CN109963113B - Method and device for monitoring interested target - Google Patents
Method and device for monitoring interested target Download PDFInfo
- Publication number
- CN109963113B CN109963113B CN201711340292.3A CN201711340292A CN109963113B CN 109963113 B CN109963113 B CN 109963113B CN 201711340292 A CN201711340292 A CN 201711340292A CN 109963113 B CN109963113 B CN 109963113B
- Authority
- CN
- China
- Prior art keywords
- information
- target
- monitoring
- wireless equipment
- wireless device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Alarm Systems (AREA)
Abstract
本发明公开了一种感兴趣目标的监控方法及装置，通过视觉检测到的身份特征与采集的无线设备信息与感兴趣目标名单进行比对，将抓拍的视频帧、目标的身份特征、无线设备信息比对结果绑定为对象包发送到后端服务器；后端服务器接收到监控摄像机上报的对象包后，将对象包存储至监控摄像机对应的存储目录中，根据对象包中包含的信息，在关联监控点存储的对象包中进行重复无线设备信息查找，以确定感兴趣目标及其对应的无线设备信息，并产生对应的告警信息。本发明基于多种身份特征来确定感兴趣目标及其无线设备信息的置信度，只要检测到一次脸部特征，就能进行感兴趣目标检测和无线设备信息查找，所受人流密度影响小，有更高的准确性和稳定性。
The invention discloses a monitoring method and device for a target of interest. By comparing visually detected identity features with collected wireless device information and a list of interested targets, the captured video frames, the target's identity features, and the wireless device are compared with each other. The information comparison result is bound as an object package and sent to the back-end server; after receiving the object package reported by the surveillance camera, the back-end server stores the object package in the storage directory corresponding to the surveillance camera. Repeated wireless device information search is performed in the object package stored in the associated monitoring point to determine the target of interest and its corresponding wireless device information, and generate corresponding alarm information. The invention determines the confidence level of the target of interest and its wireless device information based on a variety of identity features. As long as the facial feature is detected once, the target of interest detection and wireless device information search can be performed, and the impact of the crowd density is small. Higher accuracy and stability.
Description
Technical Field
The invention belongs to the technical field of video monitoring, and particularly relates to a monitoring method and device for an interested target.
Background
With the development of video monitoring technology, tracking and monitoring an interested target has gradually become a standard function in the security field. In an actual application scene, a camera is generally configured to detect and snapshot people in a monitoring scene, then comparison analysis is performed on a preset target through a rear-end server (or the detection is directly realized on the camera), and once an interested target is captured, a series of linkage schemes such as subsequent continuous tracking and the like can be triggered.
However, the actual monitoring scene is complex, such as building occlusion, dense people flow and the like, and brings certain difficulties for finding the interested target and continuously tracking the interested target through video analysis, and some insurmountable difficulties exist for finding the interested target and continuously tracking the interested target through video analysis.
In the prior art, a monitoring camera is provided with a WIFI sniffer, the WIFI sniffer can sniff wireless signals from the air, and terminal information (such as mobile phone MAC information) within a specific range can be collected. When the target person passes through the first image acquisition area, sensing and recording the wireless terminal; when the target person passes through the second image acquisition area, sensing and recording the wireless terminal, and screening out the wireless terminal which is overlapped with the first image acquisition area; and by analogy, seamless tracking can be carried out through the wireless terminal in the subsequent tracking and positioning process until the target personnel correspond to the wireless terminal one by one.
However, in an actual monitoring scene, it is still difficult to completely recognize and capture an interested target for many times through a visual signal technology, and it is difficult to completely recognize the same target at least twice or more, which causes a common problem of being unable to recognize the target face when the target face is shielded by other objects, resulting in an unstable detection output efficiency. In addition, the wireless terminal device cannot be effectively monitored and tracked when the target is replaced.
Disclosure of Invention
The invention aims to provide a method and a device for monitoring an interested target, which are used for overcoming the problems of inaccurate target identification, unstable output efficiency and incapability of tracking and monitoring caused by mobile phone exchange in the prior art.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a monitoring method of an interested target is applied to a video monitoring system, the video monitoring system comprises a monitoring camera and a back-end server, and the monitoring method of the interested target comprises the following steps:
the monitoring camera periodically collects video frames and wireless equipment information in a monitoring scene, performs target detection analysis on the collected video frames, and extracts the identity characteristics of a target;
judging whether a suspicious target exists according to the identity characteristics of the target, if so, capturing the suspicious target, binding the captured video frame, the identity characteristics of the target and the wireless equipment information in the monitoring scene into an object packet and sending the object packet to a back-end server;
comparing the identity characteristics of the target and the wireless equipment information in the monitoring scene with a stored interested target list, judging whether the interested target exists or not, if so, capturing the interested target, binding the captured video frame, the identity characteristics of the target and the wireless equipment information comparison result into an object packet and sending the object packet to a back-end server;
and after receiving the object packet reported by the monitoring camera, the back-end server stores the object packet into a storage directory corresponding to the monitoring camera, and repeatedly searches for the wireless equipment information in the object packet stored in the associated monitoring point according to the information contained in the object packet so as to determine the interested target and the wireless equipment information corresponding to the interested target and generate corresponding alarm information.
