program, a passage space detection device, a passage space detection method, and a passage space detection program.

[0002] In recent years, white canes used by the visually impaired and the elderly have been equipped with devices to help them detect and avoid obstacles. For example, Patent Document 1 discloses a walking assistance white cane that includes a camera and distance sensor to acquire information about the area around the cane body, and a control unit that determines the position of obstacles based on the surrounding information from the camera and distance sensor, determines the direction in which the user should walk to avoid the obstacles based on this position, steers the wheels in the direction in which the user should walk, and controls a guidance indicator unit to function as an indicator in the direction in which the user should walk.

[0003] Japanese Patent Publication No. 2017-42251

[0004] However, the above-described conventional walking support white cane has the following problems. That is, although the walking support white cane disclosed in the above publication can detect people and objects using a camera and a distance sensor, it does not have a function to actively detect an aggregate of people (a queue) and the direction in which the queue is arranged. An object of the present invention is to provide a queue detection device capable of detecting a queue that is an aggregate of people and the direction of the queue, a white cane including the same, a queue detection method, a queue detection program, a passage space detection device, a passage space detection method, and a passage space detection program. (Means for Solving the Problem) A queue detection device according to a first invention includes an image acquisition unit, a human body detection unit, a number of people detection unit, a face detection unit, a face direction detection unit, and a queue detection unit. The image acquisition unit acquires an image obtained by capturing an image of a predetermined direction. The human body detection unit detects the presence or absence of a human body from the image acquired by the image acquisition unit. The number of people detection unit detects the number of human bodies detected by the human body detection unit. The face detection unit detects the face of the human body detected by the human body detection unit from the image acquired by the image acquisition unit. The face direction detection unit detects the orientation of the face detected by the face detection unit from the image acquired by the image acquisition unit. The queue detection unit detects the presence or absence of a queue and the direction of the queue based on the number of human bodies detected by the number of people detection unit and the orientation of the faces of the human bodies detected by the face direction detection unit.

[0005] Here, for example, in a queue detection device mounted on a white cane or the like used by visually impaired persons, elderly people, etc., an image captured using a camera or the like is acquired, a human body and the face thereof included in the image are detected, and the presence or absence of a queue and the direction of the queue are detected based on the detection results of the number of detected human bodies and the orientation of the face of each human body. For example, when a visually impaired person or the like lines up in a queue or moves along the queue, simply detecting the presence or absence of people does not make it clear where the end of the queue is or which direction the queue is heading.

[0006] Therefore, this matrix detection device detects the presence or absence of a matrix and the direction in which the matrix is ​​moving based on the number of human bodies included in the acquired camera image and the orientation of their faces. This allows for the detection of a matrix, which is a collection of people, and its direction, enabling the user to move to the end of the matrix or follow the matrix. The matrix detection device according to the second invention is the matrix detection device according to the first invention, further comprising a occipital head detection unit that detects the occipital heads of multiple human bodies detected by the matrix detection unit.

[0007] This allows the system to determine the orientation of the detected matrix by detecting the position of the back of the heads of multiple human bodies (the hair side where there are no organs on the head). The matrix detection device according to the third invention is the matrix detection device according to the second invention, further comprising a last-end detection unit that detects the last end of the matrix from the backs of multiple human bodies detected by the back-of-head detection unit.

[0008] This allows for the detection of the orientation of a queue from the position of the back of the heads of multiple people (the side with hair where there are no organs on the head), and then the detection of the person at the end of the queue, thereby guiding visually impaired persons to join the end of the queue. The queue detection device according to the fourth invention is the queue detection device according to the third invention, further comprising a rear-end advance detection unit that detects when the rear of the queue detected by the rear-end detection unit moves forward in the direction of travel.

[0009] As a result, even if the queue moves forward after a person has joined the end of the detected queue, the change in distance from the person at the end of the queue can be detected, prompting the person to keep up with the advance of the queue. The queue detection device according to the fifth invention is the queue detection device according to the fourth invention, further comprising a last-place advance notification unit that notifies the last-place person that the last-place person has advanced in the last-place advance detection unit.

[0010] This makes it possible to notify visually impaired persons, for example, of the movement of the last person in a queue by sound or vibration. The queue detection device according to the sixth invention is the same as the queue detection device according to the fourth invention, and the detection of the last person moving forward in the last person advancement detection unit is performed using the detection result of an ultrasonic sensor.

[0011] This makes it possible to easily detect the advance of the last person in a line of people, for example, based on the change in distance to the last person detected by an ultrasonic sensor mounted on the white cane. The line of people according to the seventh invention is a line of people according to the fourth invention, wherein the detection of the advance of the last person in the line of people advance

[0012] This makes it possible to easily detect the advance of the last person in a line, for example, based on the change in distance to the last person in the line, as captured by a camera mounted on a white cane. The eighth line detection device is the fifth line detection device, wherein the last person advance notification unit provides notification by outputting a predetermined sound.

[0013] This makes it possible to easily inform visually impaired people, etc., that the last person in a queue has moved forward by outputting, for example, an audio message or a warning sound, when it is detected that the last person in the queue has moved forward. The queue detection device according to the ninth invention is the queue detection device according to the fifth invention, and the last person advancing notification unit provides notification by outputting a predetermined vibration.

[0014] This makes it possible to easily inform visually impaired people, etc., that the last in line has moved forward, for example, by outputting a vibration of a predetermined pattern corresponding to the forward movement of the last in line when it is detected that the last in line has moved forward. The matrix detection device according to the tenth invention is a matrix detection device according to the first or second invention, further comprising a human body coordinate setting unit that plots the coordinates of a plurality of human bodies constituting the matrix detected by the matrix detection unit.

[0015] This makes it possible to determine the positions of multiple human bodies that make up the detected matrix. The matrix detection device according to the eleventh invention is a matrix detection device according to the tenth invention, further comprising a passage space determination unit that determines whether or not there is space through which a person can pass based on the coordinates plotted in the human body coordinate setting unit. This makes it possible to determine whether or not there is space through which a person can pass between the detected matrices based on the coordinates plotted at the positions of the human bodies.

[0016] The queue detection device according to the twelfth invention is a queue detection device according to the eleventh invention, further comprising a direction of travel notification unit that notifies the direction of the space when the passage space determination unit determines that there is space. As a result, when it is determined that there is space in the detected queue that a person can pass through, the direction of that space can be notified by, for example, when a person is swinging their white cane from side to side, stopping the generation of sound or vibration when the white cane points in the direction of that space.

