JP2007208659A - Cooperation camera system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To continuously trace a moving body even if a plurality of moving bodies simultaneously appear. <P>SOLUTION: A cooperation camera system includes a trace importance calculation unit which calculates a trace importance degree iaB on which a camera 110 traces a moving body B in a trace region ARb. Moreover, it includes a taking over importance calculation unit which calculates a taking over importance degree iAA which a camera 110 succeeds and takes over a moving body A which a camera 100 is tracing. Further, when the camera 110 is tracing the moving body B in the tracing region ARb, the system includes a comparison unit which compares the trace importance degree for the moving body B and the taking over importance degree is tracing for the moving body to A which the camera 100. At that time, after the comparison unit detects that a degree of the taking over importance degree iAA becomes higher than a degree of the trace importance degree iaB, the system includes a switching unit which switches the moving body which the camera 110 traces to the moving body A. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a linked camera system that links a plurality of surveillance cameras and continuously photographs a plurality of moving objects.

  Conventionally, a monitoring system that monitors a wide range by distributing a plurality of monitoring cameras has been put into practical use. Many of these surveillance systems project multiple images taken by multiple wide-angle fixed surveillance cameras onto a monitor at the surveillance center, and the monitor monitors the monitor, thereby intruding suspicious individuals, incidents and accidents. It is possible to detect abnormal conditions such as.

  For this reason, in the conventional monitoring system, as the number of monitoring cameras increases, the burden on the monitoring person becomes heavier, and the discovery of an abnormal situation may be delayed or overlooked. In addition, simply shooting with a wide-angle fixed surveillance camera cannot zoom in on a specific location or object, so that the monitor cannot accurately grasp the situation.

  To improve this situation, research and development of surveillance cameras that automatically recognize moving objects and systems that link multiple surveillance cameras to each other have begun.

  Patent Document 1 and Patent Document 2 disclose an intruder automatic tracking method and a monitoring system for continuously capturing a moving body by linking a plurality of monitoring cameras.

  As shown in FIG. 5, the plurality of monitoring cameras in these prior arts first perform automatic detection of moving objects and automatic extraction of feature amounts, respectively. The plurality of monitoring cameras 1 and 2 monitor the moving body M while tracking the moving body M when the moving body M moves from the visual field that is the tracking area of one monitoring camera 1 to the visual field of the other monitoring camera 2. The feature information of the moving body M is transmitted from the camera 1 to the other monitoring camera 2. The other monitoring camera 2 can find the moving object M having the received feature information. As a result, the plurality of monitoring cameras 1 and 2 can continuously photograph the moving object M between the monitoring cameras.

  Patent Document 3 discloses an invention of a moving object measuring apparatus that accurately measures three-dimensional information of a moving object using three or more surveillance cameras at the same time. As shown in FIG. 6, this measuring device has three or more monitoring cameras 11 to 16 installed at positions suitable for measuring three-dimensional information of the moving body as the moving body moves. is doing. The measuring device dynamically selects the three or more installed monitoring cameras 11 to 16 so that the situation where the moving body M cannot be seen from the monitoring camera used for measurement does not occur.

  In addition, the measuring device disclosed in Patent Document 3 simultaneously detects a plurality of moving bodies with a fixed viewing camera (wide-angle monitoring camera) having a wide viewing angle, and three or more pans / tilts for each of the detected moving bodies. A surveillance camera with a zoom function (PTZ surveillance camera) is dynamically assigned. As a result, the measuring device can measure the three-dimensional information of a plurality of moving bodies. Furthermore, the measuring device can prioritize a plurality of moving objects, and assign a monitoring camera to a moving object with a higher priority, and can cope with the case where the monitoring camera assignments compete with the plurality of moving objects. It has become.

