JP4706535B2 - Moving object monitoring device using multiple cameras - Google Patents

Description

  The present invention relates to an apparatus for monitoring a moving object such as a human being or an automobile using a plurality of cameras installed in a monitoring area.

  In order to deal with social unrest, such as the increase in crime rates, the number of cameras installed for the purpose of monitoring suspicious persons and suspicious vehicles is increasing. As described above, monitoring using a large number of cameras requires a monitoring support technique for efficiently monitoring a monitoring area with limited monitoring personnel resources.

  As such a monitoring support technique, there is a “moving object tracking device” disclosed in Patent Document 1. Here, a technique is disclosed in which a moving object to be tracked such as a suspicious person is tracked, and image switching is automatically performed so that the tracking target is displayed on the screen as much as possible. With this technology, after a suspicious person is discovered, a supervisor can easily follow and monitor the suspicious person without gazing at a large number of monitoring images.

  In addition, the “wide area monitoring device” disclosed in Patent Document 2 discloses a technique for displaying a trajectory of a moving object on a monitoring area map and displaying it in cooperation with a monitoring video. Thereby, it becomes possible to easily grasp the past position and the current position of the moving object.

JP-A-6-325180 JP-A-2005-86626

  In the above-described prior art, since the monitoring starts after the monitoring person or device issues the tracking instruction, the monitoring person cannot grasp the situation before the starting of the tracking.

  In addition, in the above-described conventional technique, the locus of a moving object is displayed on the monitoring area map, and the history of the geometric position can be confirmed, but the past situation such as the behavior of the monitoring target is known. I can't.

  The present invention has been made in consideration of the above points, and provides a moving object monitoring apparatus that can easily grasp the status of a monitoring target.

  In the moving object monitoring apparatus according to the present invention, the moving object is selected and moved using the result of tracking the moving object from the feature quantity of the moving object calculated from the images captured by a plurality of cameras. An object movement trajectory is generated, and a monitoring screen is generated from the selected camera image and the generated movement trajectory.

  According to the present invention, the status of the tracking target can be easily grasped by displaying the selected camera video in addition to the trajectory information.

Example 1
FIG. 1 is a block diagram showing a functional configuration of a moving object monitoring apparatus according to an embodiment of the present invention.

  The camera 100 is a photographing device such as a network camera or a video camera. Multiple cameras are installed so that the entire surveillance area can be photographed. The camera installation density may be sparse or dense.

  The video acquisition unit 102 acquires video captured by the camera 100. When the camera 100 is a network camera, image data already digitized via the network is acquired at regular time intervals. In the case of a video camera, an analog signal output from the camera is acquired and digitized.

The video DB 104 is a database (DB) that stores the image data acquired by the video acquisition unit 102. The image data format is JPEG (Joint Photographic
A commonly used one such as Experts Group) or PNG (Portable Network Graphics) may be used. Further, continuous still images may be stored as moving image data. In this case, a format such as MPEG (Moving Picture Experts Group) may be used.

  The feature amount calculation unit 106 calculates the feature amount of the moving object shown in the image data stored in the video DB 104. The feature amount is a vector amount representing a feature unique to the moving object, and there are various features such as a color and texture feature, a shape feature, and a motion feature. For example, the mode luminance values of the lower and upper bodies can be used as the color characteristics of the moving object. In this case, the feature quantity is a two-dimensional vector.

  The feature amount DB 108 is a database (DB) that stores the feature amount of the moving object calculated by the feature amount calculation unit 106 and related information.

  The moving object tracking unit 110 tracks how the moving object has moved between the cameras using the feature amount of the moving object stored in the feature amount DB 108. Here, the tracking target is not limited to a person, and any tracking object may be used.

  The video selection unit 112 uses the tracking result of the moving object tracking unit 110 to select a camera video in which a moving object is shown from the video DB 104.

  The trajectory generation unit 114 uses the tracking result of the moving object tracking unit 110 to generate a moving trajectory of the moving object. The movement locus is a line segment representing the past movement history of the moving object.