[0017] The queue detection device according to the 13th invention is the same as the queue detection device according to the 12th invention, and the direction of travel notification unit notifies by outputting a predetermined sound. As a result, when a space where a person can pass through is detected between the queue, for example, when a person is swinging their white cane from side to side and the direction of the white cane points in the direction of the space, the output of sounds or warning sounds is stopped, or an audio message is sent to guide the person in the direction of the space to pass through, thereby easily informing visually impaired persons and others of the direction in which they can pass through the queue.

[0018] The queue detection device according to the 14th invention is a queue detection device according to the 12th invention, wherein the direction of travel notification unit notifies by outputting a predetermined vibration. As a result, when a space where a person can pass through is detected between the queue, for example, when a person is swinging their white cane from side to side and the direction of the white cane points in the direction of the space, the output of sound or vibration is stopped, making it easy to inform visually impaired persons and others of the direction in which they can pass through the queue.

[0019] The white cane according to the 15th invention comprises a queue detection device according to the first or second invention, a gripping part to which the queue detection device is attached and which is grasped by the user, and a rod-shaped main body attached so as to extend from the gripping part. As a result, by using a white cane with the queue detection device mounted on the gripping part described above, it is possible to detect a queue, which is a group of people, and its direction, and to move to the end of the queue or follow the queue.

[0020] The white cane according to the 16th invention is the same as the white cane according to the 15th invention, and the predetermined direction is the direction of travel of the user of the white cane. Note that the direction of travel of the user of the white cane includes not only the direction in which the user intends to go, but also the surrounding area. This makes it possible to inform visually impaired people, for example, who walk while swinging their white cane from side to side, of the presence and direction of a queue ahead.

[0021] The matrix detection method according to the 17th invention comprises an image acquisition step, a human body detection step, a person count detection step, a face detection step, a face direction detection step, and a matrix detection step. In the image acquisition step, an image is acquired that is photographed in a predetermined direction. In the human body detection step, the presence or absence of a human body is detected from the image acquired in the image acquisition step. In the person count detection step, the number of human bodies detected in the human body detection step is detected. In the face detection step, the faces of the human bodies detected in the human body detection step are detected from the image acquired in the image acquisition step. In the face direction detection step, the orientation of the faces detected in the face detection step is detected from the image acquired in the image acquisition step. In the matrix detection step, the presence or absence of a matrix and its direction are detected based on the number of human bodies detected in the person count detection step and the orientation of the faces of the human bodies detected in the face direction detection step.

[0022] Here, for example, in a queue detection device mounted on a white cane used by the visually impaired or the elderly, an image is acquired using a camera, etc., human bodies and their faces are detected in the image, and the presence or absence of a queue and its direction are detected based on the number of detected human bodies and the orientation of each human body's face. For example, when a visually impaired person is standing in a queue or moving along with a queue, simply detecting the presence or absence of people does not reveal where the end of the queue is or which direction they are heading.

[0023] Therefore, in this matrix detection method, the presence or absence of a matrix and the direction in which the matrix is ​​moving are detected from the number of human bodies and the orientation of their faces included in the acquired camera image. This makes it possible to detect a matrix, which is a collection of people, and its direction, and to move to the end of the matrix or follow the matrix. The matrix detection program according to the 18th invention causes a computer to execute a matrix detection method comprising an image acquisition step, a human body detection step, a person count detection step, a face detection step, a face direction detection step, and a matrix detection step. In the image acquisition step, an image taken in a predetermined direction is acquired. In the human body detection step, the presence or absence of human bodies is detected from the image acquired in the image acquisition step. In the person count detection step, the number of human bodies detected in the human body detection step is detected. In the face detection step, the faces of the human bodies detected in the human body detection step are detected from the image acquired in the image acquisition step. In the face direction detection step, the orientation of the faces detected in the face detection step is detected from the image acquired in the image acquisition step. In the matrix detection step, the presence or absence of a matrix and its direction are detected based on the number of human bodies detected in the person count detection step and the orientation of the faces of the human bodies detected in the face direction detection step.

[0024] Here, for example, in a queue detection device mounted on a white cane used by the visually impaired or the elderly, an image is acquired using a camera, etc., human bodies and their faces are detected in the image, and the presence or absence of a queue and its direction are detected based on the number of detected human bodies and the orientation of each human body's face. For example, when a visually impaired person is standing in a queue or moving along with a queue, simply detecting the presence or absence of people does not reveal where the end of the queue is or which direction they are heading.

[0025] Therefore, this matrix detection program detects the presence or absence of a matrix and the direction in which the matrix is ​​moving based on the number of human bodies included in the acquired camera image and the orientation of their faces. This allows for the detection of a matrix, which is a collection of people, and its direction, enabling the user to move to the end of the matrix or follow the matrix. The passage space detection device according to the 19th invention comprises an image acquisition unit, a human body detection unit, a people detection unit, a matrix detection unit, and a passage space determination unit. The image acquisition unit acquires an image taken in a predetermined direction. The human body detection unit detects the presence or absence of human bodies from the image acquired by the image acquisition unit. The people detection unit detects the number of human bodies detected by the human body detection unit. The matrix detection unit detects the presence or absence of a matrix based on the number of human bodies detected by the people detection unit. The passage space determination unit determines the presence or absence of space through which a person can pass based on the positions of the multiple human bodies constituting the matrix detected by the matrix detection unit.

[0026] Here, for example, in a passage space detection device mounted on a white cane used by the visually impaired or the elderly, an image is acquired using a camera, etc., and the number of people contained in the image is detected. Based on the number of people detected, it is determined whether or not there is space to pass through within the crowd. For example, when a visually impaired person passes through a crowd, the position of people in the crowd can be recognized by warning sounds, etc., but simply detecting the position of people does not reveal where there is space to pass through.

[0027] Therefore, this pass-through space detection device detects whether or not there is a space that can be passed through based on the number of people and the human bodies included in the acquired camera image. By detecting a space that can be passed through a crowded area of ​​people, it is possible to guide people in that direction. The pass-through space detection method according to the 20th invention comprises an image acquisition step, a human body detection step, a people detection step, a queue detection step, and a pass-through space determination step. In the image acquisition step, an image taken in a predetermined direction is acquired. In the human body detection step, the presence or absence of human bodies is detected from the image acquired in the image acquisition step. In the people detection step, the number of human bodies detected in the human body detection step is detected. In the queue detection step, the presence or absence of a queue is detected based on the number of human bodies detected in the people detection step. In the pass-through space determination step, the presence or absence of a space through which a person can pass is determined based on the positions of the multiple human bodies constituting the queue detected in the queue detection step.