Further, if the object corresponding to the object packet is a suspicious object, the repeating wireless device information search is performed in the object packet stored by the associated monitoring point according to the information contained in the object packet to determine the object of interest and the wireless device information corresponding to the object of interest, and generate corresponding alarm information, including:
according to the identity characteristics in the object package, searching repeated wireless equipment information in an interested target object package stored in the associated monitoring point, and setting the confidence corresponding to the wireless equipment information according to the repeated times;
and when the confidence of the wireless equipment information in the interested target object packet reaches or exceeds a set threshold, generating alarm information to prompt that the wireless equipment information corresponding to the interested target is detected.
Further, if the object corresponding to the object packet is the interested object, the interested object list includes the identity feature of the interested object, the comparison result of the wireless device information is the wireless device information in the monitoring scene, the repeated wireless device information search is performed in the object packet stored in the associated monitoring point according to the information included in the object packet to determine the interested object and the wireless device information corresponding to the interested object, and generate the corresponding alarm information, including:
according to the identity characteristics in the object packet, repeated wireless equipment information searching is carried out in an interested target object packet or a suspicious target object packet stored in the associated monitoring point, and the confidence corresponding to the wireless equipment information is set according to the repeated times;
and when the confidence of the wireless equipment information in the object packet reaches or exceeds a set threshold, generating alarm information to prompt that the wireless equipment information corresponding to the interested target is bound.
Further, if the target corresponding to the object packet is the target of interest, the target of interest list includes the identity of the target of interest and the wireless device information bound to the target of interest, and the comparison result of the wireless device information is a result of intersecting the wireless device information in the monitoring scene with the wireless device information bound to the target of interest, including:
when the intersection is empty, carrying a mark for detecting the wireless equipment information again in the object packet, packaging all the wireless equipment information in the monitoring scene into the object packet and reporting the object packet to a back-end server;
and when the intersection is not empty, packaging the wireless equipment information in the intersection into an object packet and reporting the wireless equipment information to a back-end server, carrying a matched mark in the object packet when the number of the wireless equipment information in the intersection is equal to that of the wireless equipment information bound by the interested target, and carrying a mark for updating the wireless equipment information of the interested target in the object packet when the number of the wireless equipment information in the intersection is less than that of the wireless equipment information bound by the interested target.
Further, when the object packet carries the wireless device information flag that needs to be re-detected, the repeated wireless device information search is performed in the object packet stored in the associated monitoring point according to the information contained in the object packet to determine the object of interest and the wireless device information corresponding to the object of interest, and generate corresponding alarm information, including:
according to the identity characteristics in the object packet, repeated wireless equipment information searching is carried out in an interested target object packet or a suspicious target object packet stored in the associated monitoring point, and the confidence corresponding to the wireless equipment information is set according to the repeated times;
and when the confidence of the wireless equipment information in the object packet reaches or exceeds a set threshold, generating alarm information to prompt that the wireless equipment information corresponding to the interested target is bound.
Further, when the object packet carries the matched flag, the repeated wireless device information search is performed in the object packet stored in the associated monitoring point according to the information contained in the object packet to determine the interested target and the wireless device information corresponding to the interested target, and generate the corresponding alarm information, including:
and directly generating alarm information to prompt the detection of the interested target.
Further, when the object packet carries information of the wireless device of the object of interest to be updated, according to the information contained in the object packet, the repeated wireless device information search is performed in the object packet stored in the associated monitoring point to determine the object of interest and the wireless device information corresponding to the object of interest, and generate corresponding alarm information, including:
and generating alarm information to prompt that the interested target is detected, updating the wireless equipment information bound in the interested target list, and issuing the information to the monitoring camera.
Further, after receiving the object packet, the back-end server deletes the object packet stored for a time exceeding T.
The invention also provides a monitoring device of the interested target, which is applied to the monitoring camera in the video monitoring system and comprises a processor and a nonvolatile memory which stores a plurality of computer instructions, wherein the computer instructions are executed by the processor to realize the steps executed by the monitoring camera in the method.
The invention also provides a monitoring device of the interested target, which is applied to a back-end server in a video monitoring system and comprises a processor and a nonvolatile memory for storing a plurality of computer instructions, wherein when the computer instructions are executed by the processor, the steps executed by the back-end server in the method are realized.
