JP2017017488A - Monitoring system, imaging control method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To appropriately associate traffic line with a movable body without requiring a troublesome work, while preventing a monitoring system from being lowered.SOLUTION: The present invention comprises: first imaging means for imaging a plurality of movable bodies corresponding to each of the plurality of traffic lines to be monitored; characteristic amount extraction means for extracting the characteristic amount of an object movable body to be processed, on the basis of the captured image obtained by the first imaging means; estimation means for estimating the identification degree of individual identification of the object movable body, on the basis of the characteristic amount of the object movable body; priority calculation means for calculating the priority in imaging the object movable body by the second imaging means, on the basis of the identification level; and transmission means for transmitting an imaging instruction of the object movable body with which the priority is associated, to the second imaging means.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a monitoring system, a photographing control method, and a program.

  Conventionally, surveillance systems represented by surveillance cameras have been used for crime prevention and security maintenance, such as detection of lost children and detection of suspicious persons in the surveillance area. On the other hand, in recent years, monitoring systems that are used for multiple purposes have become widespread, for example, by analyzing monitoring records in detail and applying it to the analysis of human consumption behavior in stores. In order to collect and analyze information (personal behavior, vehicle behavior, etc.) necessary for such advanced analysis, a system in which a plurality of surveillance cameras and sensors are linked is known. In addition, a system using a flow line tracking technique for performing detailed analysis of a movement trajectory of a person or a vehicle is also known.

  As a technique for linking a plurality of cameras, Patent Document 1 collects a plurality of captured images obtained by a plurality of cameras, and performs feedback for urging each camera to move based on the overlapping amount of the visual field of each camera. Technology is disclosed. In this way, the blind spot in the monitoring area can be reduced. In addition, as a flow tracking technique for a person, a vehicle, etc., Patent Document 2 describes the flow line information of a person who looks down on the surveillance area after linking a plurality of cameras, as well as each individual who has taken a telephoto / enlarged image. A technique for storing a detailed face image associated with a flow line is disclosed.

JP 2011-114580 A Japanese Patent No. 4585580

In the technology for tracking the flow line, if it is difficult to identify the mobile object (for example, personal authentication using a face image) due to the orientation, blurring, or hiding of the mobile object such as a person, Cannot be linked. On the other hand, it is possible to obtain an appropriate photographed image and link it by directing the camera to a moving body that cannot be individually identified. However, it is difficult to perform the tying by directing the camera to all the moving bodies, and it becomes a troublesome work.
The present invention has been made in view of such problems, and an object of the present invention is to appropriately associate a flow line and a moving body without requiring troublesome work while preventing a deterioration of a monitoring system.

  Therefore, the present invention is a monitoring system, and includes a first imaging unit that images a plurality of moving bodies corresponding to a plurality of flow lines to be monitored, and a captured image obtained by the first imaging unit. A feature amount extraction unit that extracts a feature amount of a target mobile object to be processed, an estimation unit that estimates the degree of individual identification of the target mobile body based on the feature amount of the target mobile body, and Priority calculation means for calculating the priority of shooting by the second shooting means of the target moving body based on the degree of identification, and a shooting instruction for the target moving body in association with the priority. Transmission means for transmitting to the photographing means.

  ADVANTAGE OF THE INVENTION According to this invention, a flow line and a mobile body can be appropriately linked | related, without requiring a troublesome operation | work, preventing the fall of a monitoring system.

It is a figure which shows the outline | summary of a monitoring system. It is a figure which shows the hardware constitutions of the monitoring server apparatus. It is a figure which shows the software structure of the monitoring server apparatus. It is a figure which shows the data of an external storage part. It is a figure which shows the data of an external storage part. It is a figure which shows the data of an external storage part. It is a figure which shows a display screen. It is a flowchart which shows imaging | photography control processing. 10 is a flowchart illustrating a shooting control process according to the second embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a diagram showing an overview of a monitoring system according to the present embodiment. The monitoring server device 101 is connected to the monitoring camera device 103 via the communication line 102 and monitors the monitoring space 104. Specifically, as the monitoring camera device 103, in addition to a fixed monitoring camera, a camera capable of changing the angle of view and orientation, and a camera capable of moving a pan head can be considered.