[0028] Here, for example, in a passage space detection device mounted on a white cane used by the visually impaired or the elderly, an image is acquired using a camera, etc., and the number of people contained in the image is detected. Based on the number of people detected, it is determined whether or not there is space to pass through within the crowd. For example, when a visually impaired person passes through a crowd, the position of people in the crowd can be recognized by warning sounds, etc., but simply detecting the position of people does not reveal where there is space to pass through.

[0029] Therefore, this pass-through space detection method detects whether or not there is a space that can be passed through based on the number of people and the human bodies included in the acquired camera image. By detecting a space that can be passed through a crowded area of ​​people, it is possible to guide people in that direction. The pass-through space detection program according to the 21st invention causes a computer to execute a pass-through space detection method comprising an image acquisition step, a human body detection step, a people detection step, a queue detection step, and a pass-through space determination step. In the image acquisition step, an image taken in a predetermined direction is acquired. In the human body detection step, the presence or absence of human bodies is detected from the image acquired in the image acquisition step. In the people detection step, the number of human bodies detected in the human body detection step is detected. In the queue detection step, the presence or absence of a queue is detected based on the number of human bodies detected in the people detection step. In the pass-through space determination step, the presence or absence of a space through which a person can pass is determined based on the positions of the multiple human bodies constituting the queue detected in the queue detection step.

[0030] Here, for example, in a passage space detection device mounted on a white cane used by the visually impaired or the elderly, an image is acquired using a camera, etc., and the number of people contained in the image is detected. Based on the number of people detected, it is determined whether or not there is space to pass through within the crowd. For example, when a visually impaired person passes through a crowd, the position of people in the crowd can be recognized by warning sounds, etc., but simply detecting the position of people does not reveal where there is space to pass through.

[0031] Therefore, this passage space detection program detects whether or not there is space to pass through based on the number of people and the human bodies included in the acquired camera image. By detecting space to pass through a crowded area of ​​people, it is possible to guide people in that direction. (Effects of the invention) According to the queue detection device of the present invention, it is possible to detect a queue, which is a group of people, and its direction.

[0032] Figure 1 is an overall diagram showing the configuration of a white cane equipped with a matrix detection device according to one embodiment of the present invention. Figure 1 is a conceptual diagram showing how the white cane is used by a user. Figure 1 is a control block diagram of the white cane and the matrix detection device mounted on the white cane. Figure 1 shows an example of a matrix included in a camera image captured by a camera mounted on the white cane. Figure 1 shows an example of plotting a human body in a camera image captured by a camera mounted on the white cane and detecting passing space within the matrix. (a) is a flowchart showing the basic processing flow of the white cane in Figure 1. (b) is a flowchart showing the object detection processing flow using ultrasound. (c) is a flowchart showing the human detection processing flow using camera images. Figure 3 is a flowchart showing the processing flow of the matrix detection method executed by the matrix detection device. Figure 3 is a flowchart showing the processing flow of the space detection method after matrix detection executed by the matrix detection device. Figure 3 is a flowchart showing the processing flow for following the matrix after matrix detection executed by the matrix detection device.

[0033] A matrix detection device according to one embodiment of the present invention will be described below with reference to Figures 1 to 9. Note that in this embodiment, unnecessarily detailed explanations may be omitted. For example, detailed explanations of already well-known matters and redundant explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to facilitate understanding by those skilled in the art.

[0034] Furthermore, the applicant provides the accompanying drawings and the following description so that those skilled in the art may fully understand the present invention, and not intends to limit the subject matter described in the claims by means of these. (1) Configuration of the white cane 30 The white cane 30 according to this embodiment is a cane used, for example, by visually impaired persons or elderly persons when walking, and as shown in Figure 1, comprises a gripping portion 30a, a main body portion 30b, a camera 31, an ultrasonic sensor 32, a push button 33, and a notification portion 34 (see Figure 3).

[0035] As shown in Figure 2, the white cane 30 is used by the user U1 by swinging it from side to side and striking the ground with the tip of the main body 30b. When the camera 31 and ultrasonic sensor 32, which will be described later, detect an object such as a person or object within a predetermined distance range (for example, within 1.0 to 2.0 m), they generate vibrations or sounds to notify the user U1. In the following description, the forward direction of the white cane 30 refers to, for example, the direction in which the user U1 is moving, and means the forward direction of the white cane 30 when the white cane 30 is swung from side to side by the user U1 (the direction in which the camera 31 and ultrasonic sensor 32 are facing).

[0036] Furthermore, the white cane 30 is held by the user U1 so that the side equipped with the camera 31 and ultrasonic sensor 32 shown in Figure 1 faces the direction of travel. The gripping portion 30a is to which the matrix detection device 10 is attached and is held by the user U1. A rod-shaped main body portion 30b is connected to one end of the gripping portion 30a in the longitudinal direction. The gripping portion 30a is also equipped with the camera 31, ultrasonic sensor 32, push button 33, and notification unit 34.

[0037] The main body 30b is a rod-shaped member attached to extend from the grip portion 30a, and at the end opposite to the connection side with the grip portion 30a, it contacts the ground or obstacles, generating a sound of striking the ground. The camera 31 is provided on the grip portion 30a of the white cane 30 and captures images of, for example, the front direction the white cane 30 faces when the white cane 30 is swung from side to side, or the direction in which the user U1 moves when using the white cane 30.

[0038] The camera 31, for example, takes continuous photographs at predetermined time intervals when the white cane 30 is powered on, and the image acquisition unit 11 acquires the camera image each time a predetermined time interval has elapsed. The ultrasonic sensor 32, for example, emits ultrasonic waves (sound waves with frequencies exceeding the audible range (0Hz to 20kHz)) in the direction in front of the white cane 30 when it is in use, and receives the reflected waves. This makes it possible to detect the presence or absence of an object in the direction in front of the white cane 30, the distance to the object, etc.

[0039] In this embodiment, when the ultrasonic sensor 32 detects an object within a predetermined distance range (for example, 1.0 to 2.0 m), the notification unit 34 generates sound, vibration, etc., to notify the user U1 that there is an object in the vicinity in front of the white cane 30. The push button 33 is pressed when, in the object detection using the ultrasonic sensor 32 described above, no object is detected within the predetermined distance range, and the user U1 checks for the presence or absence of an object in front of the white cane 30 (in the direction of travel of the user U1).