JP 2002-290962 A JP 2003-324720 A Japanese Patent Application Laid-Open No. 07-306012

  However, the intruder automatic tracking method and the monitoring system described in Patent Document 1 and Patent Document 2 do not take into account that a plurality of moving objects may appear simultaneously in the field of view of a plurality of linked monitoring cameras. For this reason, in the conventional intruder automatic tracking method and monitoring system, the control method of each monitoring camera and the cooperation method between the plurality of monitoring cameras when a plurality of moving objects appear simultaneously are unknown.

  In addition, the measuring device described in Patent Document 3 is suitable for a situation where a large number of surveillance cameras can be arranged around a place with good visibility such as a soccer field, and basically one surveillance camera in one range. It is not suitable for monitoring systems that monitor with This is because three or more surveillance cameras must always shoot a single moving body at the same time, so the number of surveillance cameras is three times greater than when one range is monitored by one surveillance camera. This is a significant cost increase.

  Furthermore, when a plurality of moving objects appear simultaneously, the measuring device described in Patent Document 3 measures only one moving object having the highest priority when there are only three monitoring cameras that can be assigned to these moving objects. (Or tracking / shadowing) is not possible. For this reason, there is a problem that all other moving objects are ignored.

  Therefore, even if any of the above-described prior arts is used, if a plurality of moving objects appear simultaneously within a range in which a plurality of surveillance cameras shoot together, only one moving object can be tracked and photographed at most. There was a problem that it was not possible.

  The present invention provides a linked camera system capable of simultaneously tracking and shooting as many moving objects as possible even when a plurality of moving objects appear simultaneously when a plurality of monitoring objects are linked to continuously shoot a moving object. It is to provide.

  The linked camera system of the present invention is provided in the one monitoring camera and the other monitoring camera in the linked camera system that captures and tracks the tracking target moving body from one monitoring camera to the other monitoring camera. Tracking importance setting means for obtaining tracking importance for tracking each moving camera in the tracking area of each monitoring camera, tracking importance setting means for obtaining the tracking importance, and being tracked by the one monitoring camera The takeover importance setting means for obtaining the takeover importance for the other monitoring camera to take over and track the moving body, and the other monitoring camera is provided with the other monitoring camera. When tracking a moving object in the tracking area, the tracking importance for the moving object and the moving object being tracked by the one surveillance camera are displayed. The comparison means for comparing the takeover importance level and the other monitoring camera are provided, and the comparison means has the degree of the takeover importance level higher than the tracking importance level. And switching means for switching the moving body tracked by the other monitoring camera to the moving body tracked by the one monitoring camera after being detected.

  In the cooperative camera system of the present invention, when a plurality of moving objects appear in the tracking area of the surveillance camera at the same time, each surveillance camera automatically determines the role of each person so that the role sharing between the surveillance cameras is optimized. It has become. For this reason, the cooperative camera system of the present invention can track and photograph many moving objects more effectively.

  Hereinafter, a cooperative camera system according to an embodiment of the present invention will be described with reference to FIGS. 1 to 4.

  FIG. 1 is a block diagram of the cooperative camera system 90.

  Assume that the cooperative camera system 90 includes a plurality of monitoring cameras (not shown) in addition to the monitoring camera 100 and the monitoring camera 110. These surveillance cameras have the same configuration. Each monitoring camera transmits image data and / or composite information to the monitoring device 20 via the network 10 as necessary. Here, the composite information includes at least tracking importance level information, takeover importance level information, feature information, and the like regarding the mobile objects A and B (see FIG. 4). In addition, the monitoring cameras transmit / receive composite information to / from each other via the network 10 as necessary.

  In FIG. 1, broken line arrows indicate the flow of image data (including encoded image data). Solid arrows indicate the flow of other data (composite information, various control information, etc.). The network 10 may be any communication network, communication line, or communication technology, regardless of whether it is wired / wireless or public / dedicated. In the network 10, the use of an analog dedicated line, a LAN line, a mobile phone line, etc. is increasing, and the Internet may be used in a broader concept.