  One or more persons 105 pass through the monitoring space 104. The surveillance camera device 103 records an image of a person 105 as an example of a moving object, and performs flow line measurement and feature quantity measurement of a person to be monitored. Specifically, the surveillance camera device 103 individually detects each person 105 and records the movement trajectory as the flow line data 106. Further, as the feature amount measurement, the monitoring camera device 103 shoots a face image of the person 105 and generates a face image 107, and then measures and digitizes the face feature amount. The monitoring camera device 103 transmits measurement data such as flow line data and feature quantities to the monitoring server device 101 via the communication line 102 together with the installation position information, shooting direction, and shooting angle of view of the monitoring camera device 103. The monitoring server device 101 may perform the digitization of the facial feature amount.

  On the other hand, the monitoring server device 101 is connected via a wireless communication device 108 to a wearable device 109 worn by a security guard or a store official. The wearable device 109 may be in various forms such as a tablet-type device and a mobile phone-type device in addition to a wearable device on the head. The wearable device 109 has a photographing function and plays the role of a mobile surveillance camera device. The wearable device 109 monitors the monitoring space 104 from the same viewpoint as the person 105. The wearable device 109 records the flow line data 106 or the face image 110 of the person 105 as necessary, and transmits it to the monitoring server apparatus 101 via the wireless communication apparatus 108. Note that a plurality of wearable devices 109 may exist as necessary.

  FIG. 2 is a diagram illustrating a hardware configuration of the monitoring server apparatus 101 according to the present embodiment. The calculation unit 201 performs calculation for processing information, logical determination, and the like. The arithmetic unit 201 controls each component connected to the system bus 206 via the system bus 206. A display unit 202 is a controller that controls display of image information, and an output device such as a liquid crystal panel and a projector.

  The communication unit 205 is a network controller represented by LAN, 3G, 4G, Bluetooth (registered trademark), RFID (Radio Frequency IDentification) technology, or the like. The communication unit 205 is an external communication unit that controls connection with other devices. Note that the communication method is not limited to a specific example as long as it can achieve the same purpose.

  A RAM (Random Access Memory) 207 is used for temporary storage of various data from each component. The external storage unit 208 uses a physical medium such as a flash memory, HDD, or optical disk, and stores various setting data, image data, and the like in addition to control program codes such as processing programs executed in the present embodiment. The monitoring server device 101 including each component operates in response to various inputs from the UI unit 204 and various inputs via the network supplied from the communication unit 205. That is, when an input from the UI unit 204 and an input from the communication unit 205 are supplied, an interrupt signal is first sent to the arithmetic unit 201. Then, the arithmetic unit 201 reads various control signals stored in the external storage unit 208, and various controls are performed according to the control signals.

  Functions and processes of the monitoring server device 101 described later are realized by the calculation unit 201 reading a program stored in the external storage unit 208 and executing the program. Note that the hardware configuration of the monitoring camera device 103 is substantially the same as the hardware configuration of the monitoring server device 101.

  The computing unit 201 of the monitoring server apparatus 101 also controls the photographing of a document or a person's action by the monitoring camera apparatus 103 by issuing an instruction to the monitoring camera apparatus 103 via the communication unit 205. Capture the behavior of a document or person. Examples of the document type include physical device documents such as single sheets and booklets, electronic device display documents displayed on tablet devices, and the like. The UI unit 204 is a button, a touch panel, a keyboard, a mouse, or the like, and receives an instruction from the user.

  FIG. 3 is a diagram illustrating a software configuration of the monitoring server apparatus 101. The monitoring server apparatus 101 includes an imaging processing unit 301, a feature amount extraction unit 302, an identification degree estimation unit 303, an addition determination unit 304, a priority calculation unit 305, a table management unit 306, and a communication processing unit 307. have. The imaging processing unit 301 controls imaging by the monitoring camera device 103. The feature amount extraction unit 302 identifies the face image of the person to be monitored included in the captured image captured by the monitoring camera device 103, and extracts the feature amount of the face image. The discrimination level estimation unit 303 estimates the discrimination level for the feature amount. Here, the degree of identification is an index indicating whether or not individual identification of a person can be performed based on a feature amount. The feature amount and the degree of identification will be described in detail later.

  The addition determination unit 304 determines whether or not re-shooting is necessary based on the degree of identification. The priority calculation unit 305 calculates the shooting priority for the monitoring target person based on the identification level and the distance between the wearable device 109 and the monitoring target person. The table management unit 306 manages various tables stored in the external storage unit 208. The communication processing unit 307 controls communication with an external device such as the wearable device 109.

