TW201822070A - Object tracking method - Google Patents

Abstract

An object tracking method is applied to an object tracking system and includes the following steps. A plurality of monitor point is defined at a physical area. Every monitor point includes at least one camera to capture a road image. One of the monitor points is defined an initial point according to a position signal which is related to the physical area. At least one first point chosen from the monitor points is defined. Each first point has a neighboring relation with the initial point. Every road image of the first point is determined whether captures an object under track. The first point which captures the object are defined the next initial point.

Description

Object tracking method

The invention relates to an object tracking method, in particular to an object tracking method for predicting an object moving direction and selecting a road image taken by a monitoring point for identification.

With the development of network systems and the advancement of image recognition, there are already many places where a large number of monitors are installed on the streets to investigate traffic accidents or investigate criminal cases. However, although the large number of monitors can increase the scope of surveillance, and avoid monitoring the dead angles, it is impossible to accurately grasp the occurrence of traffic accidents and criminal cases. However, the large amount of data of the monitors makes the search screen data more cumbersome and time consuming.

In general, for example, when a robbery occurs on a street, the police must monitor all the monitors in the vicinity of the area to screen out the information related to the case. This method is not only inefficient, but it is inevitable that there will be omissions in identifying whether the picture is related to the case. In addition, if it is an immediate burglary, if the surveillance system can help screen the surveillance screen more quickly and help the police to catch the gangsters more immediately, it can also increase the chances of catching the gangsters and improve the efficiency of the crime detection.

The present invention is directed to an object tracking method for solving the problem of the efficiency of screening effective picture data from a plurality of monitoring screens in the prior art.

The object tracking method disclosed in the present invention is applicable to an object tracking system, including defining a plurality of monitoring points in a physical geographic area. Each monitoring point has at least one photographic device to take a road picture. According to the position signal, select a monitoring point to define as the initial monitoring point. The location signal is related to the physical geographic area. At least one first priority point is defined, each first priority point is selected from one of the monitoring points, and the location in the physical geographic area has an adjacent relationship with the location of the initial monitoring point. It is judged whether the road picture of each of the first priority points has an object to be searched. When the road picture has an object to be searched, the first priority point for capturing the object to be searched is defined as the next initial monitoring point.

According to the object tracking method disclosed in the above, the object tracking system sets the monitoring point for searching for the object to be searched as the initial monitoring point, and defines the first priority point according to the initial monitoring point as the moving direction of the object to be searched. . When the road picture taken at any one of the first priority points recognizes the object to be searched, the first priority point of the object to be searched is defined as the next initial monitoring point. Thereby, the monitoring screen of the next position of the object to be searched can be predicted and recognized according to the position of the object to be searched. In this way, the object tracking system does not need to identify all the monitoring images in the area at one time, so that the efficiency of the identification image is improved, and the manner of predicting the moving position of the object in advance can also assist the police in tracking the objects to be searched.

The above description of the disclosure and the following description of the embodiments of the present invention are intended to illustrate and explain the spirit and principles of the invention, and to provide further explanation of the scope of the invention.

The detailed features and advantages of the present invention are set forth in the Detailed Description of the Detailed Description of the <RTIgt; </ RTI> <RTIgt; </ RTI> </ RTI> </ RTI> <RTIgt; The objects and advantages associated with the present invention can be readily understood by those skilled in the art. The following examples are intended to describe the present invention in further detail, but are not intended to limit the scope of the invention.

Referring to FIG. 1 to FIG. 4, FIG. 1 is a functional block diagram of an object tracking system according to an embodiment of the invention, and FIG. 2 is a flow chart of an object tracking method according to an embodiment of the invention. FIG. 3 is a schematic diagram of a physical geographic area according to an embodiment of the invention. FIG. 4 is a schematic diagram of a monitoring point setting photographic apparatus according to an embodiment of the invention. As shown in the figure, the object tracking method is applicable to an object tracking system. The object tracking system 10 includes the image analyzing device 101, the path establishing device 103, and the plurality of capturing devices 105, but is not limited thereto. Those skilled in the art can add, for example, a notification device or other suitable device according to actual needs, which is not limited in this embodiment.

