JP7107697B2 - POSITION DETECTION SYSTEM AND POSITION DETECTION METHOD - Google Patents

Description

The present invention relates to a position detection system and position detection method for detecting the position of a person indoors.

Conventionally, various methods have been studied for detecting the position of a person in a building or the like. For example, there are many position detection technologies using communication terminals such as beacons, wireless tags, WiFi radio waves, geomagnetism, IMES (Indoor Messaging System), PHS (Personal Handy-phone System), and combinations of these technologies. (See Patent Document 1, for example).
The methods for detecting the position of a person described above each have advantages and disadvantages. It is a point that it is necessary to arrange dedicated equipment such as position detection sensors in the building.

JP 2011-214834 A

Dedicated devices such as sensor tags can detect the location of a person and identify individual mobile devices carried by humans. cannot be identified and specified.
Therefore, even a person who is originally prohibited from entering due to security reasons can enter a predetermined building or area by taking the place of another person if he or she carries an authenticated mobile device.

The present invention has been made in view of such circumstances, and it is possible to identify individual human beings in a building and detect the position of each of the identified individuals without using a dedicated device for position detection. It is an object of the present invention to provide a position detection system and a position detection method that can

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems . An information storage unit extracts a human face image from an image captured by a surveillance camera, performs face authentication of the person in the captured image using the reference face image, and identifies the user. position detection information storage unit storing position coordinates of a mark and size information of the mark; position coordinates of the installed surveillance camera; and the mark imaged by the surveillance camera. and the imaging direction in which the reference captured image is captured; the position coordinates of the monitoring camera that captured the user; and the reference including the mark Using the imaging direction in which the captured image is captured as a reference , the imaging direction of the captured image, the magnification of each lens used to capture the reference captured image and the captured image, the position coordinates of the mark, and the size information of the mark. and the size information of the user , and a position detection unit that calculates the position coordinates of the user in a three-dimensional coordinate system from the captured image .

In the position detection system of the present invention, when the position detection unit detects the position coordinates from each of the plurality of captured images for the same user , the position coordinates are averaged, and the average value is calculated as It is characterized by outputting as position coordinates.

The position detection system of the present invention further includes a surveillance camera control unit that controls the imaging direction of the captured image of the surveillance camera and the lens magnification, and the surveillance camera control unit controls the detection of a human face when a human face is detected. and changing the magnification of the lens to a resolution that allows the face image in the captured image captured by the surveillance camera to be compared with the reference face image with a predetermined accuracy.
The position detection system of the present invention is characterized by comprising a monitoring control unit that controls a device to be controlled according to the position coordinates based on the authentication result of the face authentication and the calculated position coordinates. .

In the position detection method of the present invention, a human face image is extracted from an image captured by a surveillance camera , and a reference face image, which is an image of a user's face stored in an authentication information storage unit, is used to detect the image. Face recognition of the person in the image to identify the user ; The imaging direction of the image, the magnification of each lens used to capture the reference captured image and the captured image, the position coordinates and size information of the mark stored in the position detection information storage unit, and the information stored in the authentication information storage unit. and a position detection process of calculating position coordinates of the user in a three-dimensional coordinate system from the captured image using the size information of the user .
Further, the position detection method of the present invention is a position detection method according to the above-described position detection method, based on the authentication result of face authentication and the calculated position coordinates, monitoring control for controlling a device to be controlled according to the position coordinates. It is characterized by including a process.

According to the present invention, there is provided a position detection system and a position detection method that can identify individual human beings in a building and detect the position of each of the identified individuals without using dedicated equipment for position detection. can do.