  FIG. 4A is a diagram showing a flow line table 401 stored in the external storage unit 208 of the monitoring server apparatus 101. FIG. 4B is a diagram showing a flow line coordinate table 402 stored in the external storage unit 208 of the monitoring server apparatus 101. The monitoring server apparatus 101 stores a flow line table 401 and a flow line coordinate table 402 as the flow line data 106. In the flow line table 401, a flow line ID corresponding one-to-one with a fixed-length trajectory is registered. In the flow line coordinate table 402, the measurement device, measurement time, coordinates, speed, direction, and the like measured for the flow line ID defined in the flow line table 401 are registered. The coordinates and speed are coordinate-transformed using prior information such as the shooting position and the shooting angle of view in order to unify and manage the measurement results from a plurality of shooting devices.

  FIG. 5A is a diagram illustrating a feature amount table 501 stored in the external storage unit 208 of the monitoring server apparatus 101. FIG. 5B shows the feature quantity 502. The feature quantity 502 is a face feature quantity obtained by analyzing the face image 107 photographed by the monitoring camera apparatus 103 or the face image 110 photographed by the wearable device 109. The feature quantity 502 is used for personal identification, and is information obtained by digitizing eye / mouth / nose coordinates, estimated age, and the like from a face image. In addition, as another example of the feature amount, information using a facial expression, a skin color, clothes, belongings, and the like, and numerically representing a luminance value and a shape may be used.

  In the feature quantity table 501, a shooting ID, a person ID, a feature quantity, a blurring degree, a face orientation, and a feature quantity identification degree are recorded in association with each other. The shooting ID is feature amount identification information. The person ID is identification information of a person who is a moving object to be monitored. The degree of identification is a scale representing the ease of individual identification of a person based on the corresponding feature amount or face image. Objects with large blurring and hiding, and face images that do not face the front are likely to reduce the monitoring accuracy, so the degree of discrimination is set low. Further, when the correlation between the measured and flow line and the associated feature quantity is too high, the discrimination degree is set low in order to increase the possibility of erroneous detection.

  Examples of quantification of the identification level include the resolution of the face image, the number of pixels blurred due to movement, the estimated ratio of the area hidden by the obstacle, the relative angle from the front as the face direction, etc. is there. In addition, it is conceivable to use a distance function such as a Euclidean distance and a Mahalanobis distance for each component of the monitoring target feature quantity, and a co-occurrence rate of each component. The examples of the degree of discrimination given here may be used singly or may be determined by combining a plurality of weights. When implemented in combination with a plurality, since the tendency of the feature amount is measured without being biased, it is possible to improve the data identification accuracy. At this time, if the correlation with the measured and linked feature quantity is higher than the threshold time despite the high degree of success in photographing, it is estimated as the same person, and the same person ID is assigned to the corresponding plurality of records. Shall be allocated.

  FIG. 6 is a diagram showing the correspondence table 601 stored in the external storage unit 208 of the monitoring server device 101. The correspondence table 601 stores association information between shooting IDs and flow line IDs. FIG. 7 is a diagram showing a display screen 701 displayed on the display unit 202 of the wearable device 109 in the present embodiment. The display screen 701 displays various types of peripheral information obtained from the monitoring server device 101 in addition to the state in the field of view. The display unit 202 of the wearable device 109 is implemented by an eyepiece display, a touch panel display, or the like. The display screen 701 includes an overhead information area 702 for displaying overhead information of the monitoring space 104, a flow line tracking frame 703 for displaying the flow line tracking status of the person 105, and an action instruction for displaying an action instruction from the monitoring server device 101 and the like. There is a display area 704 and the like.

  FIG. 8 is a flowchart illustrating the imaging control process performed by the monitoring server apparatus 101. Note that the monitoring server apparatus 101 can track the position and locus of a person based on analysis of a captured image received from the monitoring camera apparatus 103 and information from an RFID tag attached to a person to be monitored. It is assumed. Note that the imaging control process shown in FIG. 8 is a process executed for each of a plurality of flow lines corresponding to a plurality of persons to be monitored. Note that the surveillance camera apparatus 103 executes the imaging control processing for each flow line asynchronously and in parallel.