[0040] The notification unit 34 is, for example, a speaker that outputs sound or a vibrator that outputs vibration, and outputs sound or vibration when an object is detected as described above, or when a queue is detected, the last person is detected, or a passage space is detected. Preferably, the sound and vibration generated by the notification unit 34 have different patterns depending on the magnitude of the distance to the detected object, the type of object detected, the queue, or the last person detected. This allows users U1, such as visually impaired or elderly people, to easily recognize what the detected object is and the distance to that object.

[0041] (2) Configuration of the Queue Detection Device 10 The queue detection device 10 according to this embodiment is provided inside the gripping portion 30a of the white cane 30 described above, and as shown in Figure 3, comprises an image acquisition unit 11, a human body detection unit 12, a person detection unit 13, a face detection unit 14, a face direction detection unit 15, a queue detection unit 16, a back of the head detection unit 17, a last-place detection unit 18, a last-place forward detection unit 19, a notification control unit (last-place forward notification control unit, direction of travel notification control unit) 20, a human body coordinate plotting unit (human body coordinate setting unit) 21, and a passage space determination unit 22.

[0042] The image acquisition unit 11 acquires camera images (see, for example, Figure 4) from the camera 31, which captures images in a predetermined direction (for example, the direction in which the user U1 is moving, or the direction in front of the white cane 30). The human body detection unit 12 detects the presence or absence of a human body from the camera images acquired by the image acquisition unit 11 (see, for example, human bodies P1 to P6 included in the camera images in Figure 4).

[0043] The number-of-people detection unit 13 detects, from the camera image acquired by the image acquisition unit 11, the number of human bodies detected by the human body detection unit 12 (for example, six human bodies P1 to P6 in the camera image of FIG. 4). The face detection unit 14 detects the faces of the human bodies P1 to P6 detected by the human body detection unit 12 from the camera image acquired by the image acquisition unit 11.

[0044] The face direction detection unit 15 detects the orientation of the face detected by the face detection unit 14 from the camera image acquired by the image acquisition unit 11. Note that the face direction detection unit 15 detects the face direction, for example, by detecting the positions of each organ such as eyes, nose, and mouth included in the face detected by the face detection unit 14, and the positions of eyebrows and hair. The queue detection unit 16 detects the presence or absence of a queue Px and its direction based on the number of human bodies detected by the number-of-people detection unit 13 (six in FIG. 4) and the face orientation of the human bodies detected by the face direction detection unit 15 (all face left in FIG. 4).

[0045] For example, when the camera image shown in FIG. 4 is acquired by the image acquisition unit 11, the queue detection unit 16 detects the queue Px including six human bodies P1 to P6, and detects that the orientation of the queue Px is left-facing. Accordingly, by detecting and notifying visually impaired persons or elderly people using a white cane 30 of the presence or absence of the queue Px, which is an aggregate of people, and the direction thereof, it is possible to encourage such users to stand behind the queue Px or avoid passing through so as not to be mixed into the queue Px.

[0046] The occiput detection unit 17 detects the occiputs of the plurality of human bodies detected by the queue detection unit 16 from the camera image acquired by the image acquisition unit 11. Note that the occiput detection unit 17 detects the position of the occiput, for example, by detecting the positions of each organ such as eyes, nose, and mouth included in the face detected by the face detection unit 14, and the positions of eyebrows and hair.

[0047] The rearmost end detection unit 18 detects the rearmost end (human body P6 in FIG. 4) of the queue Px from the positions of the occipital regions of the plurality of human bodies P1 to P6 detected by the occipital region detection unit 17, based on the camera image acquired by the image acquisition unit 11. The rearmost advancement detection unit 19 detects that the rearmost end of the queue Px detected by the rearmost end detection unit 18 has advanced in the traveling direction.

[0048] Specifically, the rearmost advancement detection unit 19 uses the detection result of the ultrasonic sensor 32 to detect that the distance to the rearmost human body P6 of the queue Px has increased, thereby detecting the advancement of the rearmost human body P6 of the queue Px. Alternatively, the rearmost advancement detection unit 19 uses a camera image (image information) acquired from the camera 31 by the image acquisition unit 11 to detect that the distance to the rearmost human body P6 of the queue Px has increased, thereby detecting the advancement of the rearmost human body P6 of the queue Px.

[0049] A notification control unit (a rearmost advancement notification control unit, a traveling direction notification control unit) 20 notifies that the advancement of the rearmost human body P6 of the queue Px has been detected by the rearmost advancement detection unit 19. Here, when the rearmost advancement detection unit 19 detects the advancement of the rearmost end of the queue Px, the notification control unit 20 outputs, for example, voice messages such as "Please move forward", "The rearmost end of the queue has advanced", or a predetermined warning sound indicating the advancement of the rearmost end of the queue Px.

[0050] Furthermore, when the rearmost advancement detection unit 19 detects the advancement of the rearmost end of the queue, the notification control unit 20 outputs, for example, a predetermined vibration indicating the advancement of the rearmost end of the queue Px. Thereby, the user U1 of the white cane 30 can easily recognize that the rearmost end of the queued queue Px has advanced and move forward. Note that for the notification of the advancement of the rearmost end of the queue Px through output of sound and vibration, either one of sound or vibration may be employed, or both may be combined.

[0051] Furthermore, the distance the last person has advanced can be communicated by sound, vibration, etc., based on, for example, camera images acquired from a camera 31 installed on the white cane 30 or detection results from an ultrasonic sensor 32. The human body coordinate plotting unit (human body coordinate setting unit) 21 plots the positions (coordinates) of multiple human bodies that constitute the matrix Px detected by the matrix detection unit 16. Specifically, as shown in Figure 5, for example, the human body coordinate plotting unit 21 plots the heads, chests, arms, and legs of human bodies P1 to P4 included in the camera images acquired by the image acquisition unit 11 to recognize the positions of human bodies P1 to P4 and recognize whether there is space to pass through among human bodies P1 to P4.

[0052] As shown in Figure 5, the passage space determination unit 22 determines the presence or absence of a passage space S1 that a person can pass through, based on the positions (coordinates) of the human body P1 to P4 plotted in the human body coordinate plotting unit 21. The presence or absence of a passage space S that a person can pass through is determined, for example, by setting a minimum width that a person can pass through in advance. If the passage space determination unit 22 detects a space wider than that width, it determines that there is a passage space S1, and if only spaces narrower than that width are detected, it determines that there is no passage space.

[0053] The notification control unit (rearmost forward notification control unit, direction of travel notification control unit) 20 notifies the user U1 of the white cane 30 of the direction in which the space is located when the space passage determination unit 22 determines that there is space in which a person can pass. Here, when the space passage determination unit 22 determines that there is space in between the matrix Px that can be passed through, the notification control unit 20 stops outputting sound, vibration, etc., or outputs an audio message, etc., at the moment when, for example, the white cane 30 is swung from side to side and the direction of the white cane 30 points in the direction of the space to be passed through.