  The thumbnail generation unit 116 generates a thumbnail so that the trajectory generated by the trajectory generation unit 114 can be easily understood by the user of the monitoring apparatus. A thumbnail is image data obtained by reducing a video shot by the camera 100. By displaying a past representative video as a thumbnail on the track, it is possible to list past video histories simply by looking at the movement track.

  The monitoring screen generation unit 118 uses the video data accumulated in the video DB 104, the video data showing the moving object selected by the video selection unit 112, and the trajectory data generated by the trajectory generation unit 114 to inform the user of this monitoring apparatus. Generate the monitoring screen to be provided.

  The input unit 120 is an input device such as a mouse or a keyboard, and accepts user operations of the monitoring device such as “tracking start” and “tracking end”.

  The output unit 122 is an output device such as a display, and displays a monitoring screen generated by the monitoring screen generation unit 118 to the user.

  Next, a display example of the monitoring screen generated by the monitoring screen generation unit 118 and displayed on the output unit 122 will be described with reference to FIGS. The monitoring screen is composed of a multiple video monitoring screen and a tracking screen. This will be described in order below.

  First, an example of a multiple video monitoring screen for browsing a plurality of videos will be described with reference to FIG.

  A map 200 on the left side of the screen is a map showing an outline of the entire monitoring area. This example shows that a three-story building is a monitoring target.

  The icon 202 indicates the geometric arrangement of the camera on the map 200. In this example, six cameras are arranged in the monitoring area.

The video monitoring area 204 is an area for displaying video captured by the camera 100. In this example, images from four cameras such as the monitoring image 206 are displayed. When displaying images from five or more cameras, the display area can be changed with the scroll bar so that the number of divisions of the image monitoring area 204 is increased or a part of the image monitoring area 204 is displayed. do it. Conversely, when displaying images from three or less cameras, the number of divisions of the video monitoring area 204 may be reduced.

  With this multiple video monitoring screen, the user of the present apparatus can simultaneously browse a plurality of videos shot by the camera 100. When a suspicious person is confirmed on the screen, the tracking start can be instructed by clicking the monitoring video in the video monitoring area 204 with the mouse. By clicking the tracking end button 208 with the mouse, the currently running tracking can be stopped.

  For example, the icon 202 and the monitoring video 206 may be connected by a line segment as a link 210 so that the correspondence between the icon 202 and the video in the video monitoring area 204 becomes clear. Further, the name of the camera corresponding to the icon 202 may be displayed on the video in the video monitoring area 204.

  Next, an example of a tracking screen for monitoring a specific moving object will be described with reference to FIG.

  The configuration of the entire screen is basically the same as the multiple video monitoring screen of FIG. The only difference is that the locus 300 of the moving object is displayed on the map 200 and only the video of the camera tracking the tracking target is displayed in the monitoring area 204.

The moving object trajectory 300 includes a thumbnail 302 and an arrow 306. The thumbnail 302 is image data obtained by reducing a video photographed by the camera. Image data captured by a camera whose tracking target corresponds to the icon 304 is used as a past history. On the other hand, an arrow 306 indicates that the thumbnail 302 has been moved to the thumbnail 308. The trail start point thumbnail 302 indicates that tracking has started with the camera corresponding to this thumbnail, and the trail end point thumbnail 310 indicates that the moving object is currently being tracked by the camera corresponding to this thumbnail. Yes. Here, the camera that is currently tracking the moving object is called a current camera.

On the monitoring area 204, a monitoring image 206 taken by the current camera is displayed. At this time, by highlighting the icon 202 corresponding to the current camera, the camera that captured the monitoring video 206 can be easily grasped. When the thumbnail 302 is clicked with a mouse, a monitoring video corresponding to the thumbnail may be displayed in the monitoring area 204.

  In this way, by displaying a past representative video as a thumbnail on the trajectory, it is possible to easily grasp the past situation of the moving object simply by looking at the moving trajectory.