In step S11, the object tracking system defines a plurality of monitoring points A to N in the physical geographic area 2. As shown in FIG. 3, the physical geographical area 2 has a plurality of road sections 21, which are defined, for example, by one block 23, for example, a crossroad system is formed by four road sections 21, and a T-shaped road system is intersected by three road sections 21. to make. Each of the monitoring points A to N is defined between at least two road segments 21, for example, at the intersection of the four road segments 21 at the intersection, or at the intersection of the three road segments 21 at the T-junction. In the present embodiment, the definition of the monitoring points A to N can be said to set the position of the photographing device 105 at the monitoring points A to N. Each of the monitoring points A to N is provided with one or more photographing devices 105 for taking a road image on the road segment 21.

As shown in FIG. 4, FIG. 4 shows a schematic diagram in which four imaging devices 105a to 105d are provided at one monitoring point. The imaging devices 105a to 105d respectively take road images in different shooting directions, for example, toward the driving direction or toward the direction of the incoming vehicle. Shooting is not limited in this embodiment. When one monitoring device is provided with four imaging devices 105a to 105d, each of the imaging devices captures a road image, and four monitoring roads are transmitted to the image analyzing device for selective image recognition or storage. For convenience of description, the following embodiments are described by taking the shooting direction as an example, but are not intended to limit the embodiment.

As shown in FIG. 3, two monitoring points are connected by one road section 21 or a plurality of road sections 21, for example, a monitoring section E and a monitoring point F are connected by one road section 21, and between the monitoring point J and the monitoring point L are two. The sections 21 are connected. In other words, when the two monitoring points are connected by a plurality of road segments, the road segments 21 connected between the two monitoring points are not restricted to extend in the same direction, and the connected road segments 21 between the two monitoring points may have an included angle. In addition, in this embodiment, only two monitoring points having adjacent relationships are connected by the link 21, and there are no other monitoring points between the monitoring points of the two adjacent relationships, for example, the monitoring point E and the monitoring point F have In the adjacent relationship, the monitoring point E and the monitoring point I are connected by the link 21, and there is a monitoring point F therebetween, so the monitoring point E and the monitoring point I do not have an adjacent relationship. The definition of the link 21 is only for the convenience of defining the monitoring points A to N, and is not intended to limit the actual situation of the physical geographical area.

In step S13, the path construction device 103 of the object tracking system 10 selects a monitoring point as an initial monitoring point according to the position signal. The location signal is related to the physical geographic area 2. In other words, the location signal is, for example, an address set, such as the address at which the event occurred, the address found by the object 3, or other information suitable to provide the location or location of the object 3. The path establishing device 103 searches for the link 21 corresponding to the physical geographical area 2 according to the address, and selects a monitoring point as the initial monitoring point according to the corresponding road segment 21. For example, in FIG. 3, depending on the position of the object to be searched, the path construction device 103 can select the monitoring point H as the initial monitoring point, and can also select the monitoring point E as the initial monitoring point. In one embodiment, the path establishing device 103 selects the closest monitoring point according to the position signal as the initial monitoring point.

In step S15, the path establishing device 103 defines at least one first priority point according to the initial monitoring point, wherein the first priority point is selected from one of the monitoring points A to N, and the first priority point is located in the physical geographical area. 2 has an adjacent relationship with the position of the initial monitoring point. In the previous step, the monitoring point H is defined as the initial monitoring point, and the monitoring points having the adjacent relationship with the monitoring point H are the monitoring point G, the monitoring point E, and the monitoring point I. After defining the monitoring point H as the initial monitoring point, the path establishing device 103 defines the monitoring point G, the monitoring point E, and the monitoring point I as the first priority point based on the monitoring point adjacent to the monitoring point H.