[0054] Furthermore, when the space passage determination unit 22 determines that there is space that can be passed through between the matrices Px, the notification control unit 20 stops the output of sound and vibration at the moment when, for example, the white cane 30 is swung from side to side and the direction of the white cane 30 points towards the space. Note that the notification of space between matrices Px using sound and vibration output may employ either sound or vibration alone, or a combination of both.

[0055] This allows visually impaired individuals to recognize the direction in which people are present and the direction in which they are not, thereby guiding them to pass through the matrix Px. <Object Detection Processing Method> The matrix detection device 10 of this embodiment has the above configuration and performs object detection processing according to the basic flow shown in Figure 6(a).

[0056] In other words, in step S11, the microcontroller and other components mounted on the white cane 30 are activated. Next, in step S12, the ultrasonic sensor 32 is used to detect whether or not there is an object within a predetermined distance range (for example, 1.0 to 2.0 m) in the direction of travel of the user U1 (direction in front of the white cane 30). The object detection process using the ultrasonic sensor 32 will be described in detail later.

[0057] Next, in step S13, the presence or absence of a human body is detected using the camera image captured by camera 31. The human detection process using the camera image captured by camera 31 will be described in detail later. Next, in step S14, it is determined whether the power is OFF or not, and the processes of steps S12 and S13 are repeated until the power is OFF.

[0058] The object detection process using the ultrasonic sensor 32 in step S12 can be explained below using the flowchart shown in Figure 6(b). Specifically, in step S21, it is determined whether the ultrasonic sensor 32 is ON or OFF, and if it is ON, the process proceeds to step S22. Next, in step S22, ultrasonic waves are emitted from the ultrasonic sensor 32 toward the front of the white cane 30 (the direction of travel of the user U1), and the reflected waves are received.

[0059] Next, in step S23, based on whether or not the reflected ultrasonic waves emitted from the ultrasonic sensor 32 in step S22 were received and their reception intensity, it is determined whether or not there is an object in the direction in front of the white cane 30. If it is determined that there is an object, the process proceeds to step S24; if it is determined that there is no object, the process returns to step S22.

[0060] Next, in step S24, since it was determined in step S23 that an object is present, the distance to the object is calculated based on the timing of the emission of ultrasonic waves from the ultrasonic sensor 32 in step S22 and the timing of the reception of the reflected waves. Next, in step S25, it is determined whether the distance to the object calculated in step S24 is less than or equal to a predetermined set distance A.

[0061] Here, if the calculated distance is less than or equal to the set distance A, the process proceeds to steps S26 and S27; if it is greater than the set distance A, the object detection process ends there. Next, in step S26, since it was determined that the distance calculated in step S25 was less than or equal to the set distance A, the notification unit 34 of the white cane 30 vibrates the vibrator with vibration pattern A to notify the user U1 that the object in front is at a predetermined distance A or less.

[0062] Similarly, in step S27, since it is determined that the distance calculated in step S25 is less than or equal to the set distance A, the notification unit 34 of the white cane 30 outputs sound pattern A from the speaker to notify the user U1 that the object in front is at a predetermined distance A or less. The person detection process using the camera image captured by the camera 31 in step S13 can be explained below using the flowchart shown in Figure 6(c).

[0063] Specifically, in step S31, the image acquisition unit 11 acquires a camera image captured by the camera 31. Next, in step S32, the human body detection unit 12 of the matrix detection device 10 uses the camera image acquired in step S31 to determine whether or not there is a person in the direction in front of the white cane 30.

[0064] If it is determined that there is a person in front of the white cane 30, the process proceeds to step S33; if it is determined that there is no person, the process returns to step S31. Next, in step S33, since it was determined that there was a person in step S32, the size of the detected human body is calculated. Next, in step S34, the distance to the person detected in step S32 is calculated from the size of the human body calculated in step S33.

[0065] Next, in step S35, it is determined whether the distance to the human body calculated in step S34 is less than or equal to a predetermined set distance B. If the calculated distance is less than or equal to the set distance B, the process proceeds to steps S36 and S37. If it is greater than the set distance B, the human detection process ends there. Next, in step S36, since it was determined that the distance calculated in step S35 is less than or equal to the set distance B, the notification unit 34 of the white cane 30 vibrates the vibrator with vibration pattern B to notify the user U1 that a person in front of them is at a distance of less than or equal to the predetermined distance B.

[0066] Similarly, in step S37, since it is determined that the distance calculated in step S35 is less than or equal to the set distance B, the notification unit 34 of the white cane 30 outputs sound pattern B from the speaker to notify the user U1 that the person in front of them is at a predetermined distance B or less. <Matrix detection processing method> The matrix detection device 10 of this embodiment, with the configuration described above, performs the following matrix detection method according to the flowchart shown in Figure 7.

[0067] In other words, as shown in Figure 7, in step S41, the image acquisition unit 11 of the matrix detection device 10 acquires a camera image captured by the camera 31. Next, in step S42, the human body detection unit 12 of the matrix detection device 10 uses the camera image acquired in step S41 to determine whether or not there is a person in the direction in front of the white cane 30.

[0068] If it is determined that there is a person in front of the white cane 30, the process proceeds to step S43; if it is determined that there is no person, the process returns to step S41. Next, in step S43, the notification control unit 20 of the queue detection device 10 notifies the user U1 that a human body was detected in step S42 by generating sound and / or vibration.

[0069] Next, in step S44, the person detection unit 13 of the queue detection device 10 determines whether or not there are multiple people detected in step S42. If multiple people are detected, the process proceeds to step S45; otherwise, the process returns to step S41. Next, in step S45, since multiple people were detected in step S44, the face detection unit 14 of the queue detection device 10 detects the faces of the multiple people, and the face direction detection unit 15 detects the orientation of the detected faces.

[0070] The face direction detection unit 15 can detect the orientation of a face by, for example, detecting the position of facial features (eyes, nose, mouth, ears), the direction of the line of sight, the orientation of the human body, etc. Next, in step S46, the matrix detection unit 16 of the matrix detection device 10 determines whether the orientations of the multiple human faces detected in step S45 are facing the same direction. If it is determined that they are facing the same direction, the process proceeds to step S47; if it is determined that they are not facing the same direction, the process returns to step S41.

