WO2016202027A1 - 一种物体移动轨迹识别方法及系统 - Google Patents

一种物体移动轨迹识别方法及系统 Download PDF

Info

Publication number
WO2016202027A1
WO2016202027A1 PCT/CN2016/076731 CN2016076731W WO2016202027A1 WO 2016202027 A1 WO2016202027 A1 WO 2016202027A1 CN 2016076731 W CN2016076731 W CN 2016076731W WO 2016202027 A1 WO2016202027 A1 WO 2016202027A1
Authority
WO
WIPO (PCT)
Prior art keywords
specific target
information
movement trajectory
acquired
determined
Prior art date
Application number
PCT/CN2016/076731
Other languages
English (en)
French (fr)
Inventor
王启贵
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2016202027A1 publication Critical patent/WO2016202027A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/292—Multi-camera tracking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries

Definitions

  • This paper relates to, but is not limited to, the field of computer video recognition and communication technology, and relates to an object moving track recognition method and system.
  • the related art object trajectory recognition method is: using mobile communication combined with Global Positioning System (GPS) technology to monitor the position of moving objects such as cars and ships, and the hardware systems depended on include: GPS satellites, mobile The mobile terminal, mobile communication base station and data processing central office installed on the object; the system can adopt both active and passive working modes; the GPS location information is processed by a single point fitting or route in the data processing central office; the passive mode is Three priorities are set in the system; after the GPS location information of the mobile object is sent to the data processing central office, it is compared and fitted with the pre-stored GPS information, and the fitted data is transmitted to the relevant higher management priority. Third party, the third party performs real-time, non-real-time monitoring or historical retrieval on the moving object as needed.
  • GPS Global Positioning System
  • the above method has the following problem: if the moving object does not carry the mobile terminal, the trajectory of the moving object cannot be tracked; in addition, when the mobile terminal fails or the GPS signal is poor, the object trajectory tracking fails.
  • the related art also recognizes the moving trajectory of the indoor object by: providing a signal according to the label, combining the real time location system (RTLS, Real Time Location Systems) and the monitoring image to obtain the specific position of the monitored object, and then converting the position information into coordinates. Information, and finally uses coordinate information to describe the movement trajectory of the monitored object.
  • RTLS Real Time Location System
  • this method requires labeling the object and cannot be monitored if there is no label on the object.
  • the embodiment of the invention provides a method and a system for recognizing an object movement trajectory, which can solve the problem that the target group cannot track the movement trajectory of the target group if the target group does not carry the corresponding device or label, so that even if the target group does not carry The corresponding device or label can also track the movement of the target group.
  • An embodiment of the present invention provides a method for recognizing an object movement trajectory, comprising: acquiring a multi-angle picture and activity information of an object at a determined location; comparing a similarity between the multi-angle picture of the acquired object and a sample picture of a specific target, If the similarity between the multi-angle picture of the acquired object and the sample picture of the specific target exceeds a set threshold, identifying the object as a specific target; according to the activity information of the object identified as the specific target and the determining The location information of the location tracks the movement trajectory of the particular target.
  • the activity information includes time information, speed information, and moving position information of the object passing through the determined location.
  • tracking the movement trajectory of the specific target according to the activity information of the object identified as the specific target and the location information of the determined location comprises: passing the plurality of determined locations according to the object identified as the specific target The time information in the activity information acquired at the time and the position information of the plurality of determined places, the trajectory of the specific target is depicted in chronological order, and the speed information and the moving direction information in the activity information acquired when passing through the plurality of determined locations are Predict the subsequent movement trajectory.
  • the method before the comparing the similarity between the multi-angle picture of the acquired object and the sample picture of the specific target, the method further includes: determining a sample picture of the specific target that needs to be tracked.
  • the embodiment of the invention further provides an object movement trajectory recognition system, comprising:
  • Camera group module set to obtain multi-angle pictures and activity letters of objects at determined locations interest
  • the image recognition module is configured to compare the similarity between the multi-angle image of the acquired object and the sample image of the specific target, if the similarity between the multi-angle image of the acquired object and the sample image of the specific target exceeds
  • the threshold is determined to identify the object as a specific target
  • the trajectory display module is configured to track the movement trajectory of the specific target according to the activity information of the object identified as the specific target and the position information of the determined location.
  • the activity information includes time information, speed information, and moving position information of the object passing through the determined location.
  • the trajectory display module is configured to: describe time information in the activity information acquired when the object identified as the specific target passes through the plurality of determined locations, and position information of the plurality of determined locations, and describe in chronological order a movement trajectory of the specific target, and predicting a subsequent movement trajectory of the specific target based on the velocity information and the movement orientation information in the activity information acquired when the location is determined.
  • the system further includes: a sample picture management module, configured to determine a sample picture of a specific target that needs to be tracked.
  • a sample picture management module configured to determine a sample picture of a specific target that needs to be tracked.
  • an embodiment of the present invention further provides a computer readable storage medium, where the computer readable storage medium stores computer executable instructions, and the computer executable trajectory recognition method is implemented when the computer executable instructions are executed.
  • the multi-angle picture and the activity information of the object at the determined location are acquired; the similarity between the multi-angle picture of the acquired object and the sample picture of the specific target is compared, and if the multi-angle picture of the acquired object is specific If the similarity of the sample picture of the target exceeds the set threshold, the object is identified as a specific target; the movement trajectory of the specific target is tracked according to the activity information of the object identified as the specific target and the position information of the determined place.
  • the technical solution provided by the embodiment of the present invention can track the movement trajectory of the target group even if the target group does not carry the corresponding device or label, so that the wanted, escaped prisoner, and stolen vehicle in the city can be
  • the effective tracking of the activity track greatly improves the efficiency of pursuing various criminals in the city. It does not need to view the video frame by frame to find suspicious molecules, and realizes automatic tracking, which in turn indicates the best time and the best place to hunt. , making the dog-hunting pursuit into reality, greatly improving the city emergency ability.
  • FIG. 1 is a flowchart of a method for recognizing an object movement trajectory according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of an object movement trajectory recognition system according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of application of an object movement trajectory recognition system according to an embodiment of the present invention.
  • FIG. 1 is a flowchart of a method for recognizing an object movement trajectory according to an embodiment of the present invention. As shown in FIG. 1 , the method for identifying an object movement track provided by an embodiment of the present invention includes the following steps:
  • Step 11 Obtain a multi-angle picture and activity information of the object at the determined location.
  • the activity information includes time information, speed information, and moving position information of the object passing through the determined location.