1 is a schematic block diagram showing a configuration example of a position detection system according to one embodiment of the present invention; FIG. 4A and 4B are diagrams for explaining operations of a monitoring camera control unit 103 in a normal state and a monitoring state; FIG. 3 is a diagram showing a configuration example of an authentication information table written and stored in an authentication information storage unit 108; FIG. 3 is a diagram showing a configuration example of a position detection information table pre-written and stored in the position detection information storage unit 109. FIG. FIG. 4 is a conceptual diagram for detecting a user's position from an image captured by a surveillance camera; 4 is a diagram showing a configuration example of a detection position history table pre-written and stored in a detection position storage unit 110. FIG. 4 is a flowchart showing an operation example of position detection processing of the position detection system according to the embodiment of the present invention;

A position detection system according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram showing a configuration example of a position detection system according to one embodiment of the present invention. The position detection system 1 includes a position detection server 10 and a surveillance camera 12, respectively. In this embodiment, a plurality of monitoring cameras 12 are provided. A plurality of surveillance cameras 12 are installed at different locations inside or outside the building. In addition, each of the position detection server 10 and the monitoring camera 12 is connected to an information communication network 500 including the Internet, a local area network, etc., and exchanges data with each other. Each of the monitoring cameras 12 is a camera whose horizontal angle (pan) and vertical angle (tilt) and lens magnification can be arbitrarily controlled, and each of the plurality of cameras is installed at different locations inside or outside the building. there is

The position detection server 10 includes a captured image input unit 101, a face detection unit 102, a monitoring camera control unit 103, a face authentication unit 104, a position detection unit 105, a monitoring control unit 106, a captured image storage unit 107, an authentication information storage unit 108, A position detection information storage unit 109 and a detection position storage unit 110 are provided.
The captured image input unit 101 inputs captured images captured by each of the monitoring cameras 12 from the respective monitoring cameras 12 via the information communication network 500 . Then, the captured image input unit 101 outputs the captured image input from each of the monitoring cameras 12 to the face detection unit 102, and also stores the captured image storage unit 107 with the monitoring camera identification information of the monitoring camera 12. A time stamp is added and written and stored.

The face detection unit 102 detects a face in the captured image supplied from the captured image input unit 101, that is, detects whether or not a human face is present in the captured image. When a human face is detected in the captured image, face detection section 102 outputs a detection signal indicating that the face has been detected to monitoring camera control section 103 . Here, any form of face detection algorithm may be used. For example, by extracting the outline of a face and determining whether the color of the face (that is, the skin of the face) within this outline is pale yellowish red (pale orange, so-called skin color), the image of the outline is the image of the human face. The configuration may be such that it is determined and detected.

When the detection signal is supplied from the face detection unit 102, that is, when the face detection in the captured image is used as a trigger, the monitoring camera control unit 103 shifts the control mode from the normal state to the face authentication state. In the normal state, the monitoring camera control unit 103 moves the imaging direction of the captured image of the monitoring camera 12 horizontally and vertically so that each part within the imaging range of the monitoring camera 12 is captured. Also, in the normal state, when moving in the imaging direction within the imaging range, it is fixed at a constant lens magnification that can always capture a constant range.

On the other hand, in the face authentication state, the monitoring camera control unit 103 tracks the position of the face in the captured image where the face is detected, and the detected face area in the captured image is the largest and the entire face area is Control the imaging direction and lens magnification of the surveillance camera to be included. At this time, the monitoring camera control unit 103 controls the monitoring camera 12 so that the lens magnification is increased compared to the normal state, and the resolution of the face in the captured image is increased. In this embodiment, an image captured in this monitoring state is shown as a high-resolution image.
Then, when an end signal indicating that the face authentication processing is completed is supplied from the face authentication unit 104, which will be described later, the surveillance camera control unit 103 shifts the control of the surveillance camera from the face authentication state to the normal state.

FIG. 2 is a diagram for explaining the operations of the surveillance camera control unit 103 in the normal state and the surveillance state. FIG. 2(a) shows the configuration of the surveillance camera 12. As shown in FIG. The monitoring camera 12 includes a monitoring camera body 12A, a driving mechanism 12B, a driving mechanism 12C, and a lens portion 12D. The surveillance camera 12 is installed on a ceiling, a wall, a pillar, or the like via a drive mechanism 12B. The drive mechanism 12B, for example, rotates the surveillance camera main body 12A parallel to a two-dimensional plane consisting of the X-axis and the Y-axis (reciprocating movement in the direction of arrow Q), that is, rotates about an axis parallel to the Z-axis (imaging Control of the horizontal angle (pan) to move back and forth in direction.
Further, the drive mechanism 12C moves the surveillance camera main body 12A parallel to the Z-axis (reciprocating movement in the arrow P direction in the imaging direction), that is, vertical movement in the imaging direction (vertical angle (tilt) control).