  First, in step S <b> 801, the arithmetic unit 201 of the monitoring camera device 103 determines whether the flow line to be processed is a newly measured flow line or a flow line designated as a retry target. When the flow line to be processed is either a new flow line or a flow line to be retried (Yes in S801), the arithmetic unit 201 advances the process to S802. When the processing target flow line does not correspond to either a new flow line or a retry target flow line (No in S801), the arithmetic unit 201 ends the process for the processing target flow line.

  In step S <b> 802, the feature amount extraction unit 302 specifies a face image corresponding to the flow line to be processed from the captured image captured by the monitoring camera device 103, and extracts the feature amount from the specified face image. Here, the process of S802 is an example of a feature amount extraction process for extracting a feature amount of a person as a target mobile object to be processed. Then, the table management unit 306 assigns a new shooting ID and registers the extracted feature amount in the feature amount table 501. Note that the monitoring server apparatus 101 can omit the process of S802 when it receives the feature amount from the monitoring camera apparatus 103. In step S <b> 803, the identification level estimation unit 303 estimates the identification level of the individual identification of the person based on the feature amount (estimation process). The discrimination level estimation unit 303 further registers the estimated discrimination level in the feature quantity table 501 in association with the feature quantity registered in S802.

In step S <b> 804, the additional determination unit 304 calculates a shooting necessity level that is an index of the necessity for additional shooting. For example, the addition determination unit 304 calculates the photographing necessity N (d) based on the identification degree d by (Equation 1). As a result, the additional determination unit 304 can calculate a higher degree of photographing necessity as the identification degree is lower.

N (d) = − d (Formula 1)

As another example, the addition determination unit 304 may calculate the shooting necessity N (d) by (Expression 2). Here, m is the estimated age of the person. In this way, by adding the estimated age of the person, the lower the estimated age, the higher the degree of photographing required can be calculated. This makes it possible to collect detailed information on events such as lost children that are expected to require a high level of monitoring, so that the accuracy of prior analysis increases.

N (d) = − dm− (Formula 2)

  In step S <b> 805, the additional determination unit 304 determines whether or not re-shooting is necessary by comparing the shooting necessity level with a threshold (determination process). The additional determination unit 304 determines that re-shooting is necessary when the shooting necessity level is equal to or greater than the threshold (Yes in S805), and advances the process to S806. If the shooting necessity level is less than the threshold, the addition determination unit 304 determines that re-shooting is unnecessary (No in S805), and advances the process to S812.

  In this case, in step S812, the table management unit 306 registers the shooting ID of the feature amount extracted in step S802 in the correspondence table 601 in association with the flow line ID of the flow line to be processed. Next, in step S813, the calculation unit 201 cancels this setting when the flow line to be processed is set as an imaging retry target. This completes the imaging control process for one flow line. At this time, if an imaging instruction has already been transmitted from the monitoring server apparatus 101 to the wearable device 109, the monitoring server apparatus 101 captures an image from the wearable device 109 in order to improve the efficiency of the wearer. A cancel instruction is transmitted (transmission process).

On the other hand, in S806, the priority calculation unit 305 calculates the priority. Specifically, the priority calculation unit 305 determines the distance r between the person corresponding to the flow line to be processed and the wearable device 109 based on the detection result of a distance measuring sensor (not shown) of the monitoring server apparatus 101. Is identified. Then, the priority calculation unit 305 calculates the priority by (Equation 3) based on the identification degree d and the distance r.

Priority = 1 / (d × r ² ) (Formula 3)

Thereby, a higher priority can be calculated as the identification degree is smaller and the distance is shorter.

  In step S807, the communication processing unit 307 transmits a shooting instruction including the person to be monitored, the priority, and the shooting conditions to the wearable device 109 (transmission processing). Here, the person to be monitored is a person corresponding to the flow line to be processed. Further, the photographing condition is determined by the calculation unit 201. That is, the calculation unit 201 determines shooting conditions for shooting a person to be monitored based on the flow line of the person to be monitored (condition determination process). Here, the shooting conditions include a shooting position, a shooting direction, a shooting timing, and the like.

  As yet another example, the arithmetic unit 201 determines whether or not the wearer 109 should move and photograph. If the communication processing unit 307 determines that the moving image can be shot with a higher degree of identification, the communication processing unit 307 may transmit a shooting instruction that further includes location information indicating the moving destination to the wearable device 109. In addition, when there are a plurality of wearable devices 109, the communication processing unit 307 may include information for designating the wearable device 109 in the nearest location as a main charge device in order to prevent redundant shooting. Good.