According to the monitoring method and device for the interested target, the identity characteristics detected visually are compared with the acquired wireless equipment information and the interested target list, different operations are performed according to the comparison result, and the user can be detected and prompted at the first time and is triggered to detect again under the condition that the wireless equipment is frequently replaced. And through the cooperative comparison with the associated monitoring points, the wireless equipment information of the interested target is quickly determined. The invention can acquire the wireless equipment information of the interested target by utilizing the existing monitoring camera without resetting the monitoring point. The method and the device determine the confidence degrees of the interested target and the wireless equipment information thereof based on various identity characteristics, can detect the interested target and search the wireless equipment information as long as the facial characteristics are detected once, and have small influence on people stream density and higher accuracy and stability.
Drawings
FIG. 1 is a diagram of an associated monitoring point according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for monitoring an object of interest according to the present invention.
Detailed Description
The technical solutions of the present invention are further described in detail below with reference to the drawings and examples, which should not be construed as limiting the present invention.
The general design idea of the technical scheme is that multiple points are continuously distributed and controlled in the possible traveling direction of personnel at related monitoring points (such as crossroads, subway exit related road sections and the like) by utilizing the fixity of multiple 'identity characteristic' information (such as human face characteristics, clothing characteristics, mobile phone MAC and the like) of an interested target within a specific time range (such as minutes).
As shown in fig. 1, the associated monitoring point is an associated camera on a possible historical travel track of the target, and if the camera 1 is a subway exit camera and the travel direction of the target is determined to be that the target exits from a subway exit, no historical travel track associated camera exists; when camera number 2 detects an object and the direction of travel is from south to north, then cameras number 1, 3, 4, 8, 9, 10, 11, 12, 13 are the associated cameras on their possible historical travel trajectories, defining them as the associated monitoring points for camera number 2. The associated monitoring point may be configured for each monitoring camera by a person, which is not described in detail below.
As shown in fig. 2, a monitoring method for an object of interest in this embodiment includes:
the monitoring camera periodically collects video frames and wireless equipment information in a monitoring scene, performs target detection analysis on the collected video frames, and extracts the identity characteristics of a target;
judging whether a suspicious target exists according to the identity characteristics of the target, if so, capturing the suspicious target, binding the captured video frame, the identity characteristics of the target and the wireless equipment information in the monitoring scene into an object packet and sending the object packet to a back-end server;
comparing the identity characteristics of the target and the wireless equipment information in the monitoring scene with a stored interested target list, judging whether the interested target exists or not, if so, capturing the interested target, binding the captured video frame, the identity characteristics of the target and the wireless equipment information comparison result into an object packet and sending the object packet to a back-end server;
and after receiving the object packet reported by the monitoring camera, the back-end server stores the object packet into a storage directory corresponding to the monitoring camera, and repeatedly searches for the wireless equipment information in the object packet stored in the associated monitoring point according to the information contained in the object packet so as to determine the interested target and the wireless equipment information corresponding to the interested target and generate corresponding alarm information.
Specifically, the surveillance camera acquires video images of a surveillance scene in real time, typically 25 frames per second. The monitoring camera of the embodiment further integrates a WIFI sniffer, and can periodically collect information of the wireless device in the monitoring scene, for example, MAC address information, which is described below by taking a MAC address as an example.
The WIFI sniffer automatically detects wireless signals within a certain range, determines the wireless signal intensity range (without special accuracy) within the range according to the effective range of pedestrian detection and analysis of each monitoring camera in the early stage of installation, takes the range as the filtering condition after subsequent normal operation, collects the MAC address of wireless equipment within the range, and records the detected wireless equipment information meeting the range into an MAC information table.
The surveillance camera itself may also perform target detection analysis on the acquired video frames, and is responsible for identifying and intelligently analyzing the video frames, and returning an analysis result, for example, identifying the facial features, the gender features, the height features, the clothing features and other identity features of the pedestrian, which is not limited to the specific content of the identity features in this embodiment.
Since the facial features are easily not easily detected due to occlusion or facing away from the camera, other assist features (gender features, height features, clothing features) are easily collected and do not change over a period of time, and can be captured multiple times. Therefore, in order to avoid the situation of being unrecognizable due to occlusion, the identity feature at least needs to include two items of content, such as a facial feature and a clothing feature, and further, may further include a gender feature, a height feature, and the like. The embodiment utilizes various identity characteristics to determine the interested target, and adds auxiliary characteristics (such as clothing characteristics, sex characteristics, height characteristics and the like) which do not change in a short time to determine the interested target, so that the information of the subsequent detection wireless device can be quickly converged to the minimum range. For any interested target in the list, the wireless device carrying the wireless device passes through the key monitoring area, the wireless device characteristic information (MAC) most relevant to the interested target can be detected and output in a full-automatic manner, and the complicated and repeated manual intervention operation of inputting time, place and the like for each interested target is avoided.