FIG. 2(b) explains control of the imaging direction of the monitoring camera 12 within the predetermined imaging range 300. As shown in FIG. An imaging range 300 indicates a range in which the monitoring camera 12 can capture an image by controlling the imaging direction (horizontal angle and vertical angle) of the monitoring camera 12 . Here, the driving mechanism 12B adjusts the imaging direction within the horizontal angle range as the maximum changeable angle range in the movement of the horizontal angle in the imaging direction. Further, the drive mechanism 12C adjusts the imaging direction within the vertical angle range as the maximum changeable angle range in the movement of the vertical angle in the imaging direction. Therefore, the imaging range 300 indicates the range that the monitoring camera 12 can capture when the horizontal angle of the imaging direction of the monitoring camera 12 is moved within the horizontal angle range and the vertical angle is moved within the vertical angle range. .
In a normal state, for example, the surveillance camera 12 fixes the vertical angle within the imaging range 300 , controls the horizontal angle in the L (left) direction, moves the imaging direction of the captured image, and moves the imaging direction to the left of the imaging range 300 . When the horizontal angle corresponding to the end of (the movement limit of the horizontal angle range) is reached, the vertical angle is changed upward (in the direction of arrow U) or downward (in the direction of arrow D) by a predetermined angle. After the predetermined angle change, the monitoring camera 12 fixes the vertical angle, controls the horizontal angle in the R (right) direction, and moves the imaging direction. The monitoring camera 12 sequentially captures images within the imaging range 300 by controlling the horizontal angle and vertical angle described above.

In addition, within the imaging range 300, the captured image S is captured in the normal state, but the monitoring camera control unit 103 includes the face image 600, which is the image of the human face, with the detection of the human face 600 as a trigger. In order to increase the resolution of the image of the area, the monitoring camera 12 is controlled to direct the imaging direction toward the human face and drive the lens portion 12D to increase the lens magnification. As a result, the monitoring camera 12 increases the ratio of the region of the face image 600 in the captured image, that is, substantially increases (enlarges) the resolution of the image 600 as a captured image captured in the monitoring state. An enlarged face image 601 of a human face in the resolution image S' is captured.

Returning to FIG. 1, the face authentication unit 104 refers to the authentication information table in the authentication information storage unit 108, and compares the face image in the high-resolution image with the reference face image in the authentication information storage unit 108. Authenticate the person in the image.

FIG. 3 is a diagram showing a configuration example of an authentication information table written and stored in the authentication information storage unit 108. As shown in FIG. The authentication information table has columns of personal identification information, reference face image #1 index to #3 index, and size information for each record. Personal identification information is, for example, identification information given to users working in a predetermined building to individually identify each user. Each of the reference face image #1 index to #3 index is an address indicating an area in which the user's front face image, right face image, and left face image corresponding to the personal identification information are written. The size information indicates the size of the user, such as height. The reference face image is, for example, a face image of a user who has obtained permission to enter the building in advance.

Returning to FIG. 1, face authentication section 104 determines whether the face image in the high-resolution image is included in the reference face images at addresses indicated by index #1 to index #3 in the authentication information table of authentication information storage section 108. Detect whether or not there is. As a result, the face authentication unit 104 can detect the presence or absence of a person (suspicious person) other than the pre-authenticated user in the building from the image captured by the monitoring camera 12 . The face authentication unit 104 outputs an end signal to the monitoring camera control unit 103 when the face authentication processing based on the authentication information table is completed.

The position detection unit 105 tracks the image of the user face-authenticated by the face authentication unit 104 in the captured image captured in the normal state, refers to the position detection information table of the position detection information storage unit 109, and determines the position detection information table. The position coordinates of the three-dimensional coordinates in the user's building are detected.