In step S17, the image analyzing device 101 determines whether the road image of each of the first priority points has the object 3 to be searched. In other words, the image analysis device 101 receives the road maps taken by the monitoring point G, the monitoring point E, and the monitoring point I, and recognizes whether or not the object to be searched 3 is present in the road image captured by the monitoring point G, the monitoring point E, and the monitoring point I. When the object to be searched for 3 is a vehicle, the image analyzing device 101 recognizes, for example, the license plate number of the vehicle to confirm whether or not the license plate number to be searched is in the road image. This embodiment does not limit the feature of the object to be searched by the image analyzing device 101, and determines whether the road image of each of the first priority points has the object to be searched for 3.

In step S19, when the road picture has the object 3 to be searched, the first priority point of the object 3 to be searched is defined as the next initial monitoring point. For example, in FIG. 3, the image analyzing device 101 recognizes that the road image captured by the monitoring point E has the object 3 to be searched. Therefore, the monitoring point E is defined as the next initial monitoring point, and steps S15 to S19 are repeated, that is, the monitoring point B, the monitoring point D, the monitoring point F, and the monitoring point H having the adjacent relationship with the monitoring point E are defined. It is the first priority point, and it is judged whether the monitoring point defined as the first priority point has the object 3 to be searched.

The object tracking system 10 tracks the object by step-by-step definition of the initial monitoring point, and predicts the moving direction of the object by defining the first priority point, so that the image analyzing device 101 can not analyze each of the imaging devices 105 in an area. In the captured road image, the image analyzing device 101 only needs to recognize whether or not the object to be searched for is captured by the monitoring point defined as the first priority point. Thereby, the amount of image data to be processed by the object tracking system 10 is effectively reduced, thereby achieving the purpose of improving identification efficiency and handling efficiency.

In an embodiment, after the monitoring point of the object 3 to be searched is defined as the initial monitoring point, in step S15, the first priority point is further defined according to the shooting direction of the photographing device 105 that captures the object to be searched. In other words, in the foregoing embodiment, when the monitoring point H has three photographing devices that respectively photograph the road map toward the monitoring point G, the monitoring point E, and the monitoring point I, the plurality of photographing devices of the monitoring point are configured to be oriented toward Search for the direction of the object 3 to shoot. Therefore, after the monitoring point H is defined as the initial monitoring point, the photographing device of the object to be searched 3 is photographed according to the monitoring point H, and the moving direction of the object to be searched 3 is obtained to move toward the monitoring point E. Therefore, the path establishing device 103 can set the monitoring point E as the first priority point according to the moving direction of the object to be searched 3, thereby further reducing the amount of image data to be processed by the image analyzing device 101.

In addition, in addition to further reducing the amount of image data to be processed by the image analyzing device 101, the path establishing device 103 defines the first priority point by determining the moving direction of the object to be searched, and can also strengthen the determination of the object to be searched. Whether to leave the link 21 between the monitoring point H and the monitoring point E from the other three sections 21 of the monitoring point E. When the object to be searched 3 does not leave the section 21 between the monitoring point H and the monitoring point E from the other three sections 21 of the monitoring point E, the section 21 indicating that the object 3 to be searched stays between the monitoring point H and the monitoring point E . At this time, the path establishing device 103 can further notify the police to the section 21 between the monitoring point H and the monitoring point E through the additional notification device. In one embodiment, the path building device 103 defines the initial monitoring point step by step to track the path that the object can construct the object 6 to be moved. The route for the object 6 to be moved can be further distributed to the online police network by the notification device in a wireless manner without interruption, and the route for the object to be searched by the police 6 is provided.