  Although FIG. 3 shows an example of the tracking screen when the number of tracking is one, the number of tracking may be plural. FIG. 18 shows an example of a tracking screen when the number of tracking is two. On this screen, a movement locus 300 and a movement locus 1800 are displayed. The movement trajectory 300 is the same as that shown in FIG. Similar to the movement locus 300, the movement locus 1800 includes thumbnails and arrows. In addition, a monitoring image 1802 corresponding to the current camera of the movement locus 1800 is displayed below the monitoring area 204. When displaying the movement locus, the movement locus 1800 is displayed with an offset in the right direction and the downward direction with respect to the movement locus 300 in order to prevent the movement locus from overlapping and disappearing.

Next, the basic principle of moving object tracking will be described with reference to FIG. The left side is a monitoring video taken by the camera, and the right side is a graph of the feature amount calculated by the feature amount calculation unit 106 for the monitoring video. In addition, when graphing the feature values, not only the feature values for a specific monitoring video but also the feature values for several monitoring videos immediately before are displayed. Therefore, feature quantities are plotted for one tracking target. This is to improve tracking accuracy in consideration of feature amount calculation errors. Note that the monitoring video acquisition times are t1, t2, and t3 (t1 <t2 <t3) from the top to the bottom.

  First, in the monitoring video 400 at time t1, a person 402 and a person 404 exist as moving objects, and correspond to the feature quantity 408 and the feature quantity 410 in the feature quantity space 406, respectively.

Next, in the monitoring video 412 at time t2, a person 414 and a person 416 exist as moving objects, and correspond to the feature quantity 420 and the feature quantity 422 in the feature quantity space 418, respectively. Here, since the feature quantity 408 in the feature quantity space 406 and the feature quantity 420 in the feature quantity space 418 are located at the same place, the person 402 is likely to be the same as the person 414, and the monitoring video 400 changes to the monitoring video 412. It can be determined that it has moved.

  Here, the Euclidean distance between the respective centroids is used for the equivalence determination of the feature quantity 408 and the feature quantity 420. If this distance is smaller than a predetermined threshold value, it is determined that both are sufficiently close and the same. In addition to the Euclidean distance, the Mahalanobis distance considering the variance of the group may be used. It is also possible to test whether there is a difference between the two groups using Wilkes' Λ statistics. Mahalanobis distance and Wilks' Λ statistic are methods often used in multivariate analysis, and are described, for example, in the publication “Multivariate Analysis Understandable” (Sadao Ishimura, Tokyo Book, 1992).

  Next, in the monitoring video 424 at time t3, a person 426 exists as a moving object, and corresponds to the feature quantity 430 in the feature quantity space 428. Here, since the feature quantity 410 in the feature quantity space 406 and the feature quantity 430 in the feature quantity space 428 are located at the same place, the person 404 is likely to be the same as the person 426, and the monitoring video 400 changes to the monitoring video 424. It can be determined that it has moved.

  Thus, by searching for feature quantities having similar values in the feature quantity space, it becomes possible to track how a specific moving object has moved between cameras.

  Next, the flow of processing of the monitoring method in the monitoring apparatus of the present embodiment will be described using FIG.

  In step 500, the processing from step 502 to step 518 is repeated at a predetermined frequency.

  In step 502, the video acquisition unit 102 acquires a monitoring video shot by the camera.

  In step 504, the feature amount calculation unit 106 calculates the feature amount of the moving object in the video acquired in step 502.

  In step 506, the monitoring screen generation unit 118 processes the operation of the user of the monitoring apparatus. Here, the user's operation is input using the input means 120.

  In step 508, the processing from step 510 to step 516 is repeated for the currently executed tracking.

In step 510, the time t to be processed is set to the current time. Thereby, the tracking process in step 512 is performed based on the current time.

  In step 512, person tracking is executed by the moving object tracking unit 110 based on the time t set in step 510.

  In step 514, it is determined whether or not a camera-to-camera transition has occurred as a result of the person tracking in step 512. If it has occurred, step 516 is executed. Here, the transition between the cameras means that the tracking target disappears from the camera originally shown and appears in another camera.

  In step 516, the locus generation unit 114 generates a movement locus of the person.

  In step 518, a monitoring screen is generated by the monitoring screen generation unit 118 and displayed on the output unit 122.

  Next, details of the input processing in step 506 in FIG. 5 will be described using the flowchart in FIG. 6.

  In step 600, an operation of the user of the monitoring apparatus from the input unit 120 is accepted.