FIG. 4 is a diagram showing a configuration example of a position detection information table pre-written and stored in the position detection information storage unit 109. As shown in FIG. The position detection information table has columns for surveillance camera identification information, surveillance camera position coordinates, reference captured image index, mark position coordinates, and mark size for each record. The monitoring camera identification information is, for example, identification information that individually identifies each of the monitoring cameras installed at various locations in a predetermined building. The monitoring camera position coordinates are three-dimensional position coordinates in the building where the monitoring camera indicated by the monitoring camera identification information is installed. The reference captured image index is an address indicating the area in which the captured image of the management range captured by the monitoring camera without the user being captured is written. Only one is described as a description example in this position detection information table. However, in the present embodiment, image data of a plurality of reference captured images corresponding to each of the imaging directions of the imaging angle for each combination of a plurality of horizontal angles and vertical angles is stored in the position detection information storage unit 109. there is The mark position coordinates are the position coordinates of the three-dimensional coordinates in the building of the reference mark in the reference captured image. The marks are signboards, signs, patterns on ceilings, floors and walls, doors and windows of rooms, plates indicating room names, and the like, which are captured in the reference captured image. The mark size indicates the dimension of the mark.

Returning to FIG. 1, the position detection unit 105 detects the size of the face-authenticated user, for example, the number of pixels SP corresponding to the height LP in the captured image captured in the normal state. In addition, the position detection unit 105 stores a reference captured image having an imaging angle indicating an imaging direction close to an imaging angle (a set of a horizontal angle and a vertical angle of the imaging direction) indicating the imaging direction of the captured image as an information storage unit for position detection. 109. Then, the position detection unit 105 corrects the reference captured image to a magnification corresponding to the lens magnification of the captured image (zoom % indicating a ratio to a predetermined magnification), and detects the number of pixels SM corresponding to the predetermined mark dimension LM. do. The position detection unit 105 obtains the distance XM between the mark in the reference captured image and the surveillance camera 12 from the respective position coordinates. The position detection unit 105 obtains the distance XP from the surveillance camera 12 to the user from the relationship of (LP/SP):XP=(LM/SM):XM. Then, the position detection unit 105 uses the position coordinates (X, Y, Z) where the monitoring camera 12 is installed, the imaging direction (horizontal angle and vertical angle), and the distance XP to determine the position of the user. The position coordinates of the three-dimensional coordinates in the building are obtained. Further, the position detection unit 105 writes and stores the position coordinates of the user in the detection position history table of the detection position storage unit 110 in association with the user's personal identification information and the time stamp.

FIG. 5 is a conceptual diagram of detecting a user's position from an image captured by a surveillance camera.
In FIG. 5, in the captured image captured by the monitoring camera 12_1, from the height LP of the user X, the height LM of the door 800 (mark), and the distance XM1 from the monitoring camera 12_1, by the method already described, A distance XP1 from the monitoring camera 12_1 to the user X is calculated. Then, the position detection unit 105 obtains the position coordinates W1 of the user X in the captured image of the monitoring camera 12_1 from the position of the monitoring camera 12_1 and the distance XP1.
Further, the position detection unit 105 detects the position of the surveillance camera 12_2 based on the height LP of the user X, the height LM of the door 800 (mark), and the distance XM2 from the surveillance camera 12_2 in the captured image captured by the surveillance camera 12_2. to the user X is calculated. Then, the position detection unit 105 obtains the position coordinates W2 of the user X in the captured image of the monitoring camera 12_2 from the position of the monitoring camera 12_2 and the distance XP2.
Then, the position detection unit 105 obtains the position coordinates of the user from each of the position coordinates W1 and the position coordinates W2.

FIG. 6 is a diagram showing a configuration example of a detection position history table pre-written and stored in the detection position storage unit 110. As shown in FIG. The detected position history table is stored in the detected position storage unit 110 with the personal identification information of each user added. The detection position history table has columns of time stamp, monitoring camera identification information, position coordinates, . . . , monitoring camera identification information, position coordinates, and average position coordinates for each record. The time stamp indicates the time when the position coordinates were detected. The monitoring camera identification information is, for example, identification information that individually identifies each of the monitoring cameras installed at various locations in a predetermined building. The position coordinates are the coordinates of the user in the three-dimensional coordinates of the building detected by the surveillance camera indicated by the surveillance camera identification information. The position coordinates indicate the coordinates of each surveillance camera that detected the same user at the same time. The average position coordinate is an average value of all position coordinates obtained from images captured by two or more monitoring cameras at the same time.