  In contrast, the wearable device 109 receives a plurality of shooting instructions in the shooting control process for each of the plurality of flow lines, and sets the shooting order of each monitoring target person based on the priority included in each shooting instruction. decide. Specifically, wearable device 109 determines the order of descending priority as the imaging order. The wearable device 109 displays a photographing instruction to instruct the wearer in the action instruction display area 704 shown in FIG. 5 according to the priority. The wearable device 109 may display, for example, a point to be photographed and a predicted arrival time of the person to be photographed as a photographing instruction. When the wearable device 109 performs shooting, the wearable device 109 extracts a feature amount from the face image in the captured image, and transmits the extraction result to the monitoring server apparatus 101. Wearable device 109 may cause the wearer to determine whether or not the extraction result is appropriate, and then transmit the extraction result. Thereby, the quality of an extraction result can be improved.

  In response to this, in step S808, the communication processing unit 307 receives the extraction result from the wearable device 109. It is assumed that the extraction result includes the coordinates of the shooting point, the orientation, and the shot image data in addition to the feature amount. The table management unit 306 registers a feature amount indicated in the extraction result in the feature amount table 501 after assigning a new shooting ID.

  As another example, the monitoring server device 101 may extract feature amounts from a face image captured by the wearable device 109. In this case, the wearable device 109 may transmit the captured image data to the monitoring server apparatus 101. If shooting is not performed within a certain time (such as 5 seconds), a time-out process is performed by using past captured image data, and the process proceeds to the next step. In step S <b> 809, the identification level estimation unit 303 estimates the identification level of the feature amount of the face image captured by the wearable device 109. Note that the process of S809 is the same as the process of S803. Next, in S810, S304 calculates the degree of photographing necessity based on the degree of identification estimated in S809. Note that the process of S810 is the same as the process of S804.

  Next, in S811, the addition determination unit 304 advances the process to S814 if the shooting necessity level is greater than or equal to the threshold (Yes in S811). If the shooting necessity level is less than the threshold (No in S811), the addition determination unit 304 advances the process to S812. In this case, in S812, the table management unit 306 registers the shooting ID of the feature amount included in the extraction result received in S808 in the correspondence table 601 in association with the flow line ID, and then advances the process to S813. Note that in the process of associating the flow line and the feature amount performed in S812 and the like, registration in a database engine or the like is assumed, but a search index may be constructed each time or periodically. Thereby, the efficiency of the extraction process of the flow line corresponding to the feature quantity to be monitored can be increased.

  In S814, the table management unit 306 determines the identification degree estimated in S809 and the registered identification degree that is already associated with the person ID of the person corresponding to the flow line to be processed in the feature amount table 501. Compare If the identification degree estimated in S809 is larger (Yes in S814), the table management unit 306 advances the process to S815. If the identification degree estimated in S809 is equal to or lower than the registered identification degree (No in S814), the table management unit 306 advances the process to S816.

  In step S815, the table management unit 306 registers the flow ID of the flow line to be processed in the correspondence table 601 in association with the imaging ID of the feature amount included in the extraction result received in step S808. Thereby, although the degree of discrimination is not sufficient, the feature amount corresponding to the flow line can be registered. In step S <b> 816, the calculation unit 201 sets a flow line to be processed as a subject to be retried for imaging. This completes the imaging control process for one flow line.

  The monitoring server apparatus 101 periodically performs imaging control processing for each flow line, thereby updating and maintaining the feature quantity to be monitored such as the feature quantity linked to each flow line with a higher degree of discrimination. can do. Further, under the control of the monitoring server apparatus 101, the wearable device 109 starts processing for specifying feature amounts in order from the person who needs to specify feature amounts earlier among the recaptured persons who continue to move. can do. Therefore, it is possible to compensate for a shortage of information due to face orientation, blurring, and hiding, and improve monitoring accuracy such as person search. From this, it is possible to improve the accuracy of monitoring and analysis such as lost child and trouble prediction using the monitoring target feature amount.

  Note that, as a modification of the first embodiment, the monitoring server apparatus 101 determines the shooting order of a plurality of persons based on the priority calculated for each flow line, and then includes the shooting order. The instruction may be transmitted to the wearable device 109. Thereby, the process of the wearable device 109 can be reduced.