In an embodiment, when the monitoring point defined as the first priority point in step S15 does not capture the road image, for example, the photography device of the monitoring point is damaged, the file is damaged, the transmission path is damaged, or other photography may cause the monitoring point. The reason why the device did not capture the road screen. The path establishing device 103 may further define a plurality of second priority points according to the first priority point in which the road picture is not captured. The second priority point is also selected from one of the monitoring points A to N, and the position of the second priority point has an adjacent relationship with the position of the first priority point in the physical geographical area 2.

For example, in step S15, when the monitoring point E is defined as the first priority point, but the monitoring point E does not capture the road picture, the monitoring point D, the monitoring point B, and the monitoring having the adjacent relationship with the monitoring point E are Point F will be selected as the second priority. Then, in step S17, the image analyzing apparatus 101 determines whether the road image of each of the second priority points has the object to be searched 3, in addition to determining whether the road image of each of the first priority points has the object to be searched for 3. When the photographing device defined as the second priority point photographs the object to be searched for 3, the second priority point can also be defined as the next initial monitoring point, and the flow of predicting the moving direction of the object to be searched 3 is performed.

Similarly, when the monitoring point defined as the second priority point does not capture the road image, the path establishing device 103 may further define a plurality of third priority points according to the second priority point of the road image not captured. The third priority point is also selected from one of the monitoring points A to N, and the position of the third priority point has an adjacent relationship with the position of the second priority point in the physical geographical area 2. Those having ordinary knowledge in the technical field can design according to actual conditions, and the present embodiment will not be described again.

In one embodiment, the road image of the object to be searched for the first priority point is collected and stored in a data storage space to be used as one of the evidence materials after completing the task of tracking the object to be searched for. Referring to FIG. 5, it can also be used to analyze the moving path of the object to be searched. In other words, the object tracking system 10 may further include a data storage space, such as a memory or other suitable device, in addition to the image analysis device 101, the path construction device 103, and the plurality of imaging devices 105.

Next, please refer to FIG. 2 and FIG. 5 to FIG. 8 together. FIG. 5 is a flow chart of steps of an object tracking method according to another embodiment of the present invention. FIG. 6 is a drawing according to another embodiment of the present invention. FIG. 7 and FIG. 8 are schematic diagrams showing physical geographic regions according to the embodiment of FIG. 6 according to the present invention. As shown, in step S401, the object tracking system 10 defines a plurality of monitoring points A to N in the physical geographic area 5. The physical geographical area has a plurality of road sections 51. The road section 51 is defined, for example, by one block 53. For example, the crossroad system is formed by the intersection of four road sections 51, and the T-shaped roadway is formed by the intersection of three road sections 51. Each of the monitoring points A to N is defined between at least two road segments 51, and each of the monitoring points A to N is a position at which at least one photographing device 105 is provided to capture a road image on the road segment 51. Two monitoring points are connected by one road segment 51 or a plurality of road segments 51, and two monitoring points having adjacent relationships are connected by a road segment 51 without other monitoring points.

In step S403, the path establishing device 103 selects a monitoring point as an initial monitoring point according to the position signal. The location signal is related to the physical geographic area 5. In other words, the path establishing device 103 searches for the link 51 corresponding to the physical geographical area 5 according to the position signal, and selects a monitoring point as the initial monitoring point according to the corresponding road segment 51. In step S405, the path establishing means 103 defines at least one first priority point according to the initial monitoring point. The first priority point is selected from the monitoring points A to N, and the position of the first priority point has an adjacent relationship with the position of the initial monitoring point in the physical geographical area 5. In this embodiment, after the monitoring point of the object 6 to be searched is defined as the initial monitoring point, in step S405, the first priority point is further defined according to the moving direction of the object to be searched. In a practical example, as shown in FIG. 6, when the monitoring point C is the first priority point, the monitoring point C captures that the object to be searched 6 moves to the monitoring point J, thus defining the monitoring point J, the monitoring point I, and the monitoring Point M is the first priority point.