Returning to FIG. 1, the position detection unit 105 averages all the position coordinates obtained from the captured images captured at the same time by two or more surveillance cameras to obtain the average position coordinates. Further, when only one monitor camera detects the position coordinates of the user, the position detection unit 105 uses the position coordinates as the average value coordinates.
For example, when a building supervisor wants to check the flow line of a given user, an input device such as a keyboard (not shown) sends the user's personal identification information and flow line confirmation processing information to the location detection server. Enter 10. As a result, the monitoring control unit 106 refers to the detection position storage unit 110 and detects the detection position history table of the input personal identification information of the specified user. Then, the monitoring control unit 106 refers to the detected position history table, attaches the position coordinates of the specified user to the structural drawing of the building of the three-dimensional coordinates with the time of the time stamp, and continuously uses the position coordinates. It is displayed as a flow line that shows the movement of people. In addition, when the face authentication unit 104 confirms a person (suspicious person) who is not the user, the monitoring control unit 106 sends an alarm corresponding to security, and the position of the monitoring camera 12 that captured the photograph of the suspicious person. , is displayed on the structural drawing of the building in three-dimensional coordinates on the display screen of the display device.

In addition, when there is a user set as a VIP (Very Important Person) in advance (hereafter referred to as a VIP user), the monitoring control unit 106 detects this VIP user in the detection position history table of the detection position storage unit 110 . At this time, VIP preferential treatment is performed for predetermined controlled objects 800 (each device provided in the building) in the building. For example, when the detection position history table detects that the VIP user has moved in front of the elevator, the monitor control unit 106 sets the elevator on any floor whose button is not pressed as the control object 800, and moves to another floor. A configuration may be adopted in which control is performed to move non-stop to the floor of the VIP user without stopping.

Further, when the face authentication unit 104 detects a suspicious person who is not registered as a user in the authentication information table of the authentication information storage unit 108, the monitor control unit 106 obtains the position coordinates of the suspicious person, An electric curtain (or an electric shutter) in a room is set as a controlled object 800, and the electric curtain is closed to shield the room, and the electric key of the door of the room requiring security management is locked so that no one can enter. It is good also as a structure which performs the process which carries out.
Further, when the face authentication unit 104 authenticates the face registered in the authentication information table of the authentication information storage unit 108 as a user, the monitor control unit 106 performs operation of each device set for each user. may be configured to perform For example, when a user is detected at predetermined position coordinates, the monitoring control unit 106 performs preset preparations corresponding to the user's work, such as turning on the personal computer.
Further, in the present embodiment, an elevator, an electric curtain, an electric key, and a personal computer have been described as examples of the controlled object 800, but the controlled object 800 is not limited to these.

FIG. 7 is a flowchart showing an operation example of position detection processing of the position detection system according to one embodiment of the present invention. The following flowchart shows control for one of a plurality of monitoring cameras 12 , and similar control is performed for other monitoring cameras 12 .
Step S1:
The monitoring camera control unit 103 performs control in a normal state, that is, captures the image of the monitoring camera 12 with a predetermined lens magnification, and controls the imaging direction of the monitoring camera 12 .

Step S2:
The monitoring camera 12 captures a captured image within the imaging range 300 using, for example, a moving image, and outputs the captured captured image to the position detection server 10 .
In other words, the monitoring camera 12 sequentially outputs captured images in units of moving image frames to the captured image input unit 101 in the position detection server 10 in chronological order.

Step S3:
The captured image input unit 101 sequentially outputs captured images supplied from the monitoring camera 12 to the face detection unit 102 .
The face detection unit 102 detects whether or not an image of a human face is included in the input captured image. At this time, if the captured image contains an image of a human face, the face detection unit 102 outputs a detection signal to the monitoring camera control unit 103, and advances the process to step S4. On the other hand, when the captured image does not include an image of a human face, the face detection unit 102 advances the process to step S1.