[0071] Next, in step S47, since it was determined in step S46 that multiple human faces were facing the same direction, the matrix detection unit 16 of the matrix detection device 10 detects that there is a matrix Px and in which direction the matrix is ​​arranged. As a result, the user U1 can receive sound and vibrations from the white cane 30 and, for example, move to the end of the matrix and join matrix Px.

[0072] Next, in step S48, the occipital head detection unit 17 of the matrix detection device 10 determines whether or not it has detected the occipital heads of multiple human bodies. The occipital heads can be detected, for example, from the direction of the human face, the position of the hair, the position of facial features (eyes, nose, mouth, ears, etc.), the orientation of the human body, etc., included in the camera image acquired in step S41.

[0073] Here, if the occipital head detection unit 17 detects the occipital heads of multiple human bodies, the process proceeds to step S49. If the occipital heads cannot be detected, the process ends because the last position of matrix Px cannot be detected. Next, in step S49, the last position detection unit 18 of the matrix detection device 10 detects the position of the last human body constituting matrix Px from the positions of the occipital heads of multiple human bodies detected in step S47, and controls the notification unit 34 to output sound and vibration to prompt the notification control unit 20 to move to the last position.

[0074] As a result, users U1 of the white cane 30, such as visually impaired or elderly people, can, for example, when swinging the white cane 30 from side to side, will hear or vibrate when the white cane 30 points towards the end of the line, allowing them to move in the direction guided and join the end of the line Px. Next, in step S50, the line-finish advance detection unit 19 of the line detection device 10 monitors the change in distance to the detected end of the line and determines whether the end of the line has moved forward.

[0075] The change in distance from the last in line may be detected using the ultrasonic sensor 32 or using camera images, as described above. If it is determined that the last in line has moved forward, the process proceeds to step S51; if it is determined that it has not moved forward, the process waits until it is determined that it has moved forward. Next, in step S51, the notification control unit 20 of the queue detection device 10 controls the notification unit 34 to output sound or vibration indicating that the last in line has moved forward, and the process ends.

[0076] <Passage Space Detection Process> In the matrix detection device 10 of this embodiment, with the configuration described above, the following passage space detection process is performed according to the flowchart shown in Figure 8. That is, in the matrix detection device 10 of this embodiment, as shown in Figure 8, in step S61, the image acquisition unit 11 of the matrix detection device 10 acquires the camera image captured by the camera 31.

[0077] Next, in step S62, the human body detection unit 12 of the queue detection device 10 uses the camera image acquired in step S61 to determine whether or not there is a person in front of the white cane 30. If it is determined that there is a person in front of the white cane 30, the process proceeds to step S63; if it is determined that there is no person, the process returns to step S61.

[0078] Next, in step S63, the person detection unit 13 of the queue detection device 10 determines whether or not there are multiple people detected in step S62. If multiple people are detected, the process proceeds to step S64; otherwise, it returns to step S61. Next, in step S64, since multiple people were detected in step S63, the person coordinate plotting unit 21 of the queue detection device 10 plots the position (coordinates) of each person.

[0079] Next, in step S65, the passage space determination unit 22 of the matrix detection device 10 determines whether or not there is space for a person to pass through, based on the position (coordinates) of the human body plotted in step S64. If it is determined that there is space for a person to pass through, the process proceeds to step S66; if it is determined that there is no space for a person to pass through, the process ends there.

[0080] Next, in step S66, since it was determined in step S65 that there is a passage space, the notification control unit 20 of the queue detection device 10 controls the notification unit 34 to output sound and vibration to guide the user U1 of the white cane 30 in the direction of the detected passage space. This allows the device to determine whether or not there is space for a person to pass through between the detected queue Px, and if it is determined that there is a passage space, it can guide the user U1 in that direction.

[0081] Furthermore, guidance in the direction of the passage space can be easily indicated by, for example, stopping the generation of sound or vibration when the white cane 30 is pointed in the direction of the passage space while the user U1 is swinging the white cane from side to side. In addition, the queue detection device 10 of this embodiment performs the following crowd passage process according to the flowchart shown in Figure 9 with the above configuration.

[0082] In other words, as shown in Figure 9, in step S71, the image acquisition unit 11 of the matrix detection device 10 acquires a camera image captured by the camera 31. Next, in step S72, the human body detection unit 12 of the matrix detection device 10 uses the camera image acquired in step S71 to determine whether or not there is a person in the direction in front of the white cane 30.

[0083] If it is determined that there is a person in front of the white cane 30, the process proceeds to step S73; if it is determined that there is no person, the process returns to step S71. Next, in step S73, the person detection unit 13 of the queue detection device 10 determines whether or not there are multiple people detected in step S72. If multiple people are detected, the process proceeds to step S74; if no multiple people are detected, the process returns to step S71.

[0084] Next, in step S74, since it was determined in step S73 that there are multiple human bodies, the matrix detection unit 16 of the matrix detection device 10 calculates the orientation of the human bodies and the distance to the human bodies. Next, in step S75, the notification control unit 20 of the matrix detection device 10 controls the notification unit 34 to guide the user U1 by notifying them of the direction to move and the distance to the human bodies, based on the orientation of the human bodies and the distance to the human bodies calculated in step S74.

[0085] This allows the user U1 of the white cane 30 to be guided through a moving crowd (queue Px). <Main Features> As shown in Figure 3, the queue detection device 10 of this embodiment includes an image acquisition unit 11, a human body detection unit 12, a person detection unit 13, a face detection unit 14, a face direction detection unit 15, and a queue detection unit 16. The image acquisition unit 11 acquires an image taken in a predetermined direction. The human body detection unit 12 detects the presence or absence of a human body from the image acquired by the image acquisition unit 11. The person detection unit 13 detects the number of human bodies detected by the human body detection unit 12. The face detection unit 14 detects the faces of the human bodies detected by the human body detection unit 12 from the image acquired by the image acquisition unit 11. The face direction detection unit 15 detects the orientation of the faces detected by the face detection unit 14 from the image acquired by the image acquisition unit 11. The matrix detection unit 16 detects the presence or absence and direction of matrix Px based on the number of people detected by the person detection unit 13 and the orientation of the faces of the people detected by the face direction detection unit 15.

[0086] This allows for the detection of a matrix Px, which is a collection of people, and its direction, enabling guidance for visually impaired individuals, the elderly, etc., to move to the end of the matrix Px or to follow the matrix Px. [Other Embodiments] Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications are possible without departing from the spirit of the invention.

[0087] (A) In the above embodiments, the present invention has been described using examples of matrix detection devices and matrix detection methods. However, the present invention is not limited thereto. For example, the present invention may be implemented as a matrix detection program that causes a computer to execute the matrix detection method described above.