  • the location is determined to be a location where the camera group is installed (eg, an intersection of a city street), and the camera group includes, for example, four cameras disposed in four relative orientations.
  • the four cameras can take pictures of the object from four different orientations, according to the movement information recorded by the four cameras when the object passes the intersection (for example, shooting at the camera)
  • the distance and time of different positions in the area determine the moving speed of the object through the intersection, and determine the moving position information according to the forward direction of the object.
  • Step 12 Compare the similarity between the multi-angle picture of the acquired object and the sample picture of the specific target. If the similarity between the multi-angle picture of the acquired object and the sample picture of the specific target exceeds the set threshold, the object is identified. For a specific goal.
  • the method further comprises: determining a sample picture of the particular target that needs to be tracked.
  • Step 13 Track the movement trajectory of the specific target according to the activity information of the object identified as the specific target and the position information of the determined location.
  • the step 13 includes: describing the movement of the specific target in chronological order according to the time information in the activity information acquired when the object identified as the specific target passes through the plurality of determined locations and the position information of the plurality of determined locations.
  • the trajectory predicts a subsequent movement trajectory of the specific target based on the velocity information and the movement orientation information in the activity information acquired when the plurality of determined locations are passed.
  • an object movement trajectory recognition system includes: a camera group module configured to acquire a multi-angle picture and activity information of an object at a determined location; and an image recognition module configured to: compare the acquisition The similarity between the multi-angle picture of the object and the sample picture of the specific target, if the similarity between the multi-angle picture of the acquired object and the sample picture of the specific target exceeds the set threshold, the object is identified as a specific target And a trajectory display module configured to track the movement trajectory of the specific target according to the activity information of the object identified as the specific target and the position information of the determined location.
  • the activity information includes time information, speed information, and moving position information of the object passing through the determined location.
  • the trajectory display module is configured to: display time information in the activity information acquired when the object identified as the specific target passes through the plurality of determined locations, and position information of the plurality of determined locations, in chronological order The movement trajectory of the specific target, and predicting the subsequent movement trajectory according to the speed information and the movement orientation information in the activity information acquired when the location is determined.
  • the system further includes: a sample picture management module configured to determine a sample picture of the specific target that needs to be tracked.
  • FIG. 3 is a schematic diagram of application of an object movement trajectory recognition system according to an embodiment of the present invention. The technical solution of this embodiment is described in detail below by taking the application scenario of FIG. 3 as an example.
  • the object movement trajectory recognition system of the present scene includes a camera group module S10, an object S11, a network S12, a background server S13, a management station S14, and a display unit S15.
  • the camera group module S10 can be configured, for example, to set four cameras (such as one of the four ports of the city street intersection, such as the cameras A to D shown in FIG. 3) at a certain location. Through four shots The cooperation of the head can completely reflect the movement orientation information of the object S11 (for example, environmental information including the moving direction and the moving direction, etc.), and recognize the object from different orientations.
  • the location information (eg, longitude and latitude) of the camera group module S10 can be directly associated with the map.
  • the camera group module S10 further includes an object moving speed analysis module and a Geographic Information System (GIS) module, wherein the object moving speed analysis module is configured to be in different positions according to the captured object S11. The distance and time are calculated and the speed of its movement is calculated.
  • GIS Geographic Information System
  • the camera group module S10 stores the activity information (including the shooting time point information, the moving speed information, the moving direction information) of the object S11 and all the captured pictures in the database of the background server S13 via the network S12.
  • the image recognition module is disposed, for example, in the background server S13 for comparing the image of the object stored in the database with the sample image of the specific target, and if the similarity exceeds the set threshold, identifying the object as a specific target, thereby It can be determined that a specific target has appeared at the location of the camera group, and at the same time, a recognition success alarm can be initiated to notify the user.
  • the object is photographed in four directions, and the activity information of the acquired object (for example, moving speed and moving position information) is analyzed.
  • the latitude and longitude of the location of the camera group location is obtained, and the location information is obtained, and the activity information, the location information, and the captured image of the object appearing at the intersection camera group (camera group 1) are stored in the database.
  • the specific target By comparing the captured image of the object stored in the database with the sample image of the specific target, the specific target is recognized in real time, and if the similarity between the two exceeds the set threshold, the recognition is successful, that is, in the camera group 1
  • the object that appears is a specific target, records the information of the object appearing in the camera group 1, and activates an alarm prompt to identify that the object information is the behavior of a specific target, and the address is at the position of the camera group 1.
  • the activity information of the object passing through the intersection is also recorded.
  • Real-time identification if the recognition is successful, that is, the object appearing in the camera group 2 is a specific target, it is also necessary to start an alarm prompt to identify the behavior of the object appearing in the camera group 2 is the behavior of a specific target, the address is in the camera group 2 position.
  • the speed and moving direction of the camera group 1 can be entered according to the specific target, and the moving direction and speed of the camera group 2 are entered, for the specific target.
  • the trajectory of the simulation is simulated, for example, simulating which particular target has passed through several streets and which street is currently heading.
  • the same specific target is recognized at the same time in the same camera group, it means that the movement track of the specific target has a return, and the line whose specific target is folded back from the corresponding street can be simulated by the moving speed and the time interval.
  • sample images of specific targets can be managed as needed. For example, when a sample picture of a specific target is added, the recognition task can be started, the picture taken by the camera group stored in the previous three months is automatically recognized, and the frequent movement of the specific target is automatically drawn on the map. A route trajectory to predict the recent movement trajectory of the particular target.
  • an embodiment of the present invention further provides a computer readable storage medium, where the computer readable storage medium stores computer executable instructions, and the computer executable trajectory recognition method is implemented when the computer executable instructions are executed.
  • each module/unit in the above embodiment may be implemented in the form of hardware, for example, by implementing an integrated circuit to implement its corresponding function, or may be implemented in the form of a software function module, for example, executing a program stored in the memory by a processor. / instruction to achieve its corresponding function.
  • This application is not limited to any specific combination of hardware and software.
  • the movement trajectory of the target group can be tracked, so that the trajectory of the wanted, escaped, and stolen vehicles in the city can be effectively tracked. It has greatly improved the efficiency of pursuing various criminals in the city.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Image Analysis (AREA)
  • Alarm Systems (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

