JP2007315841A - Moving object detection system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a moving object detection system capable of easily measuring positions of a plurality of cameras for performing area monitoring, by solving the problems wherein, though there is a method for measuring beforehand the positions of the plurality of cameras by a measure or the like when performing area monitoring, measurement of each camera position is troublesome, and thereby utilization of the moving object detection system for area monitoring is often avoided. <P>SOLUTION: The system has at least three cameras 101-1 to 3 arranged on a monitoring area, three light emitting devices provided close to each camera, and a signal processing device 105 into which each output from the cameras is input, respectively. The system has a constitution wherein the signal processing device has an operation part and a storage part 106, and the storage part stores a coordinate showing a space of the monitoring area, and emission from each light emitting device is imaged by each camera, and an operation device calculates the position on each coordinate of the three light emitting devices based on light emitting position information of the three imaged light emitting devices and the coordinate showing the space of the monitoring area stored in the storage part. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a moving object detection system, and more particularly to a moving object detection system that records an intrusion position of a moving object and a moving path of the moving object.

  Surveillance systems using an imaging device such as a TV (television) camera have been widely used in the past, but many of them are so-called manned surveillance systems in which surveillance is performed while watching an image displayed on a monitor. Monitoring system. In this manned monitoring system, it is necessary for the monitor to always see images displayed on the monitor and identify intruders such as humans and cars entering the monitored area in real time. Is a heavy burden. Due to the limited human concentration, the manned monitoring system has a problem in terms of reliability, because it is not possible to ignore the occurrence of intruding objects. In addition, with the widespread use of surveillance cameras, one surveillance person often monitors the images of many TV cameras using a plurality of monitors. Thus, even when an intruding object is simultaneously captured by a plurality of TV cameras, the observer may miss the intruding object. Therefore, instead of monitoring by such a person, the intruding object is automatically detected from the image captured by the TV camera by image processing, and the viewing direction and angle of view of the TV camera are automatically adjusted according to the movement of the intruding object. However, in recent years, there has been a strong demand for an automatic tracking monitoring system that performs predetermined notification and warning.

  In order to realize such a system, it is necessary to have a function of detecting an intruding object from an image signal and detecting the movement of the intruding object using a predetermined monitoring method. One example of a monitoring method for detecting such an intruding object is a method called a difference method (see, for example, Patent Document 1), which has been widely used. In the difference method, an input image obtained from a TV camera is compared with a reference background image created in advance, that is, an image in which an intruding object is not captured, for example, a luminance value difference is obtained for each pixel, and the difference is obtained. A region having a large value is detected as an object.

  A method called a template matching method is also well known as an example of a monitoring method for detecting the amount of movement of an intruding object. In the template matching method, an image of an intruding object detected by a difference method or the like is registered as a template, and a position most similar to the template image is detected among images sequentially input. Normally, when tracking a target object using template matching, in order to follow the change in the posture of the target object, the image of the position of the target object detected by the matching process is sequentially updated as a template, and the target object is A way to track.

  Thus, when detecting an intruding object in the monitoring target area or capturing a movement path after the intruding object has entered (this is referred to as a flow line capturing), a plurality of cameras surround the monitoring target area. It is useful to place In addition, such flow line capture technology analyzes what products customers are interested in in department stores, department stores, event venues, etc., and analyzes the arrangement of products and the display of spot products. It can be used for analysis materials. Here, for convenience, monitoring a target area using a plurality of cameras will be referred to as area monitoring.

  Now, in order to perform the area monitoring as described above, it is necessary to know the positions of a plurality of cameras in advance. As a method for that purpose, conventionally, the positions of the cameras are measured by a measure or the like, There are several methods such as reading the camera position from the floor where the coordinates are drawn, but it is complicated to measure the position including the height difference, and it is temporarily installed at a specific date and time at the event venue etc. In addition, when the position of the camera is frequently changed depending on the arrangement of objects, the use of the moving object detection system for area monitoring is often avoided. Therefore, it is desired to realize a moving object detection system that can monitor an area without requiring much labor and labor.

JP-A-9-73541

  When performing area monitoring, there is a method to measure the position of multiple cameras in advance with a measure, etc., but measuring the camera position is cumbersome, and the use of a moving object detection system for area monitoring is discouraged. Often done.

  An object of the present invention is to provide a moving object detection system that can easily measure the positions of a plurality of cameras that perform area monitoring.