(Second Embodiment)
Next, a monitoring system according to the second embodiment will be described. In the monitoring system according to the second embodiment, the monitoring server apparatus 101 calculates an expectation level indicating the possibility of shooting in the near future, and controls shooting by the wearable device 109 based on the expectation level. Hereinafter, the difference between the monitoring system according to the second embodiment and the monitoring system according to the first embodiment will be described.

  FIG. 9 is a flowchart illustrating a shooting control process performed by the monitoring server apparatus 101 according to the second embodiment. In addition, the same number is attached | subjected to the process same as each process of the imaging | photography control process demonstrated referring FIG. 8 among each process shown in FIG. In step S805, if the additional determination unit 304 determines that reshooting is necessary when the degree of shooting necessity is greater than or equal to the threshold (Yes in step S805), the process proceeds to step S901. In step S <b> 901, the addition determination unit 304 calculates an expectation level indicating the possibility of shooting (expectation level calculation process). Specifically, the addition determination unit 304 first predicts the route of the person corresponding to the flow line based on the flow line to be processed. Then, the additional determination unit 304 uses the prediction result to quantify whether the predicted route is included in the shootable area of the monitoring camera device 103 or the wearable device 109 by probability calculation or the like, thereby obtaining an expected value. Is calculated.

  In step S902, the addition determination unit 304 compares the expected value calculated in step S901 with a threshold value. If the expected value is equal to or greater than the threshold value (Yes in step S902), the process proceeds to step S814. If the expected value is less than the threshold value (No in S902), the addition determination unit 304 advances the process to S806. If the wearable device 109 is under the shooting instruction, the addition determination unit 304 stops the shooting instruction to the wearable device 109 and advances the process to step S806. Thereby, re-photographing by the wearable device 109 can be performed only when the expected value is low. The remaining configuration and processing of the monitoring system according to the second embodiment are the same as the configuration and processing of the monitoring system according to the first embodiment.

  As described above, the monitoring server apparatus 101 according to the second embodiment does not consider a flow line that can be photographed passively with the passage of time as a re-photographing target. Thereby, it is possible to minimize the shooting instruction to the wearable device 109 and maintain the monitoring accuracy.

  As an example of a change in the monitoring system according to the second embodiment, the monitoring server device 101 sets the priority of shooting so as to focus on shooting when a state where the degree of identification is less than a threshold value continues for a certain period of time. It may be changed to a higher value (change process). As a result, even when there is a monitoring target whose identification degree does not improve even after a certain amount of time has elapsed, it is possible to efficiently improve the identification degree of the entire monitoring target.

(Other embodiments)
As another embodiment, for example, when shooting from a distance, the flow line may be discontinuous due to the person being monitored hidden behind an obstacle or the shadow of another person. At this time, the monitoring server apparatus 101 performs re-measurement of the feature amount when the end / start time and position of the flow line are within the threshold value and the distance of the feature amount to be monitored is within the threshold value. Instead, the flow lines may be connected by determining that they are the same person. In this way, unnecessary remeasurement can be suppressed, and the tracking accuracy of the flow line can be increased.

  In addition, analysis accuracy may deteriorate for passers-by whose identification level is continuously low due to low height or the like. In order to deal with this case, it is also effective to specify the accompanying person to be monitored and associate the accompanying monitoring target with the flow line and the feature amount. Specifically, the monitoring server device 101 analyzes a surrounding flow line for a flow line in which the state of low discrimination continues for a certain time or more, and uses a correlation between the coordinate position and the time, etc. If you want to keep it, associate it as a companion monitoring target. And the monitoring server apparatus 101 extracts the feature-value of accompanying monitoring target, and links and registers the extracted feature-value in connection with the flow line with a lower identification degree. For passers-by who continue to have a low degree of discrimination, it is possible to detect an event such as a lost child in advance and to improve analysis accuracy.

  Further, in the above-described embodiment, the monitoring target feature quantity having a high monitoring target identification level is sequentially linked to the flow line, and thus may fall into a local solution. In order to avoid this, the monitoring server device 101 periodically monitors the feature quantities of the past monitoring targets based on the time, the number of registrations, etc., and the feature quantity so that all the flow lines have a mutually higher degree of discrimination. The table 501 and the correspondence table 601 may be updated.

  In addition, the present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors of a computer of the system or apparatus execute the program. It can also be realized by a process of reading and executing. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

  As mentioned above, according to each embodiment mentioned above, a flow line and a mobile body can be appropriately linked | related, without requiring a troublesome operation | work, preventing the fall of a monitoring system.