  Hereinafter, the configuration of the monitoring camera denoted by reference numeral 100 will be described on behalf of all the monitoring cameras. In FIG. 1, in the surveillance camera denoted by reference numeral 100 and the surveillance camera denoted by reference numeral 110, components having the same name are configured to perform the same action. For this reason, description of the surveillance camera denoted by reference numeral 110 is omitted. Also in the composite information processing unit shown in FIG. 2, components having the same name perform the same action.

  The image photographing unit 103 is a part that photographs a moving body, and includes a pan / tilt / zoom mechanism (not shown). The image photographing unit 103 can change the photographing direction and the viewing angle when the image photographing unit 103 performs pan / tilt / zoom in accordance with an instruction from the photographing parameter control unit 102. Image data captured by the image capturing unit 103 is transmitted to the composite information processing unit 105 and the image encoding unit 106.

  The imaging parameter control unit 102 receives the prediction result of the moving direction and the moving amount of the moving body on the image from the composite information processing unit 105 as will be described later. The shooting parameter control unit 102 adjusts the pan angle and the tilt angle of the image shooting unit 103 so that the moving body is shot near the center of the monitoring camera field of view based on the prediction result. When the user desires (according to user settings), the imaging parameter control unit 102 is notified of the size of the moving object on the image from the composite information processing unit 105. The shooting parameter control unit 102 controls the zoom rate of the image shooting unit 103 so that this size approaches a predetermined size determined by the user in advance (or falls within a predetermined upper limit size and lower limit size). Part. The shooting parameter control unit 102 is a control unit that transmits the pan angle, tilt angle, and zoom rate of the image shooting unit 103 to the composite information processing unit 105.

  The installation information acquisition unit 101 acquires the installation position and installation direction (the direction of the monitoring camera base) of the monitoring camera 100 and other adjacent monitoring cameras as three-dimensional absolute coordinates and direction angles. The origin of absolute coordinates and the XYZ axis directions may be arbitrarily determined, but all surveillance cameras must use the same coordinate system. Any conventional technique can be used for the installation information acquisition method. In the simplest installation information acquisition method, the user measures the installation information (installation position and installation direction) of the monitoring camera 100 and other adjacent monitoring cameras in advance using an existing measurement device, and stores the information. This is a method of inputting to 104. In this case, the installation information acquisition unit 101 is not necessary. As another installation information acquisition method, after acquiring own installation information (installation position and installation direction) using an automatic measurement device such as a GPS measurement device, communication between adjacent monitoring cameras is performed to install each other. There are ways to exchange information. The installation information acquired by the installation information acquisition unit 101 or stored in advance in the storage unit 104 is used by the composite information processing unit 105.

  Next, the composite information processing unit 105 performs various processes as listed below.

  The composite information processing unit 105 has a function of automatically detecting a moving object (moving object detection function) by processing image data in time series. Conventional methods such as a background difference method, an inter-frame difference method, and an optical flow method can be used as a moving object detection method.

  The composite information processing unit 105 also has a function (feature information extraction function) for extracting feature information on the appearance of the moving object by extracting the moving object portion from the image data and extracting predetermined information therefrom. . The predetermined information includes, for example, a shape, a size, a luminance distribution, a color distribution, a feature point distribution, a bitmap pattern, etc., or some combination thereof.

  The composite information processing unit 105 has a function (tracking importance calculation function) for calculating a tracking importance that quantitatively represents the importance of tracking a certain moving object. That is, the composite information processing unit 105 calculates the tracking importance for tracking the moving objects A and B in the shootable range ARa (ARb, ARc) that is the tracking area of the monitoring camera 100 (110, 120) shown in FIG. A tracking importance calculation unit 105a (see FIG. 2). The tracking importance calculation unit 105a (115a) corresponds to tracking importance setting means. A specific example of the tracking importance information will be described later.

  The composite information processing unit 105 also has a function of creating composite information by combining tracking importance information and feature information of a moving object and storing the composite information in the storage unit 104.