Step S4:
When the detection signal is supplied, the monitoring camera control unit 103 shifts the imaging control of the monitoring camera 12 from the normal state to the face authentication state.
Then, the monitoring camera control unit 103 controls the imaging direction of the monitoring camera control unit 103 so that the imaging direction is directed in the direction of the image of the human face in the captured image, for example, in the direction in which the center of the face image enters the center of the captured image. Control to move the direction.
In addition, the monitoring camera control unit 103 controls the lens magnification so that the face image does not protrude from the frame of the captured image in the moved imaging direction. That is, the monitoring camera control unit 103 increases the lens magnification to obtain an image in which the outline of the face image is in contact with the frame of the captured image compared to the normal state, and captures a high-resolution image of the high-resolution face image. The surveillance camera 12 is controlled so as to

Step S5:
The monitoring camera 12 sequentially supplies the position detection server 10 with high-resolution images captured with a predetermined lens magnification under the control of the monitoring camera control unit 103 in the monitoring state.
In the monitoring state, the captured image input unit 101 writes the high-resolution image supplied from the monitoring camera 12 to the captured image storage unit 107 as a data set together with the identification information of the monitoring camera 12 and the time stamp. Let
The captured image input unit 101 also outputs the high-resolution image to the face authentication unit 104 .

Step S6:
The face authentication unit 104 compares the face image in the high-resolution image supplied from the captured image input unit 101 with the reference face image stored in the authentication information storage unit 108 to identify the user. Then, face authentication section 104 outputs an end signal indicating that face authentication has ended to surveillance camera control section 103 .

Step S7:
When the detection signal is supplied, the monitoring camera control unit 103 shifts the imaging control of the monitoring camera 12 from the face authentication state to the normal state. That is, the monitoring camera control unit 103 does not track the face image, and controls the azimuth angle and elevation angle of the imaging direction of the monitoring camera 12 in the normal state described with reference to FIG. 2B.
Also, the monitoring camera control unit 103 restores the lens magnification of the monitoring camera 12 to the normal magnification, and advances the process to step S8.

Step S8:
The position detection unit 105 has already obtained the position coordinates of the user identified by face authentication based on the captured image captured by the monitoring camera 12 in the normal state and the information of the monitoring camera 12 that captured this captured image. Obtained by the process described.

Step S9:
The position detection unit 105 writes the position coordinates obtained for each user into the detected position history table corresponding to the user in the detected position storage unit 110 together with the time stamp. Here, when there are a plurality of surveillance cameras 12, the position detection unit 105 adjusts the position coordinates obtained by each of the plurality of surveillance cameras 12 to obtain the final position coordinates of each user.

As described above, according to the present embodiment, for example, a general pan-told camera is used as a monitoring camera, and the user's face is authenticated using the captured image of this monitoring camera. Since the positional coordinates are obtained from the image captured by the surveillance camera, it is possible to easily detect where each user is in the building without using a dedicated device for positional detection.
In addition, according to this embodiment, since the position coordinates of each of the user whose individual can be identified and the suspicious person whose individual cannot be identified can be obtained in real time from the surveillance camera, the convenience of the controlled object corresponding to each user can be obtained. , or control of a control target against a suspicious person in terms of security can be effectively performed.

In addition, in the present embodiment, as already described, the captured image captured by the monitoring camera fixed to the ceiling or wall is used to identify the individual user and detect the position of the user. For personal identification and location detection, images captured by imaging devices attached to mobile objects may be used as a complement to surveillance cameras. Here, the imaging device attached to a moving object means an imaging device mounted on a patrol robot, an imaging device provided on the shoulder or head of a security guard, or a mobile device (including a smartphone) of another user. imaging device, etc. When position detection is performed, if the position of a mobile object is specified, by measuring the distance from the position coordinates of the mobile object to the user to be detected, It is possible to obtain the position coordinates of the user. At this time, if the position coordinates of the moving object and the imaging direction of the imaging device are known, the reference captured image of the monitoring camera in the position detection information table of the position detection information storage unit 109, which is closest to the position coordinates, is transferred to both positions. A reference captured image to be compared with the captured image attached to the moving body is obtained by performing affine transformation on the coordinates, the imaging direction, and the like. In addition, an image of the moving object's patrol route is captured in advance as a reference captured image, and is written and stored in advance in the position detection information table of the position detection information storage unit 109 for each imaging device of the moving object. You can leave it.