  In step 602, the input result in step 600 is determined. If “start tracking” is designated, steps 604 to 608 are executed. “Start tracking” is specified by, for example, clicking the monitoring video 206 of the multiple video monitoring screen in FIG. 2 with a mouse. On the other hand, when “end tracking” is designated, step 610 is executed. “Tracking end” is specified by, for example, clicking the tracking end button 208 of the multiple video monitoring screen in FIG. 2 with a mouse.

  In step 604, the feature quantity corresponding to the person shown in the camera designated in step 600 is stored in the feature quantity DB as the trace target feature quantity.

  In step 606, the camera designated in step 600 is set as the current camera. The current camera is a camera that currently tracks a moving object at the time of tracking a person as described above.

  In step 608, a new trace is registered based on the information obtained in steps 604 and 606 in order to be added to the processing target of step 508 in FIG.

  In step 610, all currently running traces are stopped. As a result, there is no tracking to be processed in step 508 of FIG.

  Next, details of the person tracking process in step 512 of FIG. 5 will be described using the flowchart of FIG.

  In step 700, it is determined whether or not a tracking person exists in the video of the current camera. The tracked person is a person corresponding to the movement locus designated as the processing target in step 508 of FIG. In addition, although expressed as a person here, it shall include the whole moving objects, such as a vehicle. Whether or not a tracking person exists in the current camera is determined in the feature amount space as shown in FIG. If the tracking target feature amount exists in the feature amount space corresponding to the video of the current camera even at the time t to be processed, it is determined that there is no tracking person in the current camera, and steps 702 to 706 are executed. .

  In step 702, it is searched whether there is a tracking person other than the current camera. This search process will be described in detail with reference to FIG.

  In step 704, the result of the tracking person search process in step 702 is determined. If there is a tracking person in another camera, step 706 is executed.

  In step 706, based on the search result in step 702, a camera in which the tracking person is present is newly set as the current camera.

  In step 708, it is determined whether a predetermined time specified in advance has elapsed since the start of processing. This is a process for aborting the tracking process in time. If necessary, it is possible to avoid the termination of the tracking process by specifying infinity at a predetermined time. If the result of determination is that a predetermined time has elapsed, step 710 is executed.

  In step 710, a stop process is executed to stop tracking the current processing target.

  Next, details of the tracking person search processing in step 702 of FIG. 7 will be described using the flowchart of FIG.

  In Step 800, Steps 802 to 808 are repeatedly executed for all the cameras of the monitoring apparatus of this embodiment.

  In step 802, it is determined whether or not the camera to be processed in step 800 is the same as the current camera. If they are the same, step 804 is executed.

  In step 804, the current loop processing in step 800 is skipped to proceed to the next loop processing in order to improve processing efficiency. This is because it is determined in step 700 in FIG. 7 that no person to be tracked exists in the current camera.

  In step 806, it is determined whether there is a person to be tracked in the camera that has been processed in step 800. Whether or not a tracking person exists in the camera to be processed is determined in the feature space as shown in FIG. If the result of determination is that it exists, step 808 is executed.

  In step 808, the current process target camera is returned as a return value, and the process is terminated. On the call side of this flow, it can be seen that camera crossing occurs and that there is a tracking target in the current processing target camera.

  In step 810, NULL meaning invalid is returned as a return value, and the process is terminated. It can be seen that there is no camera crossing on the call side of this flow.

  Next, an example of the data structure of the feature amount stored in the feature amount DB 116 in FIG. 1 will be described with reference to FIG. The feature amount data includes a table 900 representing the entire feature amount data and a table 910 representing the feature amount vector.

A table 900 is a table representing one piece of feature amount data. This table includes data items 902 to 908. The data item 902 is an ID that uniquely represents the camera. The feature amount is calculated from the video imaged by the camera corresponding to the ID. A data item 904 is an ID representing a person to be tracked. This ID is determined in the process of tracking a person in one camera. To the last, it is uniquely determined within one camera, and is not uniquely determined for the same person among a plurality of cameras. A data item 906 is an acquisition time of the video for which the feature amount is calculated. A data item 908 is a pointer to an area for storing the calculated feature vector and points to the table 910.

  A table 910 representing a feature vector is composed of data items 912 and variable-length data items 914. Data item 912 is the order of the vector. The order of the feature vector is not uniquely determined because there are various features such as color, texture, shape, and motion. Here, the order of the feature quantity vector is held in the data item 912 so that the feature quantity to be adopted can be freely changed. For example, when the mode luminance values of the lower body and the upper body are used as the color characteristics of the moving object, the order is 2. The data item 914 is an element of the feature vector, and there are as many items as the number of orders specified in the data item 912.

  Next, FIG. 17 shows an example of functional arrangement when the monitoring apparatus of the embodiment of FIG. 1 is constructed in a network environment. This system is constructed in such a manner that a camera 100 and nodes 1702 to 1708 are connected to a network 1700.

  A network 1700 is a commonly used network such as a LAN. The functional blocks in FIG. 1 are arranged via the network 1700.

  The camera 100 is a camera having a network interface and is connected to the network 1700. By adopting such a configuration, the video of the camera 100 can be acquired from various devices connected to the network.

  The video acquisition unit 102 and the video DB 104 are arranged in the node 1702 that acquires the video.

  In the node 1704 for calculating the feature amount, the feature amount calculation unit 106 and the feature amount DB 108 are arranged.

  A moving object tracking unit 110, a trajectory generation unit 112, and a thumbnail generation unit 116 are arranged in a node 1706 that performs person tracking.

  In the node 1708 that generates the monitoring screen, the video selection unit 112, the monitoring screen generation unit 114, the input unit 116, and the output unit 118 are arranged.

  According to the embodiment described above, it is possible to track and display a moving object such as a specific person while displaying a moving locus with a thumbnail on the monitoring area map. Therefore, the user of the monitoring apparatus of this embodiment can track and monitor a moving object with a smaller load. Also, by confirming the movement trajectory, it is possible to easily grasp what route has been taken in the past and where it is now. Also, by checking the thumbnails, the past status of the tracking target can be confirmed with images. It is also possible for the user to detect the moving object tracking mistake that the apparatus tracks different moving objects as the same moving object at an early stage by checking with the thumbnail.

Example 2
In the first embodiment, the tracking target is tracked in the positive direction of the time axis, but may be tracked in the reverse direction of the time axis, that is, tracing back to the past. Here, such tracking is called reverse tracking. An example in this case is shown below. Basically, most of the configuration is the same as that of the first embodiment, and therefore, there will be described a portion having a change.

  FIG. 10 is a block diagram illustrating a functional configuration of a moving object monitoring apparatus having a reverse tracking function. The functional blocks other than the moving object reverse tracking unit 1000 in FIG. 10 are the same as those shown in FIG.

  The moving object reverse tracking unit 1000 uses the feature amount of the moving object stored in the feature amount DB 108 to trace back how the moving object has moved between the cameras.

  FIG. 11 shows an example of a tracking screen of a moving object monitoring apparatus having a reverse tracking function. Most of the tracking screen is basically the same as the tracking screen shown in FIG.

  The locus 1100 includes a thumbnail 1102 and an arrow 1104. An arrow 1104 indicates that the thumbnail 1106 existed before moving to the thumbnail 1102. In this way, the trajectory is displayed retroactively in the opposite direction to the trajectory 300 of FIG. A thumbnail 1102 serving as the start point of the locus indicates that reverse tracking has started with the camera corresponding to the thumbnail. On the other hand, the thumbnail 1108 which is the end point of the locus indicates that the reverse tracking is completed by the camera 100 corresponding to the thumbnail. The reverse tracking end condition includes a case where the person to be reverse-tracked is lost, a case where the time traced back by the reverse tracking exceeds a predetermined time, or the like.

  Next, the flow of processing of the monitoring method in the monitoring apparatus of the present embodiment having the reverse tracking function will be described using a flowchart with reference to FIG. Basically, it is the same as the flowchart shown in FIG. 5 except that step 510 is changed to step 1200.

  In step 1200, the reverse tracking processing time tb is reduced by Δt given in advance. This means that the target time of the tracking process is traced back by Δt. Thereafter, the target time t of the tracking process is set to tb. Thereby, the tracking process in step 512 is executed based on the information at time tb.

  Next, details of the input processing in step 506 in FIG. 12 will be described using the flowchart in FIG. 13. Basically, it is the same as the flowchart shown in FIG. 5 except that step 1300 is added.

  In step 1300, the current time is set as the reverse tracking processing time tb. This is because the tracking target is traced back to the past starting from the current time.

  According to this embodiment, a moving object such as a specific person can be traced back in the past while displaying a moving trajectory with a thumbnail on the monitoring area map. Therefore, the user of the monitoring apparatus of the present embodiment can easily grasp the past behavior of the moving object. In other words, by confirming the movement trajectory, it is possible to easily grasp what route has been taken in the past and where it is now. Also, by checking the thumbnails, the past status of the tracking target can be confirmed with images.

Example 3
In the first and second embodiments, the tracking target is tracked in either the forward or reverse direction of the time axis, but may be tracked in both directions, that is, in both the forward and reverse directions. Here, such tracking is called forward / reverse tracking. An example in this case is shown below. Basically, most of the configurations are the same as those in the first and second embodiments, and therefore, the portions with changes will be described.

  FIG. 14 is a block diagram showing a functional configuration of a moving object monitoring apparatus having a forward / reverse tracking function. The functional blocks other than the moving object forward / reverse tracking unit 1400 in FIG. 14 are the same as those shown in FIG.

  The moving object forward / reverse tracking unit 1400 uses the feature amount of the moving object stored in the feature amount DB 108 to determine how the moving object has moved between the cameras in both the forward and reverse directions of the time axis. Chase.

  FIG. 15 shows an example of a tracking screen of a moving object monitoring apparatus having a forward / reverse tracking function. Most of the tracking screen is basically the same as the tracking screen shown in FIG. 3, and the difference is a movement locus 1500 and a movement locus 1502.

  The movement trajectory 1500 is tracking in the forward direction starting from the thumbnail 1504, and the movement trajectory 1502 is tracking in the reverse direction similarly starting from the thumbnail 1504. It is the same as the trajectory described in FIGS. In this way, both the forward and backward movement trajectories are displayed.

  Next, the flow of processing of the monitoring method in the monitoring apparatus of this embodiment having a forward / reverse tracking function will be described with reference to FIG. Basically, this is the same as the flowchart shown in FIG. 5 except that Steps 1600 to 1606 are added. Steps 510 to 516 are forward tracking processes, while steps 1600 to 1606 are backward tracking processes. Steps 1600 to 1606 are the same processes as steps 1200 to 516 in FIG.

  According to the embodiment described above, it is possible to track and display a moving object such as a specific person while displaying a moving locus with a thumbnail on a monitoring area map, and at the same time, can perform backward tracking display retroactively. become. Therefore, the user of the monitoring apparatus of the present embodiment can track and monitor the moving object with a small load, and can easily grasp the past behavior of the moving object.

Example 4
In the first to third embodiments, the user of the monitoring device explicitly issues a tracking instruction using the input unit 120. However, the apparatus automatically detects the abnormal behavior with the apparatus, and automatically issues the tracking instruction at the time of detection. You may make it take out. An example in this case is shown below. Basically, since most of the configuration is the same as that of the first embodiment, a portion where there is a change will be described.

  FIG. 19 is a block diagram illustrating a functional configuration of a moving object monitoring apparatus having an abnormality detection function. 1 is the same as the block diagram of FIG. 1 except that an abnormal behavior detection unit 1900 is added.

  The abnormal behavior detection unit 1900 detects abnormal behavior using image data stored in the image DB 104. When abnormal behavior is detected, a tracking start instruction signal is output to the moving object tracking unit 110 to start tracking a new moving object.

  Here, as a method for detecting abnormal behavior, for example, it is described in a publication “Detection of abnormal motion from a plurality of moving images” (Information Processing Society of Japan Research Report, Computer Vision and Image Media, 2004-CVIM-145). What is necessary is just to utilize the well-known abnormal action discrimination method using a solid higher-order local autocorrelation feature. The three-dimensional higher-order local autocorrelation feature is a feature obtained by extending a higher-order local autocorrelation feature often used in face image recognition or the like in the time direction, and can obtain a feature amount of a moving image. This feature is a 251-dimensional vector quantity obtained by calculating 251 types of local autocorrelation features at each point in the voxel data in which images are arranged in time series and integrating over the entire voxel data. Then, the feature amount of the moving image obtained by this method is subjected to discriminant analysis to determine whether or not it is abnormal behavior.

  According to the present embodiment, it is possible to detect abnormal behavior in a video photographed by a camera and automatically start tracking. Therefore, it is possible to track a moving object such as a suspicious person without issuing a tracking instruction by the monitoring person, and as a result, it is possible to reduce the burden on the monitoring person.

The block diagram which shows the function structure of the moving object monitoring apparatus which is one Example of this invention. An example of a multiple video surveillance screen. Example of a tracking monitoring screen. Explanatory drawing of the basic principle of moving object tracking. The processing flow of the monitoring method in the monitoring apparatus of an Example. Details of input processing. Details of the person tracking process. Details of the tracking person search process. An example of the structure of feature data. The block diagram which shows the function structure of the moving object monitoring apparatus with a reverse tracking function. The example of the tracking screen of the moving object monitoring apparatus with a reverse tracking function. The flow of processing of a monitoring method in a monitoring device having a reverse tracking function. Details of the input processing of the moving object monitoring device having the reverse tracking function. The block diagram which shows the function structure of the moving object monitoring apparatus with a normal / reverse tracking function. An example of a monitoring screen of a moving object monitoring device having a forward / reverse tracking function. The flow of processing of a monitoring method in a monitoring device having a forward / reverse tracking function. FIG. 2 is an example of a functional layout when the monitoring apparatus of the embodiment of FIG. 1 is built in a network environment. An example of a tracking monitor screen for multiple tracking. The block diagram showing the functional structure of the moving object monitoring apparatus with an abnormal action detection function.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Camera, 102 ... Image | video acquisition part, 104 ... Image | video DB, 106 ... Feature-value calculation part,
DESCRIPTION OF SYMBOLS 108 ... Feature-value DB, 110 ... Moving object tracking part, 112 ... Image | video selection part, 114 ... Trajectory generation part, 116 ... Thumbnail generation part, 118 ... Monitoring screen generation part, 120 ... Input means, 122 ... Output means.

Claims (6)

In a moving object monitoring apparatus that monitors a moving object using a plurality of cameras,
A feature amount calculation unit for calculating a feature amount of a moving object from images captured by a plurality of cameras;
A moving object tracking unit that tracks the moving object from the feature amount;
A video selection unit that selects a camera video in which the moving object is shown using the tracking result of the moving object tracking unit;
Using the tracking result, a trajectory generating unit that generates a moving trajectory of the moving object;
A monitoring screen generation unit that generates a monitoring screen from the camera image selected by the video selection unit and the movement locus generated by the locus generation unit;
A thumbnail generation unit that generates thumbnails from videos captured by the plurality of cameras;
And the monitoring screen generating unit generates the monitoring screen so that the thumbnail is superimposed on the moving locus . Oite to claim 1, wherein the moving object tracking unit, said moving object, the moving object monitoring device, which comprises traced back to the past. Oite to claim 1, wherein the moving object tracking unit, a moving object monitoring device, characterized in that the moving object to track towards the both forward and backward direction of the time axis. In any one of claims 1 to 3, further detects the occurrence of abnormal behavior from images taken by the plurality of cameras, further comprising abnormal behavior detection unit that instructs the tracking start to the moving object tracking unit A moving object monitoring device. In any one of claims 1 to 4, wherein the monitoring screen generation unit, a moving object monitoring device and generates a monitoring screen displaying a map representing the monitored area. Characterized in any one of claims 1 to 5, wherein the moving object tracking unit, the Euclidean distance between the centers of gravity, the Mahalanobis distance between the center of gravity, to calculate the feature amount using any of Λ statistics Uirukusu A moving object monitoring device.

Images