[0088] This matrix detection program is stored in the memory (storage unit) installed in the matrix detection device, and the CPU reads the matrix detection program stored in memory and causes the hardware to execute each step. More specifically, the same effect as above can be obtained by the CPU reading the matrix detection program and executing each of the steps described above. Furthermore, the present invention may be implemented as a recording medium that stores the matrix detection program.

[0089] (B) In the above embodiment, an example was given in which the coordinates of a plurality of human bodies P1 to P4 are plotted in order to detect the passage space S1 formed between a plurality of human bodies P1 to P4 that constitute the matrix Px. However, the present invention is not limited thereto. For example, the position of the head of a human body may be detected, and the gap formed between a plurality of heads may be detected as a passage space through which a person can pass.

[0090] (C) In the above embodiment, the example of notifying the user U1 of the white cane 30 of the direction to move by outputting sound or vibration has been mainly described. However, the present invention is not limited thereto. For example, when the user is swinging the white cane from side to side, an audio message such as "Please go in this direction" or "This is the direction" may be output at the moment the white cane points in the direction to move.

[0091] (D) In ​​the above embodiment, an example was given in which a camera 31 and an ultrasonic sensor 32 are used to detect objects or people in the direction of travel of the user U1 of the white cane 30 (in front of the white cane 30). However, the present invention is not limited thereto. For example, the white cane of the present invention may be equipped with a camera but not with an ultrasonic sensor.

[0092] (E) In the above embodiment, an example was given in which a matrix Px, which is a group of people, is detected using camera images captured by a camera 31 mounted on a white cane 30 used by visually impaired people, elderly people, etc. However, the present invention is not limited thereto. For example, the matrix detection device of the present invention may detect a matrix not only using a camera mounted on something that moves with the user, such as a white cane, but also using camera images captured by a fixedly positioned camera.

[0093] (F) In the above embodiment, the queue detection device 10 of the present invention was described in an example in which it is mounted on a white cane 30 used by a visually impaired person or an elderly person. However, the present invention is not limited thereto. For example, the queue detection device of the present invention may be mounted on a mobile terminal such as a smartphone worn around the neck by a visually impaired person or an elderly person, and the camera image described above may be acquired using the camera function of the mobile terminal.

[0094] Furthermore, the matrix detection device of the present invention may be used not only by visually impaired or elderly people, but also by able-bodied people. (G) In the above embodiments, the present invention has been described as a matrix detection device and a matrix detection method. However, the present invention is not limited thereto.

[0095] For example, the present invention may be implemented as a pass-through space detection device, a pass-through space detection method, and a pass-through space detection program for detecting passable space in a crowd such as a queue. (H) In the above embodiment, an example in which the queue detection device 10 of the present invention is mounted on a white cane 30 was described. However, the present invention is not limited thereto.

[0096] For example, the queue detection device of the present invention may be mounted on wheelchairs, self-driving wheelchairs, etc., used by visually impaired persons, the elderly, injured or ill persons, etc. <Note> The queue detection device according to the first invention comprises: an image acquisition unit that acquires an image taken in a predetermined direction; a human body detection unit that detects the presence or absence of human bodies from the image acquired by the image acquisition unit; a person count detection unit that detects the number of human bodies detected by the human body detection unit; a face detection unit that detects the faces of the human bodies detected by the human body detection unit from the image acquired by the image acquisition unit; a face direction detection unit that detects the orientation of the faces detected by the face detection unit from the image acquired by the image acquisition unit; and a queue detection unit that detects the presence or absence of a queue and its direction based on the number of human bodies detected by the person count detection unit and the orientation of the faces of the human bodies detected by the face direction detection unit.

[0097] The matrix detection device according to the second invention is the matrix detection device according to the first invention, further comprising a occipital head detection unit for detecting a plurality of occipital heads of human bodies detected by the matrix detection unit. The matrix detection device according to the third invention is the matrix detection device according to the second invention, further comprising a last-end detection unit for detecting the last end of the matrix from the plurality of occipital heads of human bodies detected by the occipital head detection unit.

[0098] The matrix detection device according to the fourth invention is the matrix detection device according to the third invention, further comprising a rear-end advance detection unit that detects when the rear-end detection unit has moved forward in the direction of travel. The matrix detection device according to the fifth invention is the matrix detection device according to the fourth invention, further comprising a rear-end advance notification unit that the rear-end advance detection unit has detected that the rear-end has moved forward.

[0099] The matrix detection device according to the sixth invention is the matrix detection device according to the fourth invention, wherein the detection of the last-place moving forward in the last-place moving forward detection unit is performed using the detection result from the ultrasonic sensor. The matrix detection device according to the seventh invention is the matrix detection device according to the fourth invention, wherein the detection of the last-place moving forward in the last-place moving forward detection unit is performed using image information acquired by the image acquisition unit.

[0100] The matrix detection device according to the eighth invention is the matrix detection device according to the fifth invention, wherein the rearmost position forward notification unit provides notification by outputting a predetermined sound. The matrix detection device according to the ninth invention is the matrix detection device according to the fifth invention, wherein the rearmost position forward notification unit provides notification by outputting a predetermined vibration. The matrix detection device according to the tenth invention is the matrix detection device according to any one of the first to ninth inventions, further comprising a human body coordinate setting unit that plots the coordinates of a plurality of human bodies constituting the matrix detected by the matrix detection unit.

[0101] The matrix detection device according to the 11th invention is a matrix detection device according to the 10th invention, further comprising a passage space determination unit that determines whether or not there is a space through which a person can pass based on the coordinates plotted in the human body coordinate setting unit. The matrix detection device according to the 12th invention is a matrix detection device according to the 11th invention, further comprising a direction of travel notification unit that notifies the direction in which the space is located when the passage space determination unit determines that such a space exists.

[0102] The 13th invention is a matrix detection device according to the 12th invention, wherein the direction of travel notification unit provides notification by outputting a predetermined sound. The 14th invention is a matrix detection device according to the 12th invention, wherein the direction of travel notification unit provides notification by outputting a predetermined vibration. The 15th invention is a white cane comprising: a matrix detection device according to any one of the 1st to 14th inventions; a gripping part to which the matrix detection device is attached and which is grasped by the user; and a rod-shaped main body attached so as to extend from the gripping part.

[0103] The white cane according to the 16th invention is the white cane according to the 15th invention, wherein the predetermined direction is the direction of travel of the user of the white cane.

[0104] The matrix detection device of the present invention has the effect of being able to detect matrices, which are groups of people, and their direction, and is therefore widely applicable to various tools owned by visually impaired people, the elderly, and others.