  Another object of the present invention is to provide a moving object detection system that can easily change the positions of a plurality of cameras that perform area monitoring.

  Still another object of the present invention is to provide a moving object detection system that can easily capture the position or flow line of a moving object.

  The moving object detection system of the present invention includes at least first, second, and third cameras disposed in a monitoring area, and first, second, and third light emitting devices provided in proximity to the respective cameras. And a signal processing device to which the output of the camera is input. The signal processing device has a calculation unit and a storage unit. The storage unit stores coordinates representing the space of the monitoring area. The light emission of each of the light emitting devices is captured by the respective cameras, and the arithmetic device is stored in the storage unit and the light emission position information of the captured first, second, and third light emitting devices. The position of the first, second, and third light emitting devices on the coordinates is calculated based on the coordinates representing the space of the monitoring area.

  Further, in the moving object detection system of the present invention, the light emitting device is caused to emit light again at a predetermined time, is imaged with the camera, the position of the light emitting device is recalculated, and the light emitting device already calculated It is configured to correct the position on the coordinates.

  Further, in the moving object detection system of the present invention, a measurement having a predetermined length provided with the fourth and fifth cameras and the fourth and fifth light emitting devices different from the camera in the monitoring area. A jig is disposed, and the first to fifth cameras emit light emitted from the first to fifth light emitting devices. The arithmetic unit is arranged on the coordinates of the first, second, and third light emitting devices. Is configured to calculate the position of.

  Further, in the moving object detection system of the present invention, the detection target object that has entered the monitoring area is further imaged by the first, second, and third cameras, and the arithmetic unit is configured to capture the detected detection target object. The position of the detection target object on the coordinates is calculated based on the imaging position information and the coordinates representing the space of the monitoring area stored in the storage unit.

  In the moving object detection system of the present invention, the detection target object entering the monitoring area is imaged at a predetermined interval, and the position of the captured detection target object on the coordinates is calculated and stored. The

  As described above, according to the present invention, the positions of a plurality of cameras located in the monitoring area can be easily measured, and the positions of the plurality of cameras are measured again at an appropriate time. It is possible to easily correct or change the camera position obtained in the above. Furthermore, since it is easy to capture a flow line that records the intrusion of the detection target object and the movement state of the detection target object with time, there is a feature that various moving object detection systems other than the monitoring system can be realized.

  An embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 shows a block diagram of a schematic configuration of an embodiment of the present invention. In FIG. 1, reference numeral 100 denotes a monitoring space for an intruding object or a movement line capturing space for a moving object. Hereinafter, these spaces are abbreviated as monitoring areas. Reference numerals 101-1, 101-2, and 101-3 denote cameras. In addition, when representing a camera, the camera 101 is referred to. The plurality of cameras 101 are arranged so that there are no blind spots in the monitoring area. In this embodiment, three cameras are shown, but three or more cameras can be installed according to the situation of the monitoring area. Details of the camera 101 will be described later. Reference numerals 102-1, 102-2, and 102-3 are columns for mounting the cameras 101-1, 101-2, and 101-3, and transmission lines for signals of the camera 101 and control signals are arranged therein. ing. Reference numerals 103-1, 103-2, and 103-3 denote transmission lines that connect the cameras 101-1, 101-2, and 101-3 to the network transmission path 104. The network transmission path 104 is configured by a LAN (Local Area Network), the Internet, a public line, or the like. Hereinafter, the transmission path 104 is abbreviated.

  A signal processing device 105 is connected to the transmission path 104. Reference numeral 106 denotes an external storage device, 107 an operation unit such as a keyboard, and 108 a display unit. The signal processing device 105, the external storage device 106, the operation unit 107, and the display unit 108 will be described later. Reference numeral 109 denotes a measuring jig for measuring the length (or distance), azimuth, and horizontal. This measuring jig will also be described later. Reference numerals 110-1 and 110-2 denote cameras, which have the same configuration as the camera 101 described above. Details will be described later. P indicated by a cross indicates an intruding object or a moving object. Hereinafter, it is referred to as a detection target object P. In this embodiment, the camera 101 and the signal processing device 105 are connected by the wired transmission line 103 and the transmission path 104, but they can also be connected wirelessly.