In step S407, the image analyzing device 101 determines whether the road image of each of the first priority points has the object 6 to be searched. For example, the image analyzing device 101 recognizes whether there is an object to be searched 6 in the road image captured by the monitoring point J, the monitoring point I, and the monitoring point M, for example, the license plate number of the object to be searched 6 is used to determine each of the first priority points. Whether the road picture has an object to be searched for 6. When the road picture has the object 6 to be searched, in step S409, the first priority point of the object to be searched 6 is defined as the next initial monitoring point, and after step S409, steps S401 to S407 are repeated to continue the prediction. The process of moving the object 6 to be moved.

When the road picture does not have the object 6 to be searched, the object to be searched 6 does not leave the monitoring point C and the monitoring point J, the monitoring point I, and the monitoring point M from the monitoring point J, the monitoring point I, and the other three sections 51 of the monitoring point M. When the road segment 51 is between, it indicates that the object to be searched 6 may stay on the road segment 51 between the monitoring point C and the monitoring point J, the monitoring point I and the monitoring point M, for example, as shown in FIG. At this time, the path establishing device 103 can further notify the police to the road segment 51 between the monitoring point C and the monitoring point J, the monitoring point I and the monitoring point M through the additionally added notification device. In one embodiment, the path establishing device 103 can further construct a route for the object 6 to be moved, and wirelessly distribute the route of the object to be searched 6 to the online police network through the notification device to provide the police. The route to which the object 6 is to be moved. On the other hand, the item to be searched 6 may have replaced the road section 51 between the monitoring point C and the monitoring point J, the monitoring point I and the monitoring point M.

Therefore, in step S411, at least one search feature associated with the object to be searched 6, such as the brand model, color or other suitable search feature of the item 6 to be searched, is defined. In step S413, based on the time point at which the monitoring point is defined as the initial monitoring point, the history of the road screen of the initial monitoring point and the road screen of the first priority point is captured. In other words, in the present embodiment, since the initial monitoring point is the monitoring point C, the image analyzing device 101 captures the history of the road map by the monitoring point C, the monitoring point J, the monitoring point I, and the monitoring point M.

In step S415, a suspicious object is defined according to the captured history. That is, the image analyzing device 101 searches for the object having the search feature from the history record by the search feature of the object to be searched 6, and the path establishing device 103 defines the object having the search feature as a suspicious object, and treats the suspicious object as a suspicious object. As the object to be searched 6, the search is started with the license plate number of the suspicious object. In step S417, it is determined whether the road image of each of the first priority points and the road image of the initial monitoring point have suspicious objects, and when any of the road map of the first priority point and the road screen of the initial monitoring point has a suspicious object , the first priority point for defining the suspicious object is the next initial monitoring point. For example, in FIG. 8, the road screen of the monitoring point M has a suspicious object, and the monitoring point M is defined as the next initial monitoring point.

Specifically, the suspicious object is treated as an object to be searched, and the original object to be searched 6 is tracked by the same but independent event. For example, when the first suspicious object and the second suspicious object are defined in step S415. The first suspicious object is photographed at the monitoring point C, at which time the monitoring point C is defined as an initial monitoring point, and the monitoring point having the adjacent relationship with the monitoring point C is used as the first priority point to track the first suspicious object. The second suspicious object is captured at the monitoring point M. At this time, the monitoring point M is defined as an initial monitoring point, and the monitoring point having the adjacent relationship with the monitoring point M is used as the first priority point to track the second suspicious object. This will not be repeated here.

In summary, the embodiment of the present invention provides an object tracking method, by setting a monitoring point for searching for an object to be searched as an initial monitoring point, defining a first priority point according to an initial monitoring point, and predicting a moving direction of the object to be searched. Identifying the road image in the moving direction of the object to be searched and reducing the amount of data for image recognition by the object monitoring system. The object tracking system does not need to identify all the monitoring images in the area at one time, so that the efficiency of the identification screen can be improved, and the police can be assisted immediately. Track the items to be searched and increase the chances of solving the case.