  The composite information processing unit 105 calculates the representative position (for example, the center position) of the moving object on the image, and obtains the trajectory of the representative position of the moving object for a predetermined number of frames. It also has a function to predict the future direction and amount of movement. Then, the composite information processing unit 105 transmits this prediction result to the imaging parameter control unit 102. If the user desires (according to user settings), the size of the moving object on the image is also calculated and transmitted to the imaging parameter control unit 102.

  The composite information processing unit 105 moves based on the pan angle and tilt angle of the image shooting unit 103 transmitted from the shooting parameter control unit 102, and the prediction result of the future moving direction and moving amount on the moving object image. It is determined whether the body is moving away from the visual field of the monitoring camera 100.

  If the composite information processing unit 105 determines that the moving body is moving away, the installation information acquisition unit 101 acquires the advancement of the moving body using the installation information stored in the storage unit 104 in advance. Identify other surveillance cameras in the direction. Then, the composite information processing unit 105 transmits the composite information stored in the storage unit 104 to the identified other monitoring camera via the communication unit 107 and the network 10.

  The composite information processing unit 105 includes a function (composite information comparison function) for comparing composite information (information regarding takeover importance) received from another monitoring camera with composite information (information regarding tracking importance) created by itself. ing. Then, the composite information processing unit 105 determines a moving body to be tracked by the self-monitoring camera based on the comparison result.

  For this reason, as shown in FIG. 2, the composite information processing unit 105 of the upstream monitoring camera 100 takes over the moving object A being tracked by the monitoring camera 100 by the downstream monitoring camera 110 and takes over. A takeover importance degree calculation unit 105b for obtaining the degree is provided. The takeover importance calculating unit 105b is a takeover importance setting means. Furthermore, the downstream composite information processing unit 115 tracks the tracking importance of the moving object B and the monitoring camera 100 during tracking of the moving object B in the imageable area ARb in which the monitoring camera 110 is in the tracking area. The comparison part 115c which compares the taking over importance of the moving body A currently provided is provided. The comparison unit 115c is a comparison unit. Further, when the comparison unit 115c detects that the importance of the takeover importance for the moving object A is higher than the tracking importance for the moving object B, the moving object to be tracked by the monitoring camera 110 on the downstream side is changed from B to A. There is provided a switching portion 115d for switching to. The switching unit 115d is switching means.

  The composite information processing unit corresponding to each camera includes the tracking importance calculation unit, the takeover importance calculation unit, the comparison unit, and the switching unit described above.

  Here, three specific examples of the tracking importance level information are given below (iα, iβ, and iγ). Furthermore, the meaning of using these pieces of information and an example of assigning numerical values when these pieces of information are digitized in order to handle them uniformly are shown.

  (Iα) is information indicating whether or not the other monitoring camera is simultaneously tracking the moving object being tracked by the own monitoring camera.

  Meaning: If another surveillance camera is tracking the same mobile object as the self-monitoring camera, the self-monitoring camera can follow another mobile object.

  Numerical example: 0 or 10. 0 if other surveillance cameras are tracking the same moving object as the surveillance camera, and 10 if only the surveillance camera is tracking. If the operation of the other monitoring camera is unknown, it is assumed that only the own monitoring camera is tracking and is set to 10.

  (Iβ) is information indicating the degree of possibility that the moving body being tracked by the self-monitoring camera is out of the shootable range of the self-monitoring camera.

  Meaning: As the moving body moves away from the photographing range of the self-monitoring camera, the importance for the self-monitoring camera decreases.

  Numerical example: 0 to 3. The importance for the self-monitoring camera 100 (110, 120) is 3 when the moving object is in the vicinity of the center of the photographing range ARa (ARb, ARc) of the self-monitoring camera as shown in FIG. 2 and 1 as the distance from the camera increases, and 0 when the camera is out of the shootable range.

  (Iγ) is information indicating the degree to which the feature information of the moving object being tracked by the self-monitoring camera matches the attention information stored in advance.

  Advantage: By storing attention information for specifying the type of mobile object in advance, only a specific type of mobile object can be tracked. Thus, for example, only humans can be tracked and small animals can be ignored.