  Although the present invention has been described in detail based on preferred embodiments thereof, the present invention is not limited to these specific embodiments, and various forms within the scope of the present invention are also included in the present invention. included. A part of the above-described embodiments may be appropriately combined.

101 monitoring server device 103 monitoring camera device 109 wearable device

Claims (11)

First imaging means for imaging a plurality of moving bodies corresponding to a plurality of flow lines to be monitored;
Feature quantity extraction means for extracting a feature quantity of a target moving object to be processed based on a photographed image obtained by the first photographing means;
Estimating means for estimating the identification degree of the individual identification of the target moving body based on the feature amount of the target moving body;
Priority calculating means for calculating the priority of photographing by the second photographing means of the target moving body based on the identification degree;
A monitoring system comprising: a transmission unit configured to transmit a shooting instruction of the target moving body associated with the priority to the second shooting unit.   2. The monitoring according to claim 1, wherein the priority calculating unit further calculates the priority of the target moving body based on a distance between the target moving body and the second imaging unit. system. A determination unit that determines whether or not the second moving unit needs to re-shoot the target moving body based on the degree of identification corresponding to the target moving body;
When the determination unit determines that the re-shooting of the target moving body is necessary, the priority calculation unit calculates the priority corresponding to the target moving body, and the transmission unit includes the target The monitoring system according to claim 1, wherein the imaging instruction of the moving body is transmitted. Further comprising a condition determining means for determining a shooting condition based on a flow line of the target moving body;
The monitoring system according to any one of claims 1 to 3, wherein the transmission unit transmits the imaging instruction including the imaging condition.   The monitoring system according to claim 3 or 4, wherein the transmission unit transmits an imaging stop instruction to the second imaging unit when the determination unit determines that re-imaging is unnecessary. Based on a flow line corresponding to the target moving body, further comprising an expectation degree calculating means for calculating an expectation degree at which shooting is performed on the target moving body by the second photographing means;
The monitoring system according to claim 3, wherein the determination unit further determines whether or not re-photographing is necessary based on the expectation degree.   The apparatus according to claim 1, further comprising: a changing unit configured to change the priority of the target moving body to a higher value when the state where the identification degree of the target moving body is less than a threshold value continues for a predetermined time or longer. The monitoring system of any one of thru | or 6. The target movement based on a feature amount of a target moving body to be processed, which is extracted from a photographed image including a plurality of moving bodies corresponding to each of a plurality of flow lines to be monitored, which is shot by the first imaging means. An estimation means for estimating the degree of individual body identification;
Priority calculating means for calculating the priority of photographing by the second photographing means of the target moving body based on the identification degree;
A monitoring apparatus comprising: a transmission unit configured to transmit an imaging instruction of the target moving body associated with the priority to the second imaging unit. A shooting control method executed by a surveillance system,
A first imaging step of imaging a plurality of moving bodies corresponding to each of a plurality of flow lines to be monitored;
A feature amount extraction step of extracting a feature amount of a target moving body to be processed based on the captured image obtained in the first imaging step;
An estimation step for estimating a degree of identification of the individual identification of the target moving body based on the feature amount of the target moving body;
A priority calculating step of calculating a priority of photographing by the second photographing means of the target moving body based on the identification degree;
And a transmission step of transmitting an imaging instruction of the target moving body associated with the priority to the second imaging means. An imaging control method executed by a monitoring device,
The target movement based on a feature amount of a target moving body to be processed, which is extracted from a photographed image including a plurality of moving bodies corresponding to each of a plurality of flow lines to be monitored, which is shot by the first imaging means. An estimation step for estimating the degree of individual identification of the body;
A priority calculating step of calculating a priority of photographing by the second photographing means of the target moving body based on the identification degree;
And a transmission step of transmitting an imaging instruction of the target moving body associated with the priority to the second imaging means. Computer
The target movement based on a feature amount of a target moving body to be processed, which is extracted from a photographed image including a plurality of moving bodies corresponding to each of a plurality of flow lines to be monitored, which is shot by the first imaging means. An estimation means for estimating the degree of individual body identification;
Priority calculating means for calculating the priority of photographing by the second photographing means of the target moving body based on the identification degree;
The program for functioning as a transmission means which transmits the imaging | photography instruction | indication of the said target moving body matched with the said priority to a said 2nd imaging | photography means.

Images