Although the present invention has been disclosed above in the foregoing embodiments, it is not intended to limit the invention. It is within the scope of the invention to be modified and modified without departing from the spirit and scope of the invention. Please refer to the attached patent application for the scope of protection defined by the present invention.

10‧‧‧ Object Tracking System

101‧‧‧Image Analysis Apparatus 101

103‧‧‧Path building device 103

105, 105a ~ 105d ‧ ‧ photography equipment

2, 5‧‧‧ physical geographic area

21. 51‧‧‧ sections

23, 53‧‧‧ blocks

3,6‧‧‧ Items to be searched

A～N‧‧‧Monitoring Point

1 is a functional block diagram of an object tracking system according to an embodiment of the invention. FIG. 2 is a flow chart of steps of an object tracking method according to an embodiment of the invention. FIG. 3 is a schematic diagram of a physical geographic area according to an embodiment of the invention. FIG. 4 is a schematic diagram of a monitoring point setting photographing apparatus according to an embodiment of the invention. FIG. 5 is a flow chart showing the steps of an object tracking method according to another embodiment of the present invention. FIG. 6 is a schematic diagram of a physical geographic area according to another embodiment of the present invention. 7 and FIG. 8 are schematic diagrams showing physical geographic regions according to the embodiment of FIG. 6 according to the present invention.

Claims (8)

An object tracking method is applicable to an object tracking system. The object tracking method includes: defining a plurality of monitoring points in a physical geographic area, each of the monitoring points having at least one photographic device, each of the photographic devices being used for shooting a road picture; according to a position signal, one of the monitoring points is selected as an initial monitoring point, the position signal is related to the physical geographical area; at least one first priority point is defined, and each of the first priority points is selected And one of the monitoring points, and each of the first priority points in the physical geographic area has an adjacent relationship with the location of the initial monitoring point; determining the road picture of each of the first priority points Whether there is an object to be searched; and when the road picture has the object to be searched, the first priority point that captures the object to be searched is defined as the next initial monitoring point. The object tracking method of claim 1, wherein when the photographing device defined as the first priority point does not capture the road image, the object tracking method further includes defining a plurality of second priority points, each of the The two priority points are selected from one of the monitoring points, and the position of each of the second priority points in the geographical area of the entity has the adjacent relationship with the position of the first priority point where the road picture is not captured. The object tracking method of claim 2, further comprising determining whether the road image of each of the second priority points has the object to be searched, and the road image of each of the first priority points does not have the object to be searched And when the road picture of one of the second priority points has the object to be searched, the second priority point is defined as the next initial monitoring point. The object tracking method of claim 1, wherein after defining the monitoring point of the object to be searched as the next initial monitoring point, in the step of defining the at least one first priority point, The moving direction of the object to be searched in the picture defines the at least one first priority point. The object tracking method of claim 4, wherein when the road image of the first priority point does not have the object to be searched, the object tracking method further comprises: defining at least one search feature related to the object to be searched. Obtaining a history record of the road picture of the initial monitoring point and the road picture of the at least one first priority point according to a time point at which one of the monitoring points is defined as the initial monitoring point; The history record defines at least one suspicious object, each of the suspicious objects having the at least one search feature; determining whether the road picture of each of the first priority points and the road picture of the initial monitoring point have the suspicious object; And when the road picture of each of the first priority points and the road picture of the initial monitoring point have the suspicious object, defining the first priority point of capturing the suspicious object as the next initial monitoring point. The object tracking method of claim 1, wherein when the number of the at least one photographing device of the monitoring point is plural, each of the photographing devices of the monitoring point respectively photographs the road image in different shooting directions. The object tracking method of claim 1, wherein the physical geographic area has a plurality of road segments, each of the monitoring points being defined between at least two of the road segments, and between the monitoring points having the adjacent relationship At least one of the road segments is connected. The object tracking method of claim 1, further comprising storing the road image captured by the object to be searched in a data storage space.