  Numerical example: 0 to 3. The correlation value between the feature information of the moving object and the attention information is calculated, and the obtained correlation value is requantized to 0 to 3.

  Next, three specific examples of takeover importance level information are given (iα, iβ, iδ). iα and iβ are the same as iα and iβ described above.

  (Iδ) is information indicating the degree of takeover importance that the downstream monitoring camera 110 takes over and tracks the moving object A being tracked by the upstream monitoring camera 100.

  Meaning: Increased importance for other surveillance cameras in the direction of travel of the moving object.

  Numerical example: 0 to 3. As shown in FIG. 3B, the takeover importance for the self-monitoring camera 110 is 0 when the moving body is near the center of the photographing possible range ARa of the upstream monitoring camera 100, and is 1 as the distance from the center increases. 2. Then, when the moving body enters within the shootable range ARb of the self-monitoring camera 110, it is set to 3. In other words, the tracking importance (iβ) has an inverse relationship, iδ = 3-iβ.

  The tracking importance level information and the takeover importance level information need not be quantified as long as they can be compared. The above numerical value assignment example is described only to help understanding of the embodiment. When the cooperative camera system is actually used, the numerical value assignment method may be changed in consideration of the installation location and operation status of each surveillance camera.

  The image encoding unit 106 performs compression, encryption, error correction encoding, and the like of image data according to a predetermined procedure, and transmits the encoded image data to the communication unit 107. However, the image encoding unit 106 operates only when image data needs to be transmitted to the monitoring device 20. For this reason, it is necessary to determine in advance a control procedure for controlling whether or not to transmit image data to the monitoring device 20. As an example of the control procedure, there is one that always transmits image data to the monitoring device 20 while the composite information processing unit 105 detects a moving object. As another example, there is a system designed so that the user of the monitoring device 20 can control the start and stop of image data transmission. When it is determined that the monitoring device 20 does not display image data (only the position information of the moving body can be displayed), the image encoding unit 106 is not necessary.

  The communication unit 107 transmits and receives various data by a method (data format, modulation method, communication protocol, etc.) suitable for the network 10.

  On the other hand, the monitoring device 20 has the following configuration.

  The communication unit 21 transmits and receives various data by a method suitable for the network 10. The image decoding unit 22 decodes the received encoded image data, and then transmits the image data to the display unit 24 as a monitor. The control unit 23 includes a user interface (not shown), and performs control such as changing the display method of the display unit 24 according to an instruction from the user or transmitting control information to an arbitrary monitoring camera. The display unit 24 displays the composite information received from the communication unit 21 and the image data received from the image decoding unit 22 to the user by a predetermined display method.

  Based on FIG. 4, an example of cooperative operation between surveillance cameras will be described.

  4A to 4D are diagrams showing the operating state of the cooperative camera system 90 at times t1 to t4, respectively. It is assumed that time elapses in the order of (a) to (d).

  The monitoring cameras 100, 110, and 120 are installed side by side at intervals where the left and right ends of the shootable ranges ARa, ARb, ARc of each monitoring camera slightly overlap. The photographing possible ranges ARa, ARb, ARc of the respective monitoring cameras are semicircular ranges drawn by dotted lines in front of the monitoring camera (upward in the figure). The moving bodies A and B move substantially from left to right at times t1 to t4.

  Hereinafter, description will be made using the above-described specific examples (iα), (iβ), and (iγ) as tracking importance information of the moving object and takeover importance information (iδ).

  The sum of the numerical values obtained from (iα), (iβ), and (iγ) is set as the tracking importance (ia) ((ib) (ic). However, for (iγ), (iγ) = 3 is assigned to both of the moving bodies A and B by preliminarily storing human appearance characteristics (for example, shape and size) as attention information. Shall.

  Further, the sum of the numerical values obtained from (iα), (iγ), and (iδ) is set as the tracking importance (iA) ((iB) (iC).