一种物体移动轨迹识别方法，包括：获取经过确定地点的物体的多角度图片及活动信息；比较获取的物体的多角度图片与特定目标的样例图片的相似度，若获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标；根据识别为特定目标的物体的活动信息以及确定地点的位置信息对特定目标的移动轨迹进行跟踪。上述技术方案能够解决现有技术中若目标群体没有携带相应的设备或标签，则无法对目标群体的移动轨迹进行跟踪的问题。

Description

一种物体移动轨迹识别方法及系统 技术领域
本文涉及但不限于计算机视频识别和通信技术领域，涉及一种物体移动轨迹识别方法及系统。
背景技术
目前，随着世界安全形势越来越严峻，城市中的通缉犯、逃跑的犯人、被盗车辆等严重威胁城市公民的生活安全。人们需要对这些特殊类型的目标进行监控，并跟踪预测该类目标的移动轨迹。这需要对该类目标进行识别，并且需要把该类目标所在的位置标识在城市地图上，以方便警方对其进行控制。
相关技术的物体移动轨迹识别方式为：利用移动通信与全球定位系统(GPS，Global Positioning System)技术相结合针对汽车、轮船等移动物体的位置进行监控，所依赖的硬件系统包括：GPS卫星、移动物体上安装的移动终端、移动通信基站及数据处理中心局；系统可采用主动和被动两种工作模式；GPS位置信息在数据处理中心局的处理采用单点拟合或路线拟合；被动模式在系统中设立三个优先级；移动物体的GPS位置信息发送到数据处理中心局后，与预先存储的GPS信息进行比较和拟合，经过拟合的数据传送给相关的具有更高管理优先权的第三者，所述第三者根据需要对移动物体进行实时、非实时监控或历史检索。
然而，上述方式存在以下问题：若移动物体没有携带移动终端，则无法对移动物体的轨迹进行跟踪；另外，当移动终端出现故障或GPS信号较差时，会导致物体轨迹跟踪的失败。
另外，相关技术还采用以下方式识别室内物体移动轨迹：根据标签提供信号，再结合实时定位系统(RTLS，Real Time Location Systems)和监控图像获得被监控对象的具体位置，然后将位置信息转换成坐标信息，最后利用坐标信息描绘出被监控对象的移动轨迹。然而，该方法需要在物体上打上标签，如果物体上没有标签则无法进行监控。
可见，在相关技术中，若目标群体没有携带相应的设备或标签，则无法实现对物体移动轨迹的跟踪。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提供一种物体移动轨迹识别方法及系统，能够解决相关技术中若目标群体没有携带相应的设备或标签则无法对目标群体的移动轨迹进行跟踪的问题，从而实现即使目标群体没有携带相应的设备或标签，也可以对目标群体的移动轨迹进行跟踪。
本发明实施例提供一种物体移动轨迹识别方法，包括：获取经过确定地点的物体的多角度图片及活动信息；比较所述获取的物体的多角度图片与特定目标的样例图片的相似度，若所述获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标；根据所述识别为特定目标的物体的活动信息以及所述确定地点的位置信息对所述特定目标的移动轨迹进行跟踪。
可选地，所述活动信息包括所述物体经过该确定地点的时间信息、速度信息以及移动方位信息。
可选地，根据所述识别为特定目标的物体的活动信息以及所述确定地点的位置信息对所述特定目标的移动轨迹进行跟踪包括：根据所述识别为特定目标的物体经过多个确定地点时获取的活动信息中的时间信息以及多个确定地点的位置信息，按时间先后顺序描绘该特定目标的移动轨迹，并根据经过多个确定地点时获取的活动信息中的速度信息以及移动方位信息预测后续移动轨迹。
可选地，所述比较所述获取的物体的多角度图片与特定目标的样例图片的相似度之前，该方法还包括：确定需要跟踪的特定目标的样例图片。
本发明实施例还提供一种物体移动轨迹识别系统，包括：
摄像头组模块，设置为获取经过确定地点的物体的多角度图片及活动信 息；
图像识别模块，设置为：比较所述获取的物体的多角度图片与特定目标的样例图片的相似度，若所述获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标；
轨迹显示模块，设置为根据所述识别为特定目标的物体的活动信息以及所述确定地点的位置信息对所述特定目标的移动轨迹进行跟踪。
可选地，所述活动信息包括所述物体经过该确定地点的时间信息、速度信息以及移动方位信息。