In this embodiment, the monitoring camera stores information of an interested target list, where the interested target list includes an identity characteristic of an interested target and further includes information of a wireless device corresponding to the interested target. The list of objects of interest includes only the identity feature when the wireless device information has not been bound initially. After the wireless device information is subsequently bound, the interested target list further includes the wireless device information corresponding to the interested target, so that each monitoring point can further converge the range or correct the range conveniently through real-time detection.
For example: a typical list of objects of interest is as follows:
| object of interest | Facial features | Characteristic of clothes | Wireless device information |
| Object 1 | Feature 1, feature 2 | |
MAC1、 |
| Object | |||
| 2 | |
|
MAC3 |
TABLE 1
When the name list of the interested target only comprises the identity characteristics of the interested target, when the identity characteristics are compared, for example, if the face characteristics of the target are found to be consistent with the face characteristics of a certain interested target in the interested list, the target can be judged to be the interested target, at the moment, the target is captured, the captured video frame, the identity characteristics of the target and the wireless equipment information in the monitoring scene are bound to be an object packet and sent to a back-end server, namely, the comparison result of the wireless equipment information is the wireless equipment information in all the monitoring scenes at the moment.
When the interested target list includes the identity of the interested target and the wireless device information corresponding to the interested target, the wireless device information is also compared during comparison, which is as follows.
At this time, the identity characteristics are first compared to determine whether there is an object of interest. For example, if the facial features of the object are found to be consistent with the facial features of an object of interest in the interest list, the object may be determined to be the object of interest. When the target of interest is determined, the following comparison is further made:
the method comprises the steps that intersection is taken between wireless equipment information bound to an interested target and wireless equipment information in a monitoring scene;
when the intersection is empty, carrying a mark for detecting the wireless equipment information again in the object packet, packaging all the wireless equipment information in the monitoring scene into the object packet and reporting the object packet to a back-end server;
and when the intersection is not empty, packaging the wireless equipment information in the intersection into an object packet and reporting the wireless equipment information to a back-end server, carrying a matched mark in the object packet when the number of the wireless equipment information in the intersection is equal to that of the wireless equipment information bound by the interested target, and carrying a mark for updating the wireless equipment information of the interested target in the object packet when the number of the wireless equipment information in the intersection is less than that of the wireless equipment information bound by the interested target.
It should be noted that after comparing with the interested target list and determining that the current target is the interested target, the object package sent to the back-end server is the interested target object package.
For example, for object 1, MAC1 and MAC2 in the intersection set indicate that the wireless device information corresponding to the current object of interest is consistent with the list of objects of interest, and the object packet carries the already matched flag. If the wireless device information in the surveillance scene collected by the surveillance camera for object 1 contains only MAC1, then the intersection MAC1 is packaged into an object package and reported to the back-end server. And because the number of the intersections is less than the number (2) of the wireless device information bound in the interested target list, the object packet carries a mark for updating the interested target wireless device information.
In the embodiment, the matched mark, the wireless device information mark which needs to be re-detected or the wireless device information mark which needs to be updated are carried in the object packet, so that tracking and monitoring can be directly performed after the interested target is matched, a user can be prompted when the condition that the wireless device information of the interested target with certain counterreconnaissance capability is frequently changed is detected at the first time, and re-detection can be automatically triggered according to user configuration.
In this embodiment, a target that is too low in head or cannot capture the front face and cannot be effectively subjected to face recognition is used as a suspicious target. Because the probability of capturing the front face for multiple times is low, sometimes, whether the front face is the target of interest cannot be determined, the suspicious target is also subjected to subsequent detection, the probability that the auxiliary feature can be captured is high, and the output efficiency and the stability of the output efficiency are improved by means of the auxiliary feature and the associated monitoring point. If the suspicious target exists, the suspicious target is captured, and the captured video frame, the identity characteristics of the target and the wireless device information in the monitoring scene are bound into an object packet and sent to a back-end server. And sending the object packet corresponding to the suspicious target to a back-end server, and storing the object packet as the suspicious target object packet.
In this embodiment, after receiving the object packet, the back-end server stores the object packet in the storage directory corresponding to the monitoring camera. Only reserving the object packages within the latest T minutes in the storage directory corresponding to the monitoring camera, and setting T by a user according to the monitoring density (for example, three-level detection, the farthest average walking time consumption from a first-level camera to a third-level camera is taken as T); and deleting the object packet stored for the time exceeding T so as to ensure that the source data of the MAC extraction mechanism uses effective data in the latest time, and avoid the loss of performance or storage resources caused by traversing or storing unnecessary massive source data.
The following describes the processing procedure of the back-end server according to several situations of reporting the object packet by the monitoring camera.
In the first case, the object package corresponding to the suspicious object.
At this time, the object packet does not carry any mark, and for this situation, the back-end server needs to perform comparative analysis with the interested target object packet stored by the associated monitoring point according to the identity characteristics and the wireless device information.