  4A, the monitoring camera 100 is tracking the moving object A, and the monitoring camera 110 is tracking the moving object B. The monitoring camera 100 determines that the moving object A is approaching the direction of the monitoring camera 110 by the function of the composite information processing unit 105 described above. Therefore, the monitoring camera 100 transmits the takeover importance of the moving object A to the monitoring camera 110 as a tracking request message (tracking request 1-1 in the drawing). This message includes composite information created by combining the takeover importance (iAA) calculated by the following procedure and the feature information of the moving object A extracted using any conventional technique.

  The tracking importance (iaA) for the moving object A being tracked by the monitoring camera 100 is calculated by the composite information processing unit 105 as follows.

  Since only the surveillance camera 100 is tracking the moving object A, (iα) = 10. Since the moving object A is in the vicinity of the center of the shootable range of the monitoring camera 100, (iβ) = 3. Since the moving object A matches the attention information stored in advance, (iγ) = 3. The sum of these is (iaA) = 16.

  The takeover importance (iAA) of the downstream monitoring camera 110 with respect to the moving object A being tracked by the upstream monitoring camera 100 is calculated by the composite information processing unit 105 as follows.

  Since only the surveillance camera 100 is tracking the moving object A, (iα) = 10. Since the moving object A matches the attention information stored in advance, (iγ) = 3. Since the moving object A is near the center of the photographing range of the monitoring camera 100, (iδ) = 0. The sum of these is (iAA) = 13.

  The tracking importance (iaB) for the moving body B being tracked by the monitoring camera 110 is calculated by the composite information processing unit 115 as follows.

  Since only the surveillance camera 110 is tracking the moving object B, (iα) = 10. Since the moving body B is in the vicinity of the center of the shootable range of the monitoring camera 100, (iβ) = 3. Since the moving object B matches the attention information stored in advance, (iγ) = 3. The sum of these is (iaB) = 16.

  Accordingly, the composite information processing unit 105 of the upstream monitoring camera 100 sends a tracking request message to the composite information processing unit 115 of the monitoring camera 110 on the downstream side (taken request 1-1 in the figure). Send as.

  The comparison unit 115c of the downstream composite information processing unit 115 determines the takeover importance (iAA = 13) from the upstream composite information processing unit 105 and the tracking importance of the downstream monitoring camera 110 with respect to the moving object B. Compare. Since (iAA = 13) <(iaB = 16), the downstream composite information processing unit 115 determines that it is more important to track the mobile object B that is currently being tracked. Then, the downstream composite information processing unit 115 transmits to the upstream composite information processing unit 105 a response message (1-1NG) that rejects the tracking request for the moving object A.

  In FIG. 4B, the moving object A continues to move in the direction of the monitoring camera 110. However, the composite information processing unit 105 determines that the moving object A is in the vicinity of the center of the shootable range of the monitoring camera 100 as before, and determines that there is no change in the takeover importance. For this reason, the composite information processing unit 105 does not transmit a tracking request message to the downstream composite information processing unit 115.

  On the other hand, the moving body B is approaching the direction of the monitoring camera 120. Therefore, the takeover importance is transmitted as a tracking request message (tracking request 2-1 in the figure) from the composite information processing unit 115 of the monitoring camera 110 to the composite information processing unit of the monitoring camera 120. This message includes composite information created by combining the takeover importance (iBA) calculated by the following procedure and the feature information of the moving body B extracted using any conventional technique.

  The takeover importance (iBB) of the downstream monitoring camera 120 with respect to the moving object B being tracked by the upstream monitoring camera 110 is calculated by the composite information processing unit 115 as follows.

  Since only the surveillance camera 110 is tracking the moving object B, (iα) = 10. Since the moving object B matches the attention information stored in advance, (iγ) = 3. Since the moving body B is at a position outside the center of the shootable range of the monitoring camera 110, (iδ) = 1. As a sum of these, (iBB) = 14.