Also, by recording a program for realizing the functions of the position detection server 10 in FIG. It is also possible to control the imaging direction and lens magnification of the camera 12, perform face authentication, and detect the position coordinates of a user who has been identified as an individual. It should be noted that the "computer system" referred to here includes hardware such as an OS and peripheral devices.

The "computer system" also includes the home page providing environment (or display environment) if the WWW system is used.
The term "computer-readable recording medium" refers to portable media such as flexible discs, magneto-optical discs, ROMs and CD-ROMs, and storage devices such as hard discs incorporated in computer systems. Furthermore, "computer-readable recording medium" means a medium that dynamically retains a program for a short period of time, like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. It also includes those that hold programs for a certain period of time, such as volatile memories inside computer systems that serve as servers and clients in that case. Further, the program may be for realizing part of the functions described above, or may be capable of realizing the functions described above in combination with a program already recorded in the computer system.

Although the embodiments of the present invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments, and designs and the like are included within the scope of the gist of the present invention.

DESCRIPTION OF SYMBOLS 1... Position detection system 10... Position detection server 12... Surveillance camera 12A... Surveillance camera main bodies 12B, 12C... Drive mechanism 101... Captured image input unit 102... Face detection unit 103... Surveillance camera control unit 104... Face authentication unit 105... Position Detection unit 106 Monitoring control unit 107 Captured image storage unit 108 Authentication information storage unit 109 Position detection information storage unit 300 Imaging range 500 Information communication network 600 Face image 601 Enlarged face image 800 Control target S ... captured image S' ... high-resolution image

Claims (7)

an authentication information storage unit storing a reference face image, which is an image of a user's face, and size information of the user;
a face authentication unit that extracts a human face image from an image captured by a surveillance camera , performs face authentication of the person in the image using the reference face image, and identifies the user ;
a position detection information storage unit storing the position coordinates of the mark and the size information of the mark;
a monitoring camera information storage unit storing position coordinates of the installed monitoring camera, a reference captured image including the mark captured by the monitoring camera, and an imaging direction in which the reference captured image is captured; ,
The imaging direction of the captured image, the reference captured image, and the captured image, with reference to the position coordinates of the surveillance camera that captured the user and the imaging direction in which the reference captured image including the mark was captured . position coordinates of the user in a three-dimensional coordinate system from the captured image using the lens magnification of each of the images, the position coordinates of the mark, the size information of the mark, and the size information of the user a position detection unit that calculates ;
A position detection system comprising: When the user whose position coordinates are detected from each of the plurality of captured images is the same, the position detection unit averages the position coordinates and outputs the average value as the position coordinates of the user. The position detection system according to claim 1 , characterized by: further comprising a surveillance camera control unit for controlling each of the imaging direction and lens magnification of the image captured by the surveillance camera;
When a human face is detected, the monitoring camera control unit adjusts the lens magnification to a resolution that allows the face image in the captured image captured by the monitoring camera to be compared with the reference face image with a predetermined accuracy. change the
The position detection system according to claim 1, characterized in that: The surveillance camera is an imaging device attached to a mobile object .
The position detection system according to any one of claims 1 to 3 , characterized in that: A monitoring control unit that controls a device to be controlled according to the positional coordinates based on the authentication result of the face authentication and the calculated positional coordinates.
5. The position detection system according to any one of claims 1 to 4, comprising: A face image of a person is extracted from an image captured by a surveillance camera , and a reference face image, which is an image of a user's face stored in an authentication information storage unit, is used to authenticate the face of the person in the image. a face authentication process for performing and identifying the user ;
Based on the coordinates of the surveillance camera in the captured image and the imaging direction in which the reference captured image including the mark was captured, the imaging direction of the captured image, and the lens magnification of each of the reference captured image and the captured image. , the position coordinates and size information of the mark stored in the position detection information storage unit and the size information of the user stored in the authentication information storage unit are used to extract the three-dimensional image of the user from the captured image; a position detection process of calculating position coordinates in a coordinate system ;
A position detection method comprising: A supervisory control process for controlling a device to be controlled according to the positional coordinates based on the authentication result of the face authentication and the calculated positional coordinates.
7. The position detection method according to claim 6, comprising:

Images