The analysis process comprises the following steps:
after receiving the object packet reported by the monitoring camera, the back-end server stores the object packet into a storage directory corresponding to the monitoring camera, performs repeated wireless equipment information search in an interested target object packet stored in the associated monitoring point according to the identity characteristics in the object packet, and sets the confidence corresponding to the wireless equipment information according to the repeated times;
and when the confidence of the wireless equipment information in the interested target object packet reaches or exceeds a set threshold value, the back-end server generates alarm information to prompt that the wireless equipment information corresponding to the interested target is detected.
For example, the identity of the object 1 includes facial features and clothing features, and the facial features of the identity of the suspicious object are not consistent with the object of interest and do not need to be compared. And comparing the clothes characteristics, if object packets of the interested objects with consistent clothes characteristics are found, further comparing whether the MAC addresses of the wireless device information are repeated, and if so, increasing the confidence of the repeated MAC in the object packets of the interested objects. The confidence increase may be repeated once, adding 1 to the confidence. And when the confidence coefficient reaches or exceeds a set threshold value, the suspicious target is considered to be corresponding to a certain interested target, and warning information is sent out to prompt a user to detect the information of the wireless equipment corresponding to the interested target.
If the associated monitoring point has no object packet of the interested target, or the clothing characteristics are inconsistent, or no repeated MAC exists, the confidence of the wireless device information in the object packet is kept at an initial value, for example, 1, no alarm is generated, which means that the suspicious target does not correspond to any interested target, and the suspicious target is not considered.
The second case is an object bag of the object of interest, but not carrying any markers.
The situation corresponds to a situation that the wireless device information corresponding to the interested target does not exist in the interested target list, and the comparison result of the wireless device information is the wireless device information in all the monitoring scenes and is contained in the object packet.
For such a situation, the backend server needs to perform comparative analysis with the interested target object package or the suspicious target object package stored in the associated monitoring point according to the identity characteristics and the wireless device information. It will be readily appreciated that the comparative analysis with the target object of interest package stored in association with the monitoring point has been described in the first case and will not be described here in detail.
For example, the identity feature of the object 1 includes a facial feature and a clothing feature, and unlike the first case, the repeated wireless device information extraction is performed by first using the facial feature as an index and associating the existing object packet of interest under the monitoring point, and when one object packet with the repeated wireless device information is found, the confidence of the corresponding wireless device information is increased by 1. And if the wireless equipment information of the interested target still cannot be determined through the facial features, repeated wireless equipment information extraction can be continuously carried out with the existing interested target object packet under the associated monitoring point by taking the clothing features as indexes, and when an object packet with repeated wireless equipment information is found, the confidence coefficient of the corresponding wireless equipment information is added by 1.
And when the confidence coefficient reaches or exceeds a set threshold value, generating alarm information to prompt that the corresponding wireless equipment information is bound for the interested target. At this time, the corresponding wireless device information is found for the interested target, and the found wireless device information can be bound to the interested target, put into the interested target list and sent to the monitoring camera.
In the third case, the object packet of the object of interest carries the information tag of the wireless device to be re-detected.
When the intersection corresponding to the situation is empty, the object packet carries a mark for detecting the wireless equipment information again, and the wireless equipment information in the monitoring scene is completely packaged into the object packet and reported to the back-end server.
For such a situation, the back-end server needs to perform comparative analysis with the interested target object packet stored by the associated monitoring point according to the identity characteristics and the wireless device information, and re-detect the wireless device information of the interested target, and the specific detection method is the same as that in the second situation, and is not repeated here.
The fourth case, an object bag of the object of interest, carries the already matched flags.
This situation corresponds to the case where the number of pieces of wireless device information in the intersection is equal to the number of pieces of wireless device information bound to the object of interest when the intersection is not empty. When the matched mark is carried in the object packet, the interested target and the wireless equipment are represented to be bound, and in the interested target list, other operations are not needed, and the alarm information is directly generated to prompt the interested target to be detected.
In the fifth case, the object packet of the object of interest carries the wireless device information flag of the object of interest to be updated.
This situation is the case when the intersection is not empty, and the number of pieces of wireless device information in the intersection is smaller than the number of pieces of wireless device information bound to the object of interest. When the object packet carries the wireless equipment information mark of the target of interest to be updated, the back-end server generates alarm information to prompt that the target of interest is detected, updates the wireless equipment information bound in the list of the target of interest, and sends the wireless equipment information to the monitoring camera.
It is easy to understand that, when the object packet is the object packet of interest, when the object packet corresponds to the object packet being the object packet of interest, the object packet is a, and when the related monitoring point does not have the object packet corresponding to the object of interest that is the same as the object packet a, the related monitoring point performs the comparative analysis with the suspicious object packet, which is not described herein again.
In the above situation, when the MAC confidence in the object packet reaches or exceeds the upper limit level set by the user, the backend server considers that the determination is finished, and may generate an alarm message to prompt the user, so as to facilitate subsequent further tracking and other related operations. The confidence of each piece of wireless device information in the object packet in this embodiment is initially 1, and if only one MAC exists in the wireless device information in the monitoring scene, the confidence is set to the highest level.