  On the other hand, since the monitoring camera 120 that has received the tracking request message 2-1 does not track the moving object, the tracking request can be accepted regardless of the value of (iBB). Therefore, the composite information processing unit of the monitoring camera 120 returns a response message (2-1 OK in the drawing) for accepting the tracking request of the moving object B to the composite information processing unit 115 of the monitoring camera 110. In addition, the switching unit of the composite information processing unit in the monitoring camera 120 directs the monitoring camera 120 toward the moving body B.

  And the moving body A has moved to the position shown in FIG.4 (c) through FIG.4 (b). In FIG. 4C, the moving object A continues to move toward the monitoring camera 110. The takeover importance (iCA) of the downstream monitoring camera 110 for the moving object A being tracked by the upstream monitoring camera 100 is calculated by the composite information processing unit 105 as follows.

  Since only the surveillance camera 100 is tracking the moving object A, (iα) = 10. Since the moving object A matches the attention information stored in advance, (iγ) = 3. Since the moving object A is out of the center of the shootable range of the monitoring camera 100, (iδ) = 2. The sum of these is (iCA) = 15.

  Since the value of the takeover importance has increased from (iAA = 13) to (iCA) = 15 from the previous time, there is a high possibility that the monitoring camera 110 will accept the takeover request. Then, the tracking request message (tracking request 1-2 in the figure) is retransmitted.

  On the other hand, since the monitoring camera 110 that has received the tracking request message 1-2 is tracking the moving object B, the tracking importance (icB) of the moving object B being tracked by the own monitoring camera is calculated again.

  Since the surveillance camera 110 and the surveillance camera 120 are simultaneously tracking the moving object B, (iα) = 0. The moving object B is deviated from the center of the photographing range of the monitoring camera 110, and (iβ) = 1. Since the moving body B matches the attention information stored in advance, (iγ) = 3. As a sum of these, (icB) = 4.

  Therefore, since (iCA = 15)> (icB = 4), the comparison unit 115c of the composite information processing unit 115 in the monitoring camera 110 accepts the tracking request of the moving object A rather than keeping track of the moving object B. Judge that is more important. Therefore, the switching unit 115d of the composite information processing unit 115 changes the direction of the monitoring camera 110 from the moving body B to the moving body A. Further, the composite information processing unit 115 of the monitoring camera 110 returns a response message (1-2 OK in the drawing) for accepting the tracking request of the moving object A to the composite information processing unit 105 of the monitoring camera 100.

  As a result, as shown in FIG. 4D, the moving camera A is continuously tracked by the monitoring camera 110. The moving body B is continuously tracked by the monitoring camera 120.

  Note that when a larger number of moving objects appear at the same time than in the example of FIG. 4, it is not always possible to photograph all the moving objects at the same time. You can always shoot a moving object that you have determined. In this case, whether to increase the zoom rate so that only the moving objects that each surveillance camera determines to be the most important is captured or whether to decrease the zoom rate so that other moving objects are captured as much as possible is determined in advance by the user. It is decided according to the setting.

(Second Embodiment)
The present invention can also be realized by using a surveillance camera (fixed surveillance camera) with a fixed field of view in place of the surveillance camera with pan / tilt / zoom functions (PTZ surveillance camera) described in the first embodiment. However, when a fixed surveillance camera is used, shooting parameters such as a pan angle, a tilt angle, and a zoom rate are fixed. Therefore, the method of tracking a moving object and the meaning of tracking differ between when a PTZ monitoring camera is used and when a fixed monitoring camera is used.

  When a fixed monitoring camera is used, whether to transmit image data from each monitoring camera can be switched according to the tracking importance information described in the first embodiment. For example, it is possible to perform control such that image data is transmitted to the monitoring device 20 only when a mobile object having an importance level equal to or higher than a predetermined reference value appears.

  Alternatively, when tracking importance information and image data are transmitted from all monitoring cameras that have detected a moving object to the monitoring device 20, the method of handling image data is changed according to the tracking importance information received by the monitoring device 20. can do. For example, the image data may be rearranged in descending order of the importance of tracking, or only the image data having the highest importance may be selected and displayed.

  When a fixed surveillance camera is used as the surveillance camera, the meaning of tracking a moving body is to select or rank image data transmitted from the surveillance camera or image data to be displayed on the monitoring device.

It is a block diagram of the cooperation camera system concerning a 1st embodiment of the present invention. FIG. 2 is a detailed view of a composite information processing unit in FIG. 1. (A) is a figure which shows the relationship between the imaging | photography possible area | region of a surveillance camera, and the degree of tracking importance. (B) is a figure which shows the relationship between the imaging | photography possible area | region of a surveillance camera, and the degree of taking over importance. It is a figure for operation | movement description of the cooperation camera system of this invention. It is a figure for operation | movement description of the conventional cooperation camera system. It is a figure for operation | movement description of another conventional cooperation camera system.

Explanation of symbols

A, B Mobile objects ARa, ARb, ARc Possible shooting range (tracking area)
10 Network 20 Monitoring Device 24 Display Unit (Monitor)
90 Linked camera system 100, 110 Surveillance camera with pan / tilt / zoom function (surveillance camera)
101, 111 Installation information acquisition unit 102, 112 Imaging parameter control unit 103, 113 Image capturing unit 104, 114 Storage unit 105, 115 Complex information processing unit 105a, 115a Tracking importance calculation unit (tracking importance setting means)
105b, 115b Takeover importance calculation unit (takeover importance setting means)
105c, 115c comparison unit (comparison means)
105d, 115d switching part (switching means)

Claims (5)

In the cooperative camera system that captures and tracks the moving object to be tracked by taking over from one surveillance camera to the other surveillance camera,
Tracking importance setting means provided in the one monitoring camera and the other monitoring camera, and for each tracking camera to obtain tracking importance for tracking a moving body in the tracking area of each monitoring camera;
A takeover importance setting means for obtaining a takeover importance degree, which is provided in the one surveillance camera, and the second surveillance camera takes over and tracks the moving body being tracked by the one surveillance camera;
When the other monitoring camera is tracking a moving object within the tracking area of the other monitoring camera, the tracking importance for the moving object and the one monitoring A comparison means for comparing the takeover importance with respect to the moving object being tracked by the camera;
Provided in the other surveillance camera, and after the comparison means detects that the degree of importance of taking over is higher than the degree of importance of tracking, the other surveillance camera tracks. Switching means for switching the moving body to the moving body tracked by the one monitoring camera;
A cooperative camera system characterized by comprising: The tracking importance setting means increases the degree of tracking importance when the moving body being tracked is at the center than the tracking importance when the moving body is at the end of the tracking area of the surveillance camera,
The handover importance setting means increases the degree of tracking importance when the moving body being tracked is at the end than the tracking importance when the moving body is in the center of the tracking area of the monitoring camera.
The cooperative camera system according to claim 1. The tracking importance setting means lowers the degree of tracking importance when the moving body being tracked is less than the tracking importance when the moving body is in the tracking area of the surveillance camera,
The handover importance setting means increases the degree of tracking importance when the moving body being tracked is at the end than the tracking importance when the moving body is in the center of the tracking area of the monitoring camera.
The cooperative camera system according to claim 1. The surveillance camera can change its orientation;
When the switching means switches the moving body tracked by the other monitoring camera to the moving body tracked by the one monitoring camera, the one monitoring camera tracks the direction of the other monitoring camera. Change to a moving object,
The cooperative camera system according to any one of claims 1 to 3, wherein The surveillance camera is a fixed surveillance camera;
A monitor on which a moving object tracked by the surveillance camera is displayed;
When the switching means switches the moving body tracked by the other monitoring camera to the moving body tracked by one monitoring camera, the other monitoring camera tracks the moving body displayed on the monitor. Switch to one moving body,
The cooperative camera system according to any one of claims 1 to 3, wherein

Images