[0105] 10 Queue detection device 11 Image acquisition unit 12 Human body detection unit 13 Person detection unit 14 Face detection unit 15 Face direction detection unit 16 Queue detection unit 17 Back of head detection unit 18 Last person detection unit 19 Last person moving forward detection unit 20 Notification control unit (Last person moving forward notification control unit, Direction of travel notification control unit) 21 Human body coordinate plot unit (Human body coordinate setting unit) 22 Passage space determination unit 30 White cane 30a Grip unit 30b Main unit 31 Camera 32 Ultrasonic sensor 33 Push button 34 Notification unit P1-P6 Human body PX Queue S1 Passage space U1 User

Claims

1. A matrix detection device comprising: an image acquisition unit that acquires an image taken in a predetermined direction; a human body detection unit that detects the presence or absence of a human body from the image acquired by the image acquisition unit; a person detection unit that detects the number of human bodies detected by the human body detection unit; a face detection unit that detects the faces of the human bodies detected by the human body detection unit from the image acquired by the image acquisition unit; a face direction detection unit that detects the orientation of the faces detected by the face detection unit from the image acquired by the image acquisition unit; and a matrix detection unit that detects the presence or absence and direction of a matrix based on the number of human bodies detected by the person detection unit and the orientation of the faces of the human bodies detected by the face direction detection unit.

2. The matrix detection device according to claim 1, further comprising a occipital head detection unit for detecting a plurality of occipital heads of human bodies detected by the matrix detection unit.

3. The matrix detection device according to claim 2, further comprising a last-place detection unit for detecting the last place in the matrix from a plurality of occipital regions of human bodies detected by the occipital region detection unit.

4. The matrix detection device according to claim 3, further comprising a rear-end forward detection unit for detecting when the rear-end of the matrix detected by the rear-end detection unit has moved forward in the direction of travel.

5. The matrix detection device according to claim 4, further comprising a rearmost forward movement notification control unit that notifies the rearmost movement of the rearmost movement detected in the rearmost forward movement detection unit.

6. The matrix detection device according to claim 4, wherein the detection of the rearmost forward movement in the rearmost forward movement detection unit is performed using the detection result from the ultrasonic sensor.

7. The matrix detection device according to claim 4, wherein the detection of the rearmost position advancing in the rearmost position advancement detection unit is performed using image information acquired in the image acquisition unit.

8. The matrix detection device according to claim 5, wherein the rearmost forward notification control unit provides notification by outputting a predetermined sound.

9. The matrix detection device according to claim 5, wherein the rearmost forward notification control unit provides notification by outputting a predetermined vibration.

10. The matrix detection device according to claim 1 or 2, further comprising a human body coordinate setting unit for plotting the coordinates of a plurality of human bodies constituting the matrix detected by the matrix detection unit.

11. The matrix detection device according to claim 10, further comprising a passage space determination unit that determines whether or not there is a space through which a person can pass, based on the coordinates plotted in the human body coordinate setting unit.

12. The matrix detection device according to claim 11, further comprising a direction of travel notification control unit that notifies the direction in which the space is located when the passage space determination unit determines that the space exists.

13. The matrix detection device according to claim 12, wherein the direction of travel notification control unit provides notification by outputting a predetermined sound.

14. The matrix detection device according to claim 12, wherein the direction of travel notification control unit provides notification by outputting a predetermined vibration.

15. A white cane comprising: a matrix detection device according to claim 1 or 2; a gripping portion to which the matrix detection device is attached and which is grasped by a user; and a rod-shaped main body portion attached so as to extend from the gripping portion.

16. The white cane according to claim 15, wherein the predetermined direction is the direction in which the user of the white cane is moving.

17. A matrix detection method comprising: an image acquisition step of acquiring an image taken in a predetermined direction; a human body detection step of detecting the presence or absence of a human body from the image acquired in the image acquisition step; a person count detection step of detecting the number of human bodies detected in the human body detection step; a face detection step of detecting the faces of the human bodies detected in the human body detection step from the image acquired in the image acquisition step; a face direction detection step of detecting the orientation of the faces detected in the face detection step from the image acquired in the image acquisition step; and a matrix detection step of detecting the presence or absence of a matrix and its direction based on the number of human bodies detected in the person count detection step and the orientation of the faces of the human bodies detected in the face direction detection step.

18. A matrix detection program that causes a computer to execute a matrix detection method comprising: an image acquisition step of acquiring an image taken in a predetermined direction; a human body detection step of detecting the presence or absence of a human body from the image acquired in the image acquisition step; a person count detection step of detecting the number of human bodies detected in the human body detection step; a face detection step of detecting the faces of the human bodies detected in the human body detection step from the image acquired in the image acquisition step; a face direction detection step of detecting the orientation of the faces detected in the face detection step from the image acquired in the image acquisition step; and a matrix detection step of detecting the presence or absence of a matrix and its direction based on the number of human bodies detected in the person count detection step and the orientation of the faces of the human bodies detected in the face direction detection step.

19. A passage space detection device comprising: an image acquisition unit that acquires an image taken in a predetermined direction; a human body detection unit that detects the presence or absence of a human body from the image acquired by the image acquisition unit; a person count detection unit that detects the number of human bodies detected by the human body detection unit; a queue detection unit that detects the presence or absence of a queue based on the number of human bodies detected by the person count detection unit; and a passage space determination unit that determines the presence or absence of a space through which a person can pass based on the positions of a plurality of human bodies constituting the queue detected by the queue detection unit.

20. A method for detecting a passage space, comprising: an image acquisition step of acquiring an image taken in a predetermined direction; a human body detection step of detecting the presence or absence of a human body from the image acquired in the image acquisition step; a person count detection step of detecting the number of human bodies detected in the human body detection step; a queue detection step of detecting the presence or absence of a queue based on the number of human bodies detected in the person count detection step; and a passage space determination step of determining the presence or absence of a space through which a person can pass based on the positions of a plurality of human bodies constituting the queue detected in the queue detection step.

21. A passage space detection program that causes a computer to execute a passage space detection method comprising: an image acquisition step of acquiring an image taken in a predetermined direction; a human body detection step of detecting the presence or absence of a human body from the image acquired in the image acquisition step; a person count detection step of detecting the number of human bodies detected in the human body detection step; a queue detection step of detecting the presence or absence of a queue based on the number of human bodies detected in the person count detection step; and a passage space determination step of determining the presence or absence of a space through which a person can pass based on the positions of a plurality of human bodies constituting the queue detected in the queue detection step.

Images