It should be noted that there may be a plurality of MACs whose confidence levels reach the upper limit level set by the user, or the wireless terminal may be replaced by a subsequent interested target, in order to solve such problems, this embodiment issues these wireless device information MACs corresponding to the interested target to each monitoring point, and the monitoring point updates the MAC information to the list information of the interested target to be bound with the corresponding target. Therefore, after the interested target is shot, the MAC addresses are compared, different marks are respectively taken, and corresponding operation is carried out according to the marks.
After the rear-end server sends the alarm information, the user can perform subsequent operations aiming at various alarm information, for example, when an interested target is prompted to be detected, track monitoring is started according to the information of the specified wireless equipment, and the like; when the information of the wireless equipment of the interested target is detected, the information of the wireless equipment of the interested target is bound with the interested target, and the information is put into a list of the interested target and is sent to a monitoring camera and the like. The technical scheme is not limited to this, and the user can perform subsequent arrangement according to different alarm information, which is not described herein again.
Correspondingly to the foregoing method, an embodiment of the present technical solution further provides a monitoring apparatus for an object of interest, which is applied to a monitoring camera in a video monitoring system, and includes a processor and a nonvolatile memory storing a plurality of computer instructions, where the computer instructions, when executed by the processor, implement the steps executed by the monitoring camera in the foregoing method.
Correspondingly to the foregoing method, another embodiment of the present technical solution further provides a monitoring apparatus for an object of interest, which is applied to a backend server in a video monitoring system, and includes a processor and a nonvolatile memory storing a plurality of computer instructions, where the computer instructions, when executed by the processor, implement the steps executed by the backend server in the foregoing method.
The non-volatile Memory may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. Wherein the non-volatile memory is configured to store computer instructions that are executed by the processor.
The processor may be an integrated circuit chip having data processing capabilities. The Processor may be a general-purpose Processor including a Central Processing Unit (CPU), a Network Processor (NP), and the like. The various methods, steps and logic blocks disclosed in embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and those skilled in the art can make various corresponding changes and modifications according to the present invention without departing from the spirit and the essence of the present invention, but these corresponding changes and modifications should fall within the protection scope of the appended claims.
Claims (9)
1. A monitoring method of an interested target is applied to a video monitoring system, the video monitoring system comprises a monitoring camera and a back-end server, and the monitoring method of the interested target is characterized by comprising the following steps:
the monitoring camera periodically collects video frames and wireless equipment information in a monitoring scene, performs target detection analysis on the collected video frames, and extracts the identity characteristics of a target;
judging whether a suspicious target exists according to the identity characteristics of the target, if so, capturing the suspicious target, binding the captured video frame, the identity characteristics of the target and the wireless equipment information in the monitoring scene into an object packet and sending the object packet to a back-end server;
comparing the identity characteristics of the target and the wireless equipment information in the monitoring scene with a stored interested target list, judging whether the interested target exists or not, if so, capturing the interested target, binding the captured video frame, the identity characteristics of the target and the wireless equipment information comparison result into an object packet and sending the object packet to a back-end server;
after receiving an object packet reported by a monitoring camera, a back-end server stores the object packet into a storage directory corresponding to the monitoring camera, repeatedly searches for wireless equipment information in the object packet stored in an associated monitoring point according to the information contained in the object packet, sets a confidence corresponding to the wireless equipment information according to the number of times of repetition, determines an interested target and the wireless equipment information corresponding to the interested target when the confidence of the wireless equipment information in the object packet reaches or exceeds a set threshold, and generates corresponding alarm information; the associated monitoring point is a monitoring camera associated with the monitoring camera;
when the target corresponding to the object package is the interested target, the comparison result of the wireless device information is a result of intersection of the wireless device information in the monitoring scene and the wireless device information bound to the interested target, the interested target is snapshot, and the snapshot video frame, the identity feature of the target and the comparison result of the wireless device information are bound as the object package and sent to the back-end server, and the method comprises the following steps:
when the intersection is empty, carrying a mark for detecting the wireless equipment information again in the object packet, packaging all the wireless equipment information in the monitoring scene into the object packet and reporting the object packet to a back-end server;
and when the intersection is not empty, packaging the wireless equipment information in the intersection into an object packet and reporting the wireless equipment information to a back-end server, carrying a matched mark in the object packet when the number of the wireless equipment information in the intersection is equal to that of the wireless equipment information bound by the interested target, and carrying a mark for updating the wireless equipment information of the interested target in the object packet when the number of the wireless equipment information in the intersection is less than that of the wireless equipment information bound by the interested target.