可选地，所述轨迹显示模块是设置为：根据所述识别为特定目标的物体经过多个确定地点时获取的活动信息中的时间信息以及多个确定地点的位置信息，按时间先后顺序描绘该特定目标的移动轨迹，并根据经过确定地点时获取的活动信息中的速度信息以及移动方位信息预测所述特定目标的后续移动轨迹。
可选地，上述系统还包括：样例图片管理模块，设置为确定需要跟踪的特定目标的样例图片。
此外，本发明实施例还提供一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机可执行指令，所述计算机可执行指令被执行时实现所述物体移动轨迹识别方法。
在本发明实施例中，获取经过确定地点的物体的多角度图片及活动信息；比较获取的物体的多角度图片与特定目标的样例图片的相似度，若获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标；根据识别为特定目标的物体的活动信息以及确定地点的位置信息对特定目标的移动轨迹进行跟踪。如此，通过本发明实施例提供的技术方案，即使目标群体没有携带相应的设备或标签，也可以对目标群体的移动轨迹进行跟踪，从而可以对城市中的通缉犯、逃跑的犯人、被盗车辆的活动轨迹进行有效跟踪，大大提高了对城市内各类犯罪分子的追捕效率，无需一帧一帧查看视频来查找可疑分子，实现了自动跟踪，进而对追捕最佳时间及最佳地点进行指示，使守株待兔式的追捕变为现实，大大提高了城市应急 能力。
在阅读并理解了附图和详细描述后，可以明白其它方面。
附图说明
图1为本发明实施例提供的物体移动轨迹识别方法的流程图；
图2为本发明实施例提供的物体移动轨迹识别系统的示意图；
图3为本发明实施例提供的物体移动轨迹识别系统的应用示意图。
具体实施方式
以下结合附图对本发明的实施例进行详细说明，应当理解，以下所说明的实施例仅用于说明和解释本申请，并不用于限定本申请。
图1为本发明实施例提供的物体移动轨迹识别方法的流程图。如图1所示，本发明实施例提供的物体移动轨迹识别方法包括以下步骤：
步骤11：获取经过确定地点的物体的多角度图片及活动信息。
其中，活动信息包括物体经过确定地点的时间信息、速度信息以及移动方位信息。
于一实施例中，确定地点为安装摄像头组的地点(如城市街道的路口)，摄像头组例如包括设置在四个相对方位的四个摄像头。举例而言，当一物体经过设置摄像头组的路口时，四个摄像头可从四个不同方位对该物体进行拍照，根据该物体在经过该路口时四个摄像头记录的移动信息(例如在摄像头拍摄区域内不同位置的距离和时间)，确定该物体经过该路口的移动速度，并根据该物体的前进方向，确定移动方位信息。
步骤12：比较获取的物体的多角度图片与特定目标的样例图片的相似度，若获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标。
于一实施例中，在步骤12之前，该方法还包括：确定需要跟踪的特定目标的样例图片。
步骤13：根据识别为特定目标的物体的活动信息以及确定地点的位置信息对特定目标的移动轨迹进行跟踪。
于一实施例中，步骤13包括：根据识别为特定目标的物体经过多个确定地点时获取的活动信息中的时间信息以及多个确定地点的位置信息，按时间先后顺序描绘该特定目标的移动轨迹，并根据经过多个确定地点时获取的活动信息中的速度信息以及移动方位信息预测特定目标的后续移动轨迹。
图2为本发明实施例提供的物体移动轨迹识别系统的示意图。如图2所示，本发明实施例提供的物体移动轨迹识别系统包括：摄像头组模块，设置为获取经过确定地点的物体的多角度图片及活动信息；图像识别模块，设置为：比较所述获取的物体的多角度图片与特定目标的样例图片的相似度，若所述获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标；轨迹显示模块，设置为根据所述识别为特定目标的物体的活动信息以及所述确定地点的位置信息对所述特定目标的移动轨迹进行跟踪。
其中，活动信息包括所述物体经过该确定地点的时间信息、速度信息以及移动方位信息。
于一实施例中，轨迹显示模块是设置为：根据所述识别为特定目标的物体经过多个确定地点时获取的活动信息中的时间信息以及多个确定地点的位置信息，按时间先后顺序描绘该特定目标的移动轨迹，并根据经过确定地点时获取的活动信息中的速度信息以及移动方位信息预测后续移动轨迹。
于一实施例中，该系统还包括：样例图片管理模块，设置为确定需要跟踪的特定目标的样例图片。
图3为本发明实施例提供的物体移动轨迹识别系统的应用示意图。下面以图3的应用场景为例，详细说明本实施例的技术方案。
如图3所示，本场景的物体移动轨迹识别系统包括：摄像头组模块S10、物体S11、网络S12、后台服务器S13、管理台S14以及显示部件S15。
摄像头组模块S10，例如可以配置为在一个确定地点设置四个摄像头(如城市街道路口的四个岔口各一个，如图3所示的摄像头A～D)。通过四个摄 像头的配合，能够完整地反映物体S11的移动方位信息(例如，包括移动方向以及移动方向的环境信息等)，并从不同方位对该物体进行识别。另外，摄像头组模块S10的位置信息(如，经度及纬度)可直接关联至地图。于一实施例中，摄像头组模块S10例如还包括物体移动速度分析模块和地理信息系统(GIS，Geographic Information System)模块，其中，物体移动速度分析模块，用于根据拍摄到的物体S11在不同位置的距离和时间，计算得出其移动速度。
摄像头组模块S10通过网络S12，将物体S11的活动信息(包括拍摄时间点信息、移动速度信息、移动方位信息)以及所有拍摄图片存入后台服务器S13的数据库。
图像识别模块例如设置在后台服务器S13内，用于将存储在数据库中的物体的图片与特定目标的样例图片进行比较，若相似度超过设定阀值，则识别该物体为特定目标，从而可确定特定目标曾出现在该摄像头组所在位置，同时，可启动识别成功报警以通知用户。
根据识别出的特定目标出现的摄像头组所在位置，按照时间先后顺序在地图上标识特定目标的移动轨迹，并根据特定目标的移动方位信息以及速度信息预测后续移动轨迹，并显示在显示部件S15(轨迹显示模块)上。
另外，通过管理台S14(样例图片管理模块)，可以录入特定目标的图片特征。
举例而言，通过摄像头组1(至少包括前后左右四个方位的四个摄像头)，分四个方位对物体进行拍照，并分析获取物体的活动信息(例如，移动速度及移动方位信息)，通过摄像头组所在地图位置的经纬度，得出位置信息，将在该路口摄像头组(摄像头组1)出现的物体的活动信息、位置信息及拍摄图片存储到数据库中。通过将存储在数据库中的物体的拍摄图片与特定目标的样例图片进行比对，实时对特定目标进行识别，若两者的相似度超过设定的阀值，则识别成功，即在摄像头组1出现的物体为特定目标，记录在摄像头组1出现的物体的信息，并启动告警提示，标识上述物体信息是某特定目标的行为，地址在摄像头组1所在位置。下一时段，在某路口摄像头组2，采用与摄像头组1相同的方式，同样记录经过该路口的物体的活动信息，同 样进行实时识别，若识别成功，即在摄像头组2出现的物体为特定目标，也需启动告警提示，标识在摄像头组2出现的物体的信息是某特定目标的行为，地址在摄像头组2所在位置。当识别出经过摄像头组1和摄像头组2的物体为同一个特定目标时，可根据该特定目标进入摄像头组1的速度和移动方向，以及进入摄像头组2的移动方向和速度，对该特定目标的移动轨迹进行模拟，例如模拟该特定目标经过了几个街道，目前正在前往哪个街道。
若在同一摄像头组不同时间识别出同一个特定目标，则说明特定目标的移动轨迹有返回，可通过移动速度及相隔时间，模拟出特定目标从相应街道折返的线路。
另外，特定目标的样例图片可根据需要进行管理。举例而言，当新增特定目标的样例图片之后，可启动识别任务，对之前三个月内存储的摄像头组拍摄的图片进行自动识别，并自动在地图上描绘出该特定目标的经常移动路线轨迹，从而对该特定目标近期的移动轨迹进行预测。
此外，本发明实施例还提供一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机可执行指令，所述计算机可执行指令被执行时实现所述物体移动轨迹识别方法。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序来指令相关硬件(例如处理器)完成，所述程序可以存储于计算机可读存储介质中，如只读存储器、磁盘或光盘等。可选地，上述实施例的全部或部分步骤也可以使用一个或多个集成电路来实现。相应地，上述实施例中的各模块/单元可以采用硬件的形式实现，例如通过集成电路来实现其相应功能，也可以采用软件功能模块的形式实现，例如通过处理器执行存储于存储器中的程序/指令来实现其相应功能。本申请不限制于任何特定形式的硬件和软件的结合。
以上显示和描述了本申请的基本原理和主要特征和本申请的优点。本申请不受上述实施例的限制，上述实施例和说明书中描述的只是说明本申请的原理，在不脱离本申请精神和范围的前提下，本申请还会有各种变化和改进，这些变化和改进都落入要求保护的本申请范围内。
工业实用性
通过上述技术方案，即使目标群体没有携带相应的设备或标签，也可以对目标群体的移动轨迹进行跟踪，从而可以对城市中的通缉犯、逃跑的犯人、被盗车辆的活动轨迹进行有效跟踪，大大提高了对城市内各类犯罪分子的追捕效率。