  FIG. 2 is a block diagram for explaining specific configurations of the signal processing device 105, the external storage device 106, the operation unit 107, and the display unit 108 shown in FIG. In FIG. 2, reference numeral 201 denotes an input / output terminal connected to the transmission path 104. The signal processing device 105 includes an image input unit 202, a light source control output unit 203, an operation input unit 204, an image memory 205, an MPU (Micro Processing Unit) 206, a work memory 207, an external input / output unit 208, an image output unit 209, an alarm. An output unit 210 and a data bus 211 are included. 212 is an alarm device, for example, a warning light. In addition, the same code | symbol is attached | subjected to the same thing as FIG. The external storage device 106 can also be provided inside the signal processing device 105, but will be described as the external storage device 106 in the following description.

  The output of the operation unit 107 is connected to the data bus 211 via the operation input unit 204. The external storage device 106 is connected to the data bus 211 via the external input / output unit 208, the display unit 108 is connected to the data bus 211 via the image output unit 209, and the warning device 212 is an alarm output unit. It is connected to the data bus 211 via 210. The MPU 206 and the work memory 207 are connected to the data bus 211 as they are. The warning device 212 is a device that is attached when the present embodiment is used as a monitoring system, and is not necessary except for the monitoring system.

  The camera 101 has a function of photographing each camera in order to specify the position of each camera and a function of capturing the detection target object P in the field of view and capturing a detection target object P and outputting a video signal. The captured video signal is stored in the image memory 205 from the input / output terminal 201 via the image input unit 202 and the data bus 211. The external storage device 106 functions to store programs, data, etc., and programs, data, etc. are read into the work memory 207 via the external input / output unit 208 as necessary, and vice versa. It is stored in the device 106.

  The MPU 206 reads a program stored in the external storage device 106 during the operation of the signal processing device 105 into the work memory 207 and executes necessary signal processing. That is, the analysis processing of the image stored in the image memory 205 is performed in the work memory 207. Then, according to the processing result, the MPU 206 detects the detection target object P, captures the dynamics of the detection target object P (or tracks the intruding object), and displays the detection target object on the display unit 108 as necessary. P is displayed or dynamic capture is displayed. In addition, the warning device 212 is driven to warn as necessary.

  FIG. 3 is a diagram illustrating a schematic configuration of an embodiment of the camera 101. The camera shown in FIG. 3 is an omnidirectional camera. In FIG. 3, reference numeral 301 denotes a metal convex mirror, for example, which is attached to and supported by the support member 302. A camera unit 303 having a lens (not shown), a CCD (Charge Coupled Device), a signal processing circuit, and the like is disposed at a position facing the top of the convex mirror 301. The camera unit 303 includes a transparent cylindrical body 305, one end of the cylindrical body 305 is fixed to the support 302 on the outer periphery of the convex mirror 301, and a connecting member 306 having a light transmitting window hole 307 on the other end. Via the camera unit 303. Reference numeral 308 denotes a light emitting device constituted by an LED (Light Emitting Diode) or the like.

  In the omnidirectional camera 101 configured as described above, light from all azimuths of 360 degrees centered on the optical axis of the camera unit 303 strikes the convex mirror 301 via the cylindrical body 305 and is condensed on the lens of the camera unit 303. Thus, an image in the entire circumferential direction can be acquired. Note that a fish-eye lens can be used instead of the convex mirror 301.

  FIG. 4 is a diagram showing a schematic configuration of another embodiment of the camera 110, and shows an omnidirectional camera using a fisheye lens. In FIG. 4, reference numeral 310 denotes a fisheye lens, which is attached to and supported by the cylindrical support member 305. A camera unit 303 having a lens (not shown), an imaging device such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor), a signal processing circuit, and the like is disposed at a position facing the fisheye lens 310. In the omnidirectional camera 110 configured as described above, light from all azimuths of 360 degrees centering on the optical axis of the camera unit 303 is collected by the fisheye lens 310, and an image in the entire circumferential direction can be acquired. . Reference numerals 311 and 312 denote light emitting devices composed of LEDs (Light Emitting Diodes) or the like. Note that the camera 101 configured with the convex mirror 301 shown in FIG. 3 and the camera 110 using the fisheye lens 310 shown in FIG. 4 can be appropriately used depending on the usage. In addition, when the fisheye lens 310 shown in FIG. 4 is used, the cylindrical support member 305 does not need to be transparent. Further, in the above embodiment, the position of the imaging device and the position of the light emitting device are not physically the same, but this difference is negligible based on the size of the monitoring area. In the description, it is assumed that the position of the imaging device and the position of the light emitting device are located at substantially the same place.