2. The method for monitoring an object of interest according to claim 1, wherein if the object corresponding to the object packet is a suspicious object, the repeating wireless device information search is performed in the object packet stored in the associated monitoring point according to information contained in the object packet to determine the object of interest and the wireless device information corresponding thereto, and generating corresponding alarm information, including:
according to the identity characteristics in the object package, searching repeated wireless equipment information in an interested target object package stored in the associated monitoring point, and setting the confidence corresponding to the wireless equipment information according to the repeated times;
and when the confidence of the wireless equipment information in the interested target object packet reaches or exceeds a set threshold, generating alarm information to prompt that the wireless equipment information corresponding to the interested target is detected.
3. The method for monitoring an object of interest according to claim 1, wherein if the object corresponding to the object package is the object of interest, the list of objects of interest includes an identity of the object of interest, the comparison result of the wireless device information is wireless device information in the monitoring scene, and the repeated wireless device information search is performed in the object package stored in the associated monitoring point according to information included in the object package to determine the object of interest and the wireless device information corresponding to the object of interest, and generate corresponding warning information, including:
according to the identity characteristics in the object packet, repeated wireless equipment information searching is carried out in an interested target object packet or a suspicious target object packet stored in the associated monitoring point, and the confidence corresponding to the wireless equipment information is set according to the repeated times;
and when the confidence of the wireless equipment information in the object packet reaches or exceeds a set threshold, generating alarm information to prompt that the wireless equipment information corresponding to the interested target is bound.
4. The method for monitoring an object of interest according to claim 1, wherein when the object packet carries the tag of information of the wireless device to be re-detected, the repeating of the wireless device information search is performed in the object packet stored in the associated monitoring point according to the information contained in the object packet to determine the object of interest and the wireless device information corresponding thereto, and generate the corresponding alarm information, including:
according to the identity characteristics in the object packet, repeated wireless equipment information searching is carried out in an interested target object packet or a suspicious target object packet stored in the associated monitoring point, and the confidence corresponding to the wireless equipment information is set according to the repeated times;
and when the confidence of the wireless equipment information in the object packet reaches or exceeds a set threshold, generating alarm information to prompt that the wireless equipment information corresponding to the interested target is bound.
5. The method for monitoring an object of interest according to claim 1, wherein when the object packet carries the matched flag, the repeating wireless device information search is performed in the object packet stored in the associated monitoring point according to the information contained in the object packet to determine the object of interest and the wireless device information corresponding thereto, and generate the corresponding alarm information, including:
and directly generating alarm information to prompt the detection of the interested target.
6. The method for monitoring an object of interest according to claim 1, wherein when the object packet carries information of a wireless device of the object of interest to be updated, the repeating of the wireless device information search is performed in the object packet stored in the associated monitoring point according to the information contained in the object packet to determine the object of interest and the wireless device information corresponding thereto and generate corresponding alarm information, including:
and generating alarm information to prompt that the interested target is detected, updating the wireless equipment information bound in the interested target list, and issuing the information to the monitoring camera.
7. The method for monitoring an object of interest according to claim 1, wherein the backend server further deletes the object packets stored for more than T time after receiving the object packets.
8. Monitoring device of an object of interest for use in a monitoring camera in a video monitoring system, comprising a processor and a non-volatile memory in which computer instructions are stored, wherein the computer instructions, when executed by the processor, perform the steps performed by the monitoring camera in the method according to any one of claims 1 to 7.
9. Monitoring device of an object of interest, applied to a backend server in a video monitoring system, comprising a processor and a non-volatile memory storing computer instructions, wherein the computer instructions, when executed by the processor, implement the steps performed by the backend server in the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711340292.3A CN109963113B (en) | 2017-12-14 | 2017-12-14 | Method and device for monitoring interested target |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711340292.3A CN109963113B (en) | 2017-12-14 | 2017-12-14 | Method and device for monitoring interested target |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109963113A CN109963113A (en) | 2019-07-02 |