Claims (9)

  1. 一种物体移动轨迹识别方法，包括：
    获取经过确定地点的物体的多角度图片及活动信息；
    比较所述获取的物体的多角度图片与特定目标的样例图片的相似度，若所述获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标；
    根据所述识别为特定目标的物体的活动信息以及所述确定地点的位置信息对所述特定目标的移动轨迹进行跟踪。
  2. 如权利要求1所述的方法，其中，所述活动信息包括所述物体经过该确定地点的时间信息、速度信息以及移动方位信息。
  3. 如权利要求2所述的方法，其中，所述根据所述识别为特定目标的物体的活动信息以及所述确定地点的位置信息对所述特定目标的移动轨迹进行跟踪包括：
    根据所述识别为特定目标的物体经过多个确定地点时获取的活动信息中的时间信息以及多个确定地点的位置信息，按时间先后顺序描绘该特定目标的移动轨迹，并根据经过多个确定地点时获取的活动信息中的速度信息以及移动方位信息预测所述特定目标的后续移动轨迹。
  4. 如权利要求1所述的方法，所述比较所述获取的物体的多角度图片与特定目标的样例图片的相似度之前，所述方法还包括：确定需要跟踪的特定目标的样例图片。
  5. 一种物体移动轨迹识别系统，包括：
    摄像头组模块，设置为获取经过确定地点的物体的多角度图片及活动信息；
    图像识别模块，设置为：比较所述获取的物体的多角度图片与特定目标的样例图片的相似度，若所述获取的物体的多角度图片与特定目标的样例图片的相似度超过设定的阈值，则识别该物体为特定目标；
    轨迹显示模块，设置为根据所述识别为特定目标的物体的活动信息以及 所述确定地点的位置信息对所述特定目标的移动轨迹进行跟踪。
  6. 如权利要求5所述的系统，其中，所述活动信息包括所述物体经过该确定地点的时间信息、速度信息以及移动方位信息。
  7. 如权利要求6所述的系统，其中，所述轨迹显示模块是设置为：根据所述识别为特定目标的物体经过多个确定地点时获取的活动信息中的时间信息以及多个确定地点的位置信息，按时间先后顺序描绘该特定目标的移动轨迹，并根据经过确定地点时获取的活动信息中的速度信息以及移动方位信息预测所述特定目标的后续移动轨迹。
  8. 如权利要求5所述的系统，所述系统还包括：样例图片管理模块，设置为确定需要跟踪的特定目标的样例图片。
  9. 一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机可执行指令，所述计算机可执行指令被执行时实现权利要求1～4任一项所述的方法。
PCT/CN2016/076731 2015-06-18 2016-03-18 一种物体移动轨迹识别方法及系统 WO2016202027A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510342334.1A CN106326240A (zh) 2015-06-18 2015-06-18 一种物体移动轨迹识别方法及系统
CN201510342334.1 2015-06-18

Publications (1)

Publication Number Publication Date
WO2016202027A1 true WO2016202027A1 (zh) 2016-12-22

Family

ID=57544971

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/076731 WO2016202027A1 (zh) 2015-06-18 2016-03-18 一种物体移动轨迹识别方法及系统

Country Status (2)

Country Link
CN (1) CN106326240A (zh)
WO (1) WO2016202027A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109614897A (zh) * 2018-11-29 2019-04-12 平安科技（深圳）有限公司 一种室内查找物品的方法及终端
CN110147471A (zh) * 2019-04-04 2019-08-20 平安科技（深圳）有限公司 基于视频的轨迹跟踪方法、装置、计算机设备及存储介质
CN111126807A (zh) * 2019-12-12 2020-05-08 浙江大华技术股份有限公司 行程切分方法和装置、存储介质及电子装置
CN111723835A (zh) * 2019-03-21 2020-09-29 北京嘀嘀无限科技发展有限公司 车辆移动轨迹区分方法、装置和电子设备
CN113515982A (zh) * 2020-05-22 2021-10-19 阿里巴巴集团控股有限公司 轨迹还原方法及设备、设备管理方法及管理设备
CN114331662A (zh) * 2022-03-11 2022-04-12 支付宝(杭州)信息技术有限公司 识别相同业务主体的方法及装置
CN114584746A (zh) * 2022-04-29 2022-06-03 深圳市边海物联科技有限公司 一种安防监控系统及安防监控方法
CN115394025A (zh) * 2021-05-20 2022-11-25 中国移动通信集团有限公司 监控方法、装置、电子设备及存储介质

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109740462B (zh) * 2018-12-21 2020-10-27 北京智行者科技有限公司 目标的识别跟随方法
CN109960264A (zh) * 2019-03-28 2019-07-02 潍柴动力股份有限公司 一种目标识别方法及系统
CN110430524A (zh) * 2019-06-24 2019-11-08 深圳云天励飞技术有限公司 一种基于gps定位的告警方法及装置
CN112348545A (zh) * 2019-08-09 2021-02-09 上海红星美凯龙悦家互联网科技有限公司 用户信息获取及关联系统、装置和存储介质
CN112348544A (zh) * 2019-08-09 2021-02-09 上海红星美凯龙悦家互联网科技有限公司 用户信息获取及关联方法、装置、设备、系统和存储介质
CN110728617A (zh) * 2019-09-30 2020-01-24 上海电机学院 一种基于fpga的动态目标识别与实时跟踪系统
CN110599776A (zh) * 2019-10-15 2019-12-20 福州市协成智慧科技有限公司 一种交通数据处理系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101207638A (zh) * 2007-12-03 2008-06-25 浙江树人大学 一种基于预测的无线传感器网络目标跟踪方法
US20110135150A1 (en) * 2009-12-08 2011-06-09 Texas Instruments Incorporated Method and apparatus for tracking objects across images
CN103020607A (zh) * 2012-12-27 2013-04-03 Tcl集团股份有限公司 一种人脸识别方法及装置
CN103942811A (zh) * 2013-01-21 2014-07-23 中国电信股份有限公司 分布式并行确定特征目标运动轨迹的方法与系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102843547B (zh) * 2012-08-01 2014-01-08 安科智慧城市技术（中国）有限公司 一种嫌疑目标智能追踪方法和系统
CN104034316B (zh) * 2013-03-06 2018-02-06 深圳先进技术研究院 一种基于视频分析的空间定位方法
CN104539909A (zh) * 2015-01-15 2015-04-22 安徽大学 一种视频监控方法及视频监控服务器

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101207638A (zh) * 2007-12-03 2008-06-25 浙江树人大学 一种基于预测的无线传感器网络目标跟踪方法
US20110135150A1 (en) * 2009-12-08 2011-06-09 Texas Instruments Incorporated Method and apparatus for tracking objects across images
CN103020607A (zh) * 2012-12-27 2013-04-03 Tcl集团股份有限公司 一种人脸识别方法及装置
CN103942811A (zh) * 2013-01-21 2014-07-23 中国电信股份有限公司 分布式并行确定特征目标运动轨迹的方法与系统

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109614897A (zh) * 2018-11-29 2019-04-12 平安科技（深圳）有限公司 一种室内查找物品的方法及终端
CN111723835A (zh) * 2019-03-21 2020-09-29 北京嘀嘀无限科技发展有限公司 车辆移动轨迹区分方法、装置和电子设备
CN110147471A (zh) * 2019-04-04 2019-08-20 平安科技（深圳）有限公司 基于视频的轨迹跟踪方法、装置、计算机设备及存储介质
CN111126807A (zh) * 2019-12-12 2020-05-08 浙江大华技术股份有限公司 行程切分方法和装置、存储介质及电子装置
CN111126807B (zh) * 2019-12-12 2023-10-10 浙江大华技术股份有限公司 行程切分方法和装置、存储介质及电子装置
CN113515982A (zh) * 2020-05-22 2021-10-19 阿里巴巴集团控股有限公司 轨迹还原方法及设备、设备管理方法及管理设备
CN113515982B (zh) * 2020-05-22 2022-06-14 阿里巴巴集团控股有限公司 轨迹还原方法及设备、设备管理方法及管理设备
CN115394025A (zh) * 2021-05-20 2022-11-25 中国移动通信集团有限公司 监控方法、装置、电子设备及存储介质
CN114331662A (zh) * 2022-03-11 2022-04-12 支付宝(杭州)信息技术有限公司 识别相同业务主体的方法及装置
CN114584746A (zh) * 2022-04-29 2022-06-03 深圳市边海物联科技有限公司 一种安防监控系统及安防监控方法

Also Published As

Publication number Publication date
CN106326240A (zh) 2017-01-11

Similar Documents

Publication Publication Date Title
WO2016202027A1 (zh) 一种物体移动轨迹识别方法及系统
WO2020224375A1 (zh) 定位方法、装置、设备和计算机可读存储介质
CN102436738B (zh) 一种基于无人机的交通监测装置
JP6815262B2 (ja) 交通違反検知装置、システム、交通違反検知方法およびプログラム
KR100533033B1 (ko) 디지털 영상 처리 기술을 이용한 위치 추적 시스템 및 방법
Liu et al. A Vision‐Based Target Detection, Tracking, and Positioning Algorithm for Unmanned Aerial Vehicle
CN102595103B (zh) 一种基于gis地图推演智能视频的方法
US20150009327A1 (en) Image capture device for moving vehicles
CN107645653A (zh) 一种摄像机跟踪拍摄的方法、装置、设备及存储介质
JP6013923B2 (ja) ビデオエピソードの閲覧及び検索のためのシステム及び方法
JP2023516502A (ja) 画像ベースの場所決定及び駐車モニタリングのためのシステム及び方法
KR20130123092A (ko) 실내 네비게이션 시스템 및 방법
CN106295598A (zh) 一种跨摄像头目标跟踪方法及装置
KR101678004B1 (ko) 노드-링크 기반 카메라 네트워크 통합 감시 시스템 및 감시 방법
US11657623B2 (en) Traffic information providing method and device, and computer program stored in medium in order to execute method
US10896513B2 (en) Method and apparatus for surveillance using location-tracking imaging devices
US10388132B2 (en) Systems and methods for surveillance-assisted patrol
CN111024061A (zh) 一种导航方法、装置、设备和介质
US11100656B2 (en) Methods circuits devices systems and functionally associated machine executable instructions for image acquisition identification localization and subject tracking
RU2693926C1 (ru) Система контроля и воздействия на объекты, представляющие интерес, и выполняемые ими процессы и соответствующий способ
CN109345567B (zh) 物体运动轨迹识别方法、装置、设备和存储介质
CN112365520B (zh) 一种基于视频大数据资源效能测评的行人目标实时追踪系统及方法
CN115424465B (zh) 停车场地图的构建方法、装置及存储介质
CN108398133A (zh) 一种导航方法、装置及系统
US20190286876A1 (en) On-Demand Outdoor Image Based Location Tracking Platform

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16810768

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16810768

Country of ref document: EP

Kind code of ref document: A1