  The light emitting devices 308, 311 and 312 provided in the cameras 101 and 110 are configured to be able to control light emission / non-issuance in the case of a single color via the light source control output unit 203 by the operation of the operation unit 107. In addition, when the emission colors are different, they can be emitted simultaneously. Of course, the light emission color can also be controlled by the light source control output unit 203 of the signal processing device 105. Note that the light emitting devices 308, 311 and 312 are installed at locations that can be seen from any direction so that any of the cameras 101 and 110 can capture images. Moreover, you may install the light-emitting device in the multiple places of a camera.

  FIG. 5 is a diagram showing an embodiment of the measurement jig 109 for measuring the length (or distance), the azimuth, and the horizontal. The measuring jig 109 is graduated so that the length (distance) can be understood, and the length when captured by the cameras 101 and 110 can be understood from the captured image. In this embodiment, it can be used as a measure having a length of 300 cm. The jig 109 is provided with screw holes 501 for attaching the camera 110 at regular intervals. For example, the camera 110-1 is installed at a position of 0 cm, and the camera 110-2 is fixed at a position of 300 cm. Has been made. Accordingly, the distance between the camera 110-1 and the camera 110-2 in this case is 300 cm. Further, the measuring jig 109 is provided with an azimuth meter (azimuth magnet) or azimuth sensor 502 for determining the azimuth of the measuring jig 109 and a level sensor 503 for measuring the level of the measuring jig 109. These azimuth sensor 502 and level sensor 503 are used for calibration of the X-axis, Y-axis, and Z-axis when displaying a monitoring area to be described later in orthogonal coordinates.

  Next, the operation of the present invention will be described. First, a method for obtaining the positions of the cameras 101-1, 101-2, and 101-3 arranged in the monitoring area 100 will be described. First, the measurement jig 109 is placed on the monitoring area 100, for example, on the floor. And each camera apparatus 101-1, 101-2, 101-3, 110-1, 110-2 is light-emitted sequentially, and it images with each camera and calculates | requires the position of each camera. This will be described in more detail with reference to FIGS.

  FIG. 6 is a diagram illustrating an arrangement relationship between the cameras 101-1, 101-2, 101-3, 110-1, 110-2 and the measurement jig 109.

  In FIG. 6, how to obtain θ1 to θ9 and D1 to D6 will be described. First, the lengths D1 and D2 of the two sides of the triangle formed by the camera 101-1 and the cameras 110-1 and 110-2 are expressed by the following equations.

D1 = Lsinθ3 / sin (θ2 + θ3) (1)
D2 = Lsinθ2 / sin (θ2 + θ3) (2)
Here, L represents the distance between the cameras 110-1 and 110-2, that is, the length of the measurement jig 109. Therefore, it can be understood that the angles θ2 and θ3 may be measured in order to obtain D1 and D2.

  Next, how to obtain the angles θ2 and θ3 will be described with reference to FIG. FIG. 7A is a diagram for explaining how to obtain the angle θ 1, and shows the orthogonal coordinate axes (X, Y, Z) and the unit pole surface 701. Here, the unit pole face refers to a pole face having a distance 1 from the origin, for example, a radius of 1 m. FIG. 7B is a diagram for explaining an image photographed by the camera. In FIG. 7B, reference numeral 710 represents image data when captured by an omnidirectional camera. In the present embodiment, the image data 710 is rectangular as in a general camera. Then, in imaging through a convex mirror or lens of an omnidirectional camera, effective image information has a circular shape as indicated by 711. This circular image data 711 is mapped onto the unit pole surface 701 in FIG. 7A using a conversion table. Since the mapping from the circular image data 711 onto the unit spherical surface 701 differs depending on the characteristics of the convex mirror or lens of the omnidirectional camera, it is measured in advance with an actual machine, a conversion table is created, and stored in the external recording device 106. Keep it.

  In this state, the operator operates the operation unit 107 to send a light emission command from the light source control output unit 203 to the light emitting device of the camera 110-1, for example, the light emitting device 311 to emit light. The light emitted from the light emitting device 311 of the camera 110-1 is imaged by the camera 101-1, and the intersection M1 (Xm1, Ym1, Zm1) of the line segment connecting the camera 101-1 and the camera 110-1 and the unit polar surface 701 at that time is captured. ) Is calculated by the MPU 206. Next, the light emitting device 311 of the camera 110-2 is caused to emit light in the same manner as described above, and the intersection M2 (Xm2, Ym2, Zm2) between the line segment connecting the camera 101-1 and the camera 110-2 and the unit pole surface 701 at that time. ) Is calculated by the MPU 206. As a result, the angle θ1 formed by the point M1, the camera 101-1, and the point M2 can be obtained by calculating the inner product of the following equation (3).

cosθ1 = Xm1Xm2 + Ym1Ym2 + Zm1Zm2 (3)
Similarly, the angle θ2 causes the light emitting device 308 of the camera 101-1 to emit light, then causes the light emitting device 311 of the camera 110-2 to emit light, the camera 110-1 takes an image, and the origin is 110-1. When the intersection point on the surface 701 is obtained by calculation with the MPU 206, the angle θ2 is obtained in the same manner as the angle θ1 described above. Similarly, the light emitting devices of the camera 101-1 and the camera 110-1 are caused to emit light, the camera 110-2 captures an image, and θ3 can also be obtained using the unit pole surface 701 with the origin as 110-2. θ3 can also be obtained by the following equation.

θ3 = 180− (θ1 + θ2) (4)
By the method as described above, the angles (distances) D1 of the angles θ1, θ2, θ3 of the triangles constituted by the omnidirectional cameras 101-1, 110-1, 110-2 and the above expressions (1) and (2), D2 is obtained. Similarly, angles [theta] 4, [theta] 5, [theta] 6 and lengths (distances) D3, D4 of the triangle formed by the omnidirectional cameras 101-2, 110-1, 110-2 shown in FIG. 6 are obtained. Similarly, the angles θ7, θ8, θ9 and the lengths (distances) D5, D6 of the triangle constituted by the omnidirectional cameras 101-3, 110-1, 110-2 are also obtained.

  Next, how to determine the positions of the omnidirectional cameras 101-1, 101-2, and 101-3 installed in the monitoring area 100 will be described. First, orthogonal coordinates (X axis, Y axis, Z axis) for specifying the positions of the omnidirectional cameras 101-1, 101-2, and 101-3 are determined as shown in FIG. In FIG. 8, the origin of orthogonal coordinates (X, Y, Z) is the position of the omnidirectional camera 110-1 provided in the measurement jig 109, and the plane including the measurement jig 109 is defined as the X-axis-Y-axis plane. . Therefore, the direction perpendicular to the X-axis / Y-axis plane is the Z-axis. In setting the coordinates (X, Y, Z), for example, the azimuth sensor 502 and the level sensor 503 provided in the measurement jig 109 described above are used to make the X axis coincide with the north-south direction. Orthogonal coordinates (X, Y, Z) are determined so that the Y-axis plane is horizontal. These settings are all calculated by the MPU 206 by setting input from the operation unit 107 and stored in the external storage device 106 in advance. In the above embodiment, the origin of the Cartesian coordinates (X, Y, Z) has been described as the position of the omnidirectional camera 110-1. However, the present invention is not limited to this and can be set as appropriate. Absent.

Now, when set as described above, the coordinates of each omnidirectional camera shown in FIG. 6 can be easily obtained as described below. In FIG. 8, points M1 (Xm1, Ym1, Zm1), M2 (Xm2, Ym2, Zm2), M3 (Xm3, Ym3, Zm3), M4 (1, 0, 0) are respectively omnidirectional cameras 101-1, The point which light-emitted the light-emitting device of 101-2, 101-3, and 110-2 and imaged the light emission with the camera 110-1 on the unit pole surface 801 is shown. Since the point M4 is a point on the X axis, the X, Y, and Z coordinates are (1, 0, 0). Here, the distances D1, D3, and D5 are obtained by the method described above, and the distance L is known in advance. Points M1, M2, M3, and M4 are points on the unit pole surface 801, and the distance from the origin is 1, which can be regarded as a unit vector. Therefore, the coordinates of each omnidirectional camera can be obtained by unit vector × distance. That is,
Position of camera 101-1 (X1, Y1, Z1) = (D1 × Xm1, D1 × Ym1, D1 × Zm1) (5)
Camera 101-2 position (X2, Y2, Z2) = (D3 × Xm2, D3 × Ym2, D3 × Zm2) (6)
Position of camera 101-3 (X3, Y3, Z3) = (D5 x Xm3, D5 x Ym3, D5 x Zm3) (7)
Camera 110-2 position (X4, Y4, Z4) = (L, 0, 0) (8)
The position of the omnidirectional camera located in the monitoring area can be obtained by the above method. Then, the position (X1, Y1, Z1) of the omnidirectional camera 101-1, the position (X2, Y2, Z2) of the omnidirectional camera 101-2, and the position of the omnidirectional camera 101-3 (as described above) X3, Y3, Z3) are stored in the external storage device 106.

  As described above, in the present invention, the omnidirectional cameras 101-1, 101-2, and 101-3 are installed in the monitoring area where the detection target object P is detected, and the omnidirectional cameras 110-1 and 110- A measuring jig 109 with 2 is placed in this monitoring area. Next, the light emitting devices provided in the respective omnidirectional cameras are caused to emit light sequentially, and the emitted light is sequentially imaged by the omnidirectional cameras 101 and 110, whereby the omnidirectional cameras 101-1 and 101- arranged in the monitoring area. The positions 2 and 101-3 (substantially the same position as the position of the light emitting device) can be obtained. In the above embodiment, the case where the cameras 101 and 110 emit light sequentially has been described, but the light emitting devices of the cameras 101 and 110 may be configured to emit different colors. For example, the light emitting device of the camera 101-1 emits red light, and the light emitting device of the camera 101-2 emits orange light. The light emitting device of the camera 101-3 emits green light, the light emitting device of the camera 110-1 emits light blue, the light emitting device of the camera 110-2 emits blue light, and so on. In such a case, even if each light emitting device emits light at the same timing, the MPU 206 can distinguish the light emitting position because the omnidirectional camera can capture images with different colors for each light emitting device. Therefore, the position of the omnidirectional camera can be obtained by this method. In addition, when the light emitting device emits light at the same timing as described above, the position of the omnidirectional camera can be obtained in a short time, and the moving object detection system can be operated quickly.

  When the coordinate (X, Y, Z) positions of the omnidirectional cameras 101-1, 101-2, and 101-3 are obtained as described above, the measurement jig 109 and the omnidirectional camera 110 attached to the measurement jig are used. -1, 110-2 may be removed from the monitoring area. That is, since the position coordinates in the monitoring area 100 of the omnidirectional cameras 101-1, 101-2, and 101-3 are clear, the detection target object P that enters the monitoring area can be detected. In the above-described embodiment, the case where there are three omnidirectional cameras 101 has been described. However, three or more omnidirectional cameras 101 may be used.

  In addition, in the state configured as described above, when there is a need to move one omnidirectional camera, for example, the omnidirectional camera 101-1, due to some reason, for example, redesign of an exhibit in the monitoring area 100 Even if the omnidirectional camera 101-1 is moved, the other two omnidirectional cameras 101-2 and 101-3 hold the previous positions. The position of the moved omnidirectional camera 101-1 can be obtained by causing the light emitting devices 308 of 101-2 and 101-3 to emit light and capturing images with the omnidirectional camera. That is, since the distance between the omnidirectional cameras 101-2 and 101-3 is not changed, it can be obtained in the same manner as described above. Thus, once the position of the omnidirectional camera 101 located in the monitoring area 100 is obtained, the position can be easily changed.

  Further, in the above embodiment, when one omnidirectional camera is moved, the light emitting device 308 of the omnidirectional camera 101 is caused to emit light again to obtain the position of the moved omnidirectional camera 101. The light emitting device 308 of the omnidirectional camera 101 emits light periodically, for example, at a rate of once every 2 hours, and an image captured by the omnidirectional camera 101 is compared with a past image by a conventionally known background subtraction method or the like. Then, when the moving distance of the omnidirectional camera 101 exceeds a certain distance, for example, a threshold of 30 cm, the position registered in the external storage device 106 as the possibility that the omnidirectional camera 101 has moved is corrected. You can also re-register.

  Next, another embodiment of the present invention will be described with reference to FIG. FIG. 9 is a diagram illustrating intrusion of the detection target object P and flow line capture. In FIG. 9, the positions of the omnidirectional cameras are indicated by 101-1, 101-2, and 101-3 in the coordinates (X, Y, Z) indicated by the monitoring area 100. The positions 101-1, 101-2, and 101-3 of these omnidirectional cameras are obtained in advance by measurement and calculation by the method described in the above embodiment, and are stored in the external storage device 106. And each omnidirectional camera 101-1, 101-2, and 101-3 is always imaging the inside of the monitoring area 100 (it is in the monitoring state). When the detection target object P enters the monitoring area 100 in this state, each of the omnidirectional cameras 101-1, 101-2, and 101-3 images the detection target object P and stores each captured image in the image memory 205. To do. Note that the detection target object P is a person, a car, an animal, or the like. For convenience of explanation, for example, the detection target object P is a person and detects the position of the center of gravity of the person. It is shown as a target object P. In addition, the recognition of the detection target object P includes a method called a difference method or a template matching method described in the prior art, and details thereof are already well known, and detailed description thereof is omitted. .

  The detection target object P stored in the image memory 205 is the position of each omnidirectional camera 101-1, 101-2, and 101-3 in the monitoring area 100 already stored in the external storage device 106 by the MPU 206. Then, the position of the detection target object P is calculated, and the calculated position of the detection target object P is stored in the external storage device 106. Here, a method for calculating the position of the detection target object P will be briefly described. First, the detection target object P is imaged by the omnidirectional camera 101-1. Since the position coordinates of the omnidirectional camera 101-1 and the omnidirectional camera 101-2 are already registered, the position and length (distance) of the line segment D7 are clear. Therefore, the angle θ10 formed by the line segment D7 and the line segment D8 can be obtained by calculation of the MPU 206. Similarly, when the detection target object P is imaged by the omnidirectional camera 101-2, the angle θ11 formed by the line segment D7 and the line segment D9 can also be obtained by the calculation of the MPU 206. Therefore, since the distance of the line segment D7 is known, the distance between the line segment D8 and the line segment D9 can be obtained by the calculation of the MPU 206. As a result, the position (Xp, Yp, Zp) of the detection target object P within the monitoring area 100 of the detection target object P can be obtained by calculation of the MPU 206.

  When the position (Xp, Yp, Zp) of the detection target object P is detected in this manner, the moving object detection system starts monitoring the detection target object P, and the detection target object P is, for example, an entry prohibition target. Assuming that the vehicle has entered the area, the alarm output unit 210 is driven, the alarm 211 is sounded, an alarm with an intruder is sounded, and at the same time the position of the detection target object P is displayed on the display unit 108.

  Further, each of the omnidirectional cameras 101-1, 101-2, and 101-3 continues to image the detection target object P, and sets the position of the detection target object P every predetermined time, for example, every 5 minutes. It is calculated by the same method, stored in the external storage device 106, and displayed on the display unit 108. In this way, the operator knows the movement positions P1, P2, P3,... Of the detection target object P every 5 minutes. That is, the intrusion position of the detection target object P and the movement (flow line capture) of the detection target object P can be monitored. The moving object detection system described above is not limited to the monitoring camera system, and can be applied in various ways. This will be described below.

  FIG. 10 is a view for explaining still another embodiment of the present invention. In FIG. 10, reference numeral 1001 denotes a plan view of the first floor of the department store, for example. 1002 is an entrance, 1003 is an exit, and A, B, C,... N are exhibition tables displaying various products. Reference numerals 101-1, 101-2, and 101-3 denote omnidirectional cameras installed on the ceiling of the first floor of the department store. The floor plan of the first floor of the department store, the display tables A, B, C,... N, and the positions of various products are stored in the external storage device 106 in advance. Of course, when the omnidirectional cameras 101-1, 101-2, and 101-3 are installed, the measurement jig 109 equipped with the omnidirectional cameras 110-1 and 110-2 is first placed on the floor in the manner described above. It is installed, and the positions in the monitoring area 100 of the omnidirectional cameras 101-1, 101-2, and 101-3 are measured and calculated and stored in the external storage device 106. In this state, the department store starts imaging the shopper as soon as the store opens. In FIG. 10, the flow line capture of two people of the shoppers N1 and N2 is shown. That is, the position of the shopper N1 every predetermined time is indicated by a circle, and the position of the shopper N2 every predetermined time is indicated by a cross. In this way, the flow line captures of the two shoppers N1 and N2 are recorded, and the administrator analyzes this to analyze what the shopper is interested in. The arrangement of the product can be examined.

  In this practical example, the light emission amount of each camera is described as being constant. However, if the light emission amount of the light emitting device is adjusted by the operation unit 107 and each camera captures an image according to the light emission amount, the measurement time is Although it becomes longer, it is possible to increase the measurement accuracy. Furthermore, in this practical example, the omnidirectional camera has been described as an example of a convex mirror or a fisheye lens. However, a camera capable of calculating an angle with another camera, for example, a swivel camera, or a camera that can capture images for each direction. A camera such as an eye camera may be used. Moreover, although the said Example demonstrated the method to measure length with the measuring jig provided with the camera, for example, when the distance between any two cameras 101 and the installation height can be measured, There is no need to provide a measuring jig with a camera in the monitoring area. For example, when two cameras are provided on the ceilings at both ends of the floor of the department store, the distance and the height of both ends of the floor are known in advance, so that a measurement jig equipped with cameras is not necessary.

  Although the present invention has been described in detail above, the present invention is not limited to the embodiment of the moving object detection system described herein, and can be widely applied to other moving object detection systems. Needless to say.

It is a figure which shows schematic structure of one Example of this invention. It is a figure which shows the block diagram of schematic structure of the signal processing apparatus of one Example of this invention shown in FIG. It is a figure which shows one Example of the omnidirectional camera used by one Example of this invention. It is a figure which shows one Example of the other omnidirectional camera used by one Example of this invention. It is a figure which shows one Example of the measuring jig used in one Example of this invention. It is a figure for demonstrating the method to obtain | require the position of the camera used by this invention. It is a figure for demonstrating the method of calculating | requiring the angle used by this invention. It is a figure for demonstrating the method of calculating | requiring the position of the camera of this invention on a unit pole surface. It is a figure for demonstrating another Example of this invention. It is a figure for demonstrating another Example of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100: Surveillance area 101, 110: Omni-directional camera, 102: Camera support column, 103: Transmission line, 104: Transmission path, 105: Signal processing device, 106: External storage device, 107: Operation unit, 108: Display unit 109: Measurement jig 201: Input / output terminal 202: Image input unit 203: Light source control output unit 204: Operation input unit 205: Image memory 206: MPU 207: Work memory 208: External input / output 209: Image output unit 210: Alarm output unit 211: Alarm device 301: Convex mirror 302: Support member 303: Camera unit 305: Tube body 306: Connecting member 307: Window hole 308 311, 312: light emitting device, 310: fisheye lens, 501: screw hole, 502: orientation sensor, 503: level sensor, 701 and 801: polar surface, 10 1: a plan view of a floor, 1002: Inlet, 1003: exit.

Claims (4)

  At least the first, second, and third cameras arranged in the monitoring area, the first, second, and third light emitting devices provided in proximity to each of the cameras, and the output of the camera are input. The signal processing device includes a calculation unit and a storage unit. The storage unit stores coordinates representing the space of the monitoring area, and emits light from each of the light emitting devices. The image is captured by each of the cameras, and the arithmetic unit is configured to display the captured light emission position information of the first, second, and third light emitting devices and coordinates indicating the space of the monitoring area stored in the storage unit. The moving object detection system characterized in that the position on the coordinates of the first, second and third light emitting devices is calculated based on the above.   The moving object detection system according to claim 1, wherein the light emitting device is caused to emit light again at a predetermined time, is imaged by the camera, the position of the light emitting device is recalculated, and the light emitting device already calculated is A moving object detection system characterized by correcting a position on coordinates.   2. The moving object detection system according to claim 1, wherein the camera in the monitoring area has a predetermined length including the fourth and fifth cameras and the fourth and fifth light emitting devices different from the camera. A jig is disposed, and the first to fifth cameras emit light emitted from the first to fifth light emitting devices. The arithmetic unit is arranged on the coordinates of the first, second, and third light emitting devices. A moving object detection system characterized in that the position of the moving object is calculated.   2. The moving object detection system according to claim 1, wherein the detection target object that has entered the monitoring area is imaged by the first, second, and third cameras, and the arithmetic unit is configured to capture the detected detection target object. A moving object detection system that calculates the position of the detection target object on the coordinates based on the image pickup position information and coordinates representing the space of the monitoring area stored in the storage unit.

Images