| CN109963113B true CN109963113B (en) | 2021-06-22 |
Family
ID=67018147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711340292.3A Active CN109963113B (en) | 2017-12-14 | 2017-12-14 | Method and device for monitoring interested target |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109963113B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110705476A (en) * | 2019-09-30 | 2020-01-17 | 深圳市商汤科技有限公司 | Data analysis method and device, electronic equipment and computer storage medium |
| CN111274432B (en) * | 2020-02-06 | 2023-05-09 | 浙江大华技术股份有限公司 | Method and device for processing cloth control |
| CN111784896B (en) * | 2020-06-17 | 2021-02-23 | 深圳南亿科技股份有限公司 | Access control monitoring image storage method, system and storage medium |
| CN112532934B (en) * | 2020-11-23 | 2022-11-15 | 国网山东省电力公司利津县供电公司 | Multi-dimensional cooperative monitoring system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015130744A1 (en) * | 2014-02-28 | 2015-09-03 | Tyco Fire & Security Gmbh | Correlation of sensory inputs to identify unauthorized persons |
| CN105279496A (en) * | 2015-10-26 | 2016-01-27 | 浙江宇视科技有限公司 | Human face recognition method and apparatus |
| CN105357480A (en) * | 2015-11-10 | 2016-02-24 | 杭州敦崇科技股份有限公司 | Public place wireless internet access security management system and operation method thereof |
| CN105790955A (en) * | 2016-04-06 | 2016-07-20 | 深圳市博康智能信息技术有限公司 | Method and system for associating MAC addresses with face information |
| CN106303400A (en) * | 2015-05-12 | 2017-01-04 | 杭州海康威视数字技术股份有限公司 | The monitoring method and apparatus of vehicle-mounted vidicon |
| CN106548148A (en) * | 2016-10-26 | 2017-03-29 | 北京意泰物联软件科技有限公司 | Method and system for identifying unknown face in video |
| CN106899827A (en) * | 2015-12-17 | 2017-06-27 | 杭州海康威视数字技术股份有限公司 | Image data acquiring, inquiry, video frequency monitoring method, equipment and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8184154B2 (en) * | 2006-02-27 | 2012-05-22 | Texas Instruments Incorporated | Video surveillance correlating detected moving objects and RF signals |
| US20130216107A1 (en) * | 2012-02-20 | 2013-08-22 | Chih-Hsung Huang | Method of surveillance by face recognition |
-
2017
- 2017-12-14 CN CN201711340292.3A patent/CN109963113B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015130744A1 (en) * | 2014-02-28 | 2015-09-03 | Tyco Fire & Security Gmbh | Correlation of sensory inputs to identify unauthorized persons |
| CN106303400A (en) * | 2015-05-12 | 2017-01-04 | 杭州海康威视数字技术股份有限公司 | The monitoring method and apparatus of vehicle-mounted vidicon |
| CN105279496A (en) * | 2015-10-26 | 2016-01-27 | 浙江宇视科技有限公司 | Human face recognition method and apparatus |
| CN105357480A (en) * | 2015-11-10 | 2016-02-24 | 杭州敦崇科技股份有限公司 | Public place wireless internet access security management system and operation method thereof |
| CN106899827A (en) * | 2015-12-17 | 2017-06-27 | 杭州海康威视数字技术股份有限公司 | Image data acquiring, inquiry, video frequency monitoring method, equipment and system |
| CN105790955A (en) * | 2016-04-06 | 2016-07-20 | 深圳市博康智能信息技术有限公司 | Method and system for associating MAC addresses with face information |
| CN106548148A (en) * | 2016-10-26 | 2017-03-29 | 北京意泰物联软件科技有限公司 | Method and system for identifying unknown face in video |
Non-Patent Citations (1)
| Title |
|---|
| 视频信息与定位信息融合技术研究;王冬海;《中国电子科学研究院学报》;20140220;第9卷(第1期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109963113A (en) | 2019-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107590439B (en) | Target person identification and tracking method and device based on surveillance video | |
| CN108062349B (en) | Video surveillance method and system based on video structured data and deep learning | |
| CN109446942B (en) | Target tracking method, device and system | |
| CN109963113B (en) | Method and device for monitoring interested target | |
| CN108038176B (en) | Method and device for establishing passerby library, electronic equipment and medium | |
| CN108540752B (en) | Method, device and system for identifying target object in video monitoring | |
| CN104317918B (en) | Abnormal behaviour analysis and warning system based on compound big data GIS | |
| CN109426785B (en) | Human body target identity recognition method and device | |
| CN110706247B (en) | Target tracking method, device and system | |
| US8266174B2 (en) | Behavior history retrieval apparatus and behavior history retrieval method | |
| CN109377694B (en) | Monitoring method and system for community vehicles | |
| CN105279496B (en) | Method and device for face recognition | |
| CN111814510B (en) | Method and device for detecting legacy host | |
| CN102170563A (en) | Intelligent person capture system and person monitoring management method | |
| CN108960124A (en) | The image processing method and device identified again for pedestrian | |
| CN113920585A (en) | Behavior recognition method and device, equipment and storage medium | |
| JP5758165B2 (en) | Article detection device and stationary person detection device | |
| KR20190047748A (en) | Method for object of image and apparatus for the same | |
| CN111291129B (en) | Target person tracking method and device based on multidimensional data research and judgment | |
| CN114187666B (en) | Identification method and system for watching mobile phone while walking | |
| CN113450236B (en) | Suspicious person identification method, device, system and medium based on spatio-temporal data | |
| CN113449563B (en) | Personnel tracking and marking method, device, electronic device and storage medium | |
| TWI515667B (en) | Vehicle recognition and detection system, vehicle information collection method, vehicle information detection method and vehicle information inquiry method | |
| KR102410244B1 (en) | A method for detecting danger by analazing obtained image and an electronic device for the method | |
| CN113936124A (en) | Personnel identity identification method and device based on big data collision detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |