CN111489380B - Target object track analysis method - Google Patents
Target object track analysis method Download PDFInfo
- Publication number
- CN111489380B CN111489380B CN202010290588.4A CN202010290588A CN111489380B CN 111489380 B CN111489380 B CN 111489380B CN 202010290588 A CN202010290588 A CN 202010290588A CN 111489380 B CN111489380 B CN 111489380B
- Authority
- CN
- China
- Prior art keywords
- max
- target object
- information
- camera
- identification
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 39
- 238000001514 detection method Methods 0.000 claims description 16
- 238000012544 monitoring process Methods 0.000 claims description 15
- 238000000034 method Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 4
- 239000003550 marker Substances 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 abstract description 3
- 230000004044 response Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011897 real-time detection Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
Classifications
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a target object track analysis method, which is characterized in that through object identification, the cross rate of overlapping areas of adjacent frames of a target object in a video frame sequence, such as a vehicle, a person or other moving objects in an intelligent community, is calculated, so that whether the object is the same object or not is determined, and the track analysis of the target object is realized. The target object track analysis method can effectively solve the problem that the target object symbolic information cannot be identified and the identification rate is low, so that the dependence on the identification of the target object symbolic information is reduced, and the efficiency and the accuracy of a target object track analysis mechanism are improved.
Description
Technical Field
The invention relates to a target object track analysis method in a video, in particular to a target object track analysis method with high efficiency and good accuracy.
Background
With the continuous maturity of the concept and specification of the smart community and the application of the intelligent application technology in the smart community, for example, the technologies such as face recognition, face access control, object recognition, vehicle recognition are continuously updated, so that the intelligent degree of the smart community is continuously increased to another level. In the development process of the intelligent community, the track analysis accuracy of the mobile object in the community is very important for the security field of the intelligent community. For example, the travel records of the elderly living alone or the people concerned in the community are analyzed and recorded, so that the real-time detection of the life health of the elderly living alone and the people concerned and the alarm of abnormal conditions are effectively guaranteed, and the method is particularly important for the precise and efficient processing of community assistance work and the timeliness and accuracy of information; and the illegal behavior prejudgment and the real-time detection of the travel track of illegal personnel entering the community and the informatization management of the transactions of vehicle travel, parking and the like in the community are particularly important for the analysis accuracy of the moving track of the moving target object. At present, the moving object is often monitored by means of object identification, identification information identification and the like, dotting and marking analysis is carried out on information on a time axis, and the track of a target object is obtained. However, the discrete point marking method often causes inaccuracy of analysis of the moving track of the object due to the limitation of the recognition rate, and causes that the identification information cannot be recognized and the track of the target object cannot be detected for a long time due to the weather reason and the occlusion of a specific scene. For example, in the license plate recognition, at the entrance and exit of a community brake, because of the adoption of a light supplement and high-quality license plate recognition device, the license plate recognition accuracy is basically close to 100%, however, in the entrance of a vehicle into a community, because a common camera is mostly adopted, due to factors such as a shooting angle and a direction of a vehicle body in the vehicle moving process, the license plate recognition is basically difficult to obtain, intermittent data of a vehicle route action track is lost, and only a track recommendation algorithm can be used, and the most probable line is recommended approximately to make up.
Therefore, there is a need for an improvement to overcome the deficiencies of the prior art.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides a target object track analysis method with high efficiency and good accuracy.
The technical scheme of the invention is as follows:
a target object trajectory analysis method includes the following steps:
s1, when the object enters the community, the certainty information authentication is performed to generate the object mark information and the initial coordinate information [ [ x ]0min,y0min],[x0max,y0max]]；
S2, judging the recording condition of the target object in the cameras in the community and carrying out targeted analysis:
if the target object moves in the same camera picture, analyzing the track of the target object and executing the step S3;
if the target object enters the adjacent area of the pictures shot by the two cameras, analyzing the track of the target object and executing the step S4;
if the target object moves out of one camera picture and no other camera pictures are adjacent to the target object, executing step S5;
s3, setting video frame acquisition frequency according to a target object, selecting video frames according to the video frame acquisition frequency for detection, and performing object detection and cross rate calculation on each detected frame image to perform target object track analysis;
s4, according to the shooting positions of the cameras in the community, the coverage area of the cameras in the whole community is processed in advance, if the coverage areas are adjacent, the adjacent cameras are associated and labeled in the adjacent direction, and when a target object moves from one camera monitoring area to another camera monitoring area along one direction, the following operations are carried out:
s4a, storing track information of a target object in the current camera video;
s4b, acquiring information of target object identification and mark information identification at a critical point of adjacent cameras in the moving direction, and performing accurate information link matching;
s4c, if the information link matching beyond the boundary point is successful, namely the object identification at the boundary point acquires the coordinate information of the target object, linking the object tracks of the front camera and the back camera; otherwise, the link fails;
s5, if the target object moves out of a camera monitoring area and no other camera monitoring area is adjacent to the camera monitoring area or no critical point connection information is matched, namely the step S4c link fails, regularly matching the mark information of which the mark information identification accuracy rate of all objects in the cell is less than 80%, and if the matching is successful, considering the moving track information of the same target object and combining the moving track information; otherwise, the matching fails and no processing is performed.
Preferably, in step S1, the "entering of the target object into the community" is performed to authenticate the certainty information, and target object identification information and initial coordinate information [ [ x ] ], are generated0min,y0min],[x0max,y0max]]"is: when a vehicle enters a community, a camera is snapped at the entrance and exit of a brake to shoot a license plate, license plate information is generated, and the vehicle body is in the communityVehicle body coordinate information [ x ] corresponding to snap camera at brake0min,y0min],[x0max,y0max]]。
Preferably, in step S1, the "entering of the target object into the community" is performed to authenticate the certainty information, and target object identification information and initial coordinate information [ [ x ] ], are generated0min,y0min],[x0max,y0max]]"is: when a pedestrian enters a community, a camera at an entrance and an exit of the community carries out face snapshot to generate face information and corresponding pedestrian coordinate information [ [ x ] of the camera at the entrance and the exit of the community0min,y0min],[x0max,y0max]]。
As a preferred technical solution, in the step S3, the video frame capturing frequency is determined by the number of frames per second x captured by the camera, the moving speed of the target object y m/S, the length of the target object a m, and the crossing rate threshold b: the maximum interval acquisition frequency n ═ (a × b)/(y/x).
As a preferred technical solution, the specific method of "performing object detection and cross rate calculation for each detected frame image to perform target object trajectory analysis" in step S3 includes:
s3a, acquiring a frame coordinate set Object _ registration _ set of the target Object through target Object identification and position coordinate statistics, judging whether the designated mark information and accuracy of the Object exist in the system when the Object identification detects that the target type Object enters from the boundary of the camera detection area, and recording the track information Object _ registration _ i [ [ x ] of the target Object according to the mark information, the preparation rate and the current time if the mark information and the accuracy of the designated Object exist in the systemimin,yimin],[ximax,yimax]](ii) a If not, the object is uniquely marked and the coordinate information of the object is marked [ x ]imin,yimin],[ximax,yimax]]The accuracy rate is 0 and the current time information, wherein i is the number of the object identification detected target object;
s3b, identifying and acquiring the Identification-reco of the frame coordinate set of the target object type with the Identification information through the Identification informationA determination _ repetition _ set; identifying the mark information of each frame of video, if the mark information of the target object can be obtained, recording the mark information of the target type object of the current frame and the object frame coordinate Identification _ registration _ j [ [ x ] of the target type objectjmin,yjmin],[xjmax,yjmax]]Wherein j is the number of the object of the target class detected by the object identification;
s3c, calculating the two frame coordinate sets and the target type Object frame intersection rate in the same frame of video through a two-rectangle intersection rate algorithm for the Object _ registration _ set and the Identification _ registration _ se acquired in the steps S3a and S3 b; and if the crossing rate reaches 80%, determining that the objects are the same target type, comparing the accuracy rate of the marking information, selecting the information with the highest preparation rate as the marking information, and recording the coordinate track of the marking information.
As a further preferable technical solution, in the step S3c, a specific method for "calculating the intersection rate of the two frame coordinate sets and the target type object frame in the same frame video by using a two-rectangle intersection rate algorithm" includes:
a. calculating two rectangular areas s1 and s 2;
b. if (x)imin≥xjmax) or (x)imax≤xjmin) or (y)imin≥yjmax) or (y)imax≤yjmin), the intersection area of the two rectangles is considered to be 0, and the intersection rate is also 0; otherwise, turning to the step c;
c. if two rectangles are overlapped, the overlapped area must be a rectangle, and then the point pair set ([ x ]imin,yimin],[ximin,yimax],[ximin,yjmin],[ximin,yjmax],[ximax,yimin],[ximax,yimax],[ximax,yjmin],[ximax,yjmax],[xjmin,yimin],[xjmin,yimax],[ximin,yjmin],[xj min,yjmax],[xjmax,yimin],[xjmax,yimax],[xjmax,yjmin],[xjmax,yjmax]) Totally 16 points are used for judging whether boundaries are included in the two matrixes simultaneously, namely if one point P is included, ximin≤xP≤ximax and yimin≤yP≤yimax and xjmin≤xP≤xjmax and yjmin≤yP≤yjmax, then record the point into the point set points _ set of the rewrite area;
d. removing the duplicate of the point set points _ set to obtain four points which are four point coordinates of an overlapped region rectangle;
e. calculating the area s of the overlapping area according to the coordinates of the four points of the rectangle of the overlapping area;
f. calculating the crossing rate W of the overlapping area as (s/s1+ s/s 2)/2; and the crossing rate W of the areas of the overlapping areas is the crossing rate of the target type object frames.
According to the target object track analysis method, through object recognition, the cross rate of the overlapping area of adjacent frames of a target object in a video frame sequence, such as a vehicle, a person or other moving objects in an intelligent community, is calculated, so that whether the object is the same object or not is determined, and the track analysis of the target object is realized. The target object track analysis method can effectively solve the problem that the target object symbolic information cannot be identified and the identification rate is low, so that the dependence on the identification of the target object symbolic information is reduced, and the efficiency and the accuracy of a target object track analysis mechanism are improved.
Drawings
Fig. 1 is a flow chart of a target object trajectory analysis method according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terminology used in the embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the examples of the present invention and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, and "a" and "an" generally include at least two, but do not exclude at least one, unless the context clearly dictates otherwise.
It should be understood that the term "and/or" as used herein is merely one type of association that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
The words "if", as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination" or "in response to a detection", depending on the context. Similarly, the phrases "if determined" or "if detected (a stated condition or event)" may be interpreted as "when determined" or "in response to a determination" or "when detected (a stated condition or event)" or "in response to a detection (a stated condition or event)", depending on the context.
It is also noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a good or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such good or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a commodity or system that includes the element.
Fig. 1 shows a target object trajectory analysis method according to the present invention, which includes the following steps:
s1, when the object enters the community, the certainty information authentication is performed to generate the object mark information and the initial coordinate information [ [ x ]0min,y0min],[x0max,y0max]](ii) a For example: when a vehicle enters a community, license plate information accuracy recording and authentication can be carried out at a gate exit, and when pedestrians enter the community, face snapshot is also needed.
S2, judging the recording condition of the target object in the cameras in the community and carrying out pertinence analysis, wherein the distribution of the cameras in the community has a certain irregular shape, so that all roads in the community are not covered, and the object track analysis is carried out by using a cross rate algorithm, and the method is divided into three conditions:
if the target object moves in the same camera picture, analyzing the track of the target object and executing the step S3;
if the target object enters the adjacent area of the pictures shot by the two cameras, analyzing the track of the target object and executing the step S4;
if the target object moves out of one camera picture and no other camera pictures are adjacent to the target object, executing step S5;
s3, setting video frame acquisition frequency according to a target object, selecting video frames according to the video frame acquisition frequency for detection, and performing object detection and cross rate calculation on each detected frame image to perform target object track analysis;
s4, according to the shooting positions of the cameras in the community, the coverage area of the cameras in the whole community is processed in advance, if the coverage areas are adjacent, the adjacent cameras are associated and labeled in the adjacent direction, and when a target object moves from one camera monitoring area to another camera monitoring area along one direction, the following operations are carried out:
s4a, storing track information of a target object in the current camera video;
s4b, acquiring information of target object identification and mark information identification at a critical point of adjacent cameras in the moving direction, and performing accurate information link matching;
s4c, if the information link matching beyond the boundary point is successful, namely the object identification at the boundary point acquires the coordinate information of the target object, linking the object tracks of the front camera and the back camera; otherwise, the link fails;
s5, if the target object moves out of a camera monitoring area and no other camera monitoring area is adjacent to the camera monitoring area or no critical point connection information is matched, namely the step S4c link fails, regularly matching the mark information of which the mark information identification accuracy rate of all objects in the cell is less than 80%, and if the matching is successful, considering the moving track information of the same target object and combining the moving track information; otherwise, the matching fails and no processing is performed.
In practical applications, when the target object is a vehicle, the deterministic information authentication is performed to generate target object marker information and initial coordinate information [ [ x ] when the target object enters the community as described in S10min,y0min],[x0max,y0max]]"is: when a vehicle enters a community, a license plate is shot under a snapshot camera at the entrance and exit of a brake, and license plate information and vehicle body coordinate information [ x ] corresponding to the snapshot camera of the vehicle body at the brake are generated0min,y0min],[x0max,y0max]]. When the target object is a pedestrian, the certainty information authentication is performed to generate target object identification information and initial coordinate information [ [ x ] [ when the target object enters the community region in step S10min,y0min],[x0max,y0max]]"is: when a pedestrian enters a community, a camera at an entrance and an exit of the community carries out face snapshot to generate face information and corresponding pedestrian coordinate information [ [ x ] of the camera at the entrance and the exit of the community0min,y0min],[x0max,y0max]]。
As a preferable scheme, in the step S3, the video frame capturing frequency is determined by the number of frames per second x captured by the camera, the moving speed of the target object y meters/second, the length of the target object a meters, and the crossing rate threshold b: the maximum interval acquisition frequency n ═ (a × b)/(y/x). For example, the target object is a vehicle, the video frame of the camera is about 25 frames/s, the moving speed of the vehicle in the cell is limited to 20km/h (i.e. 5.6m/s), the length of the vehicle body is 3.8 m to 4.3 m, and the width is 1.6 m to 1.8 m, so if the threshold value of the intersection rate is selected to be 30%, the maximum interval acquisition frequency n is (0.3 × (3.8)/(5.6/25) × (5.1), the maximum interval acquisition frequency is not more than 5.1, and one frame can be selected for detection every 4 frames, thereby reducing the amount of calculation and ensuring the detection effect.
As a preferable scheme, the specific method of "performing object detection and cross rate calculation for each detected frame image to perform target object trajectory analysis" in step S3 includes:
s3a, acquiring a frame coordinate set Object _ registration _ set of the target Object through target Object identification and position coordinate statistics, judging whether the designated mark information and accuracy of the Object exist in the system when the Object identification detects that the target type Object enters from the boundary of the camera detection area, and recording the track information Object _ registration _ i [ [ x ] of the target Object according to the mark information, the preparation rate and the current time if the mark information and the accuracy of the designated Object exist in the systemimin,yimin],[ximax,yimax]](ii) a If not, the object is uniquely marked and the coordinate information of the object is marked [ x ]imin,yimin],[ximax,yimax]]The accuracy rate is 0 and the current time information, wherein i is the number of the object identification detected target object;
s3b, identifying and acquiring the Identification _ registration _ set of the frame coordinate set of the target object type with the Identification information through the Identification information; identifying the mark information of each frame of video, if the mark information of the target object can be obtained, recording the mark information of the target type object of the current frame and the object frame coordinate Identification _ registration _ j [ [ x ] of the target type objectjmin,yjmin],[xjmax,yjmax]]Wherein j is the number of the object of the target class detected by the object identification;
s3c, calculating the two frame coordinate sets and the target type Object frame intersection rate in the same frame of video through a two-rectangle intersection rate algorithm for the Object _ registration _ set and the Identification _ registration _ se acquired in the steps S3a and S3 b; and if the crossing rate reaches 80%, determining that the objects are the same target type, comparing the accuracy rate of the marking information, selecting the information with the highest preparation rate as the marking information, and recording the coordinate track of the marking information.
In step S3c, the specific method for calculating the intersection rate of the two frame coordinate sets and the target type object frame in the same frame of video through the two rectangle intersection rate algorithm includes:
a. calculating two rectangular areas s1 and s 2;
b. if (x)imin≥xjmax) or (x)imax≤xjmin) or (y)imin≥yjmax) or (y)imax≤yjmin), the intersection area of the two rectangles is considered to be 0, and the intersection rate is also 0; otherwise, turning to the step c;
c. if two rectangles are overlapped, the overlapped area must be a rectangle, and then the point pair set ([ x ]imin,yimin],[ximin,yimax],[ximin,yjmin],[ximin,yjmax],[ximax,yimin],[ximax,yimax],[ximax,yjmin],[ximax,yjmax],[xjmin,yimin],[xjmin,yimax],[ximin,yjmin],[xj min,yjmax],[xjmax,yimin],[xjmax,yimax],[xjmax,yjmin],[xjmax,yjmax]) Totally 16 points are used for judging whether boundaries are included in the two matrixes simultaneously, namely if one point P is included, ximin≤xP≤ximax and yimin≤yP≤yimax and xjmin≤xP≤xjmax and yjmin≤yP≤yjmax, then record the point into the point set points _ set of the rewrite area;
d. removing the duplicate of the point set points _ set to obtain four points which are four point coordinates of an overlapped region rectangle;
e. calculating the area s of the overlapping area according to the coordinates of the four points of the rectangle of the overlapping area;
f. calculating the crossing rate W of the overlapping area as (s/s1+ s/s 2)/2; and the crossing rate W of the areas of the overlapping areas is the crossing rate of the target type object frames.
According to the target object track analysis method, through object recognition, the cross rate of the overlapping area of adjacent frames of a target object in a video frame sequence, such as a vehicle, a person or other moving objects in an intelligent community, is calculated, so that whether the object is the same object or not is determined, and the track analysis of the target object is realized. The target object track analysis method can effectively solve the problem that the target object symbolic information cannot be identified and the identification rate is low, so that the dependence on the identification of the target object symbolic information is reduced, and the efficiency and the accuracy of a target object track analysis mechanism are improved.
In summary, the embodiments of the present invention are merely exemplary and should not be construed as limiting the scope of the invention. All equivalent changes and modifications made according to the content of the claims of the present invention should fall within the technical scope of the present invention.
Claims (5)
1. A target object trajectory analysis method is characterized in that: the method comprises the following steps:
s1, when the object enters the community, the certainty information authentication is performed to generate the object mark information and the initial coordinate information [ [ x ]0min,y0min],[x0max,y0max]]；
S2, judging the recording condition of the target object in the cameras in the community and carrying out targeted analysis:
if the target object moves in the same camera picture, analyzing the track of the target object and executing the step S3;
if the target object enters the adjacent area of the pictures shot by the two cameras, analyzing the track of the target object and executing the step S4;
if the target object moves out of one camera picture and no other camera pictures are adjacent to the target object, executing step S5;
s3, setting video frame acquisition frequency according to a target object, selecting video frames according to the video frame acquisition frequency for detection, and performing object detection and cross rate calculation on each detected frame image to perform target object track analysis; the specific method for performing object detection and cross rate calculation on each detected frame image to perform target object trajectory analysis includes:
s3a, acquiring a frame coordinate set Object _ registration _ set of the target Object through target Object identification and position coordinate statistics, judging whether the designated mark information and accuracy of the Object exist in the system when the Object identification detects that the target type Object enters from the boundary of the camera detection area, and recording the track information Object _ registration _ i = [ [ x ] of the target Object according to the mark information, accuracy and current time if the mark information, accuracy and accuracy existimin,yimin],[ximax,yimax]](ii) a If not, the object is uniquely marked and the coordinate information of the object is marked [ x ]imin,yimin],[ximax,yimax]]The accuracy rate is 0 and the current time information, wherein i is the number of the object identification detected target object;
s3b, identifying and acquiring the Identification _ registration _ set of the frame coordinate set of the target object type with the Identification information through the Identification information; identifying the mark information of each frame of video, if the mark information of the target object can be obtained, recording the mark information of the target type object of the current frame and the object frame coordinate Identification _ recognition _ repetition _ j = [ [ x ] of the target type objectjmin,yjmin],[xjmax,yjmax]]Wherein j is the number of the object of the target class detected by the object identification;
s3c, calculating the two frame coordinate sets and the target type Object frame intersection rate in the same frame of video through a two-rectangle intersection rate algorithm for the Object _ registration _ set and the Identification _ registration _ se acquired in the steps S3a and S3 b; if the crossing rate reaches 80%, the objects are considered to be the same target type of objects, the accuracy rates of the marking information are compared, the information with the highest accuracy rate is selected as the marking information, and the coordinate track of the marking information is recorded;
s4, according to the shooting positions of the cameras in the community, the coverage area of the cameras in the whole community is processed in advance, if the coverage areas are adjacent, the adjacent cameras are associated and labeled in the adjacent direction, and when a target object moves from one camera monitoring area to another camera monitoring area along one direction, the following operations are carried out:
s4a, storing track information of a target object in the current camera video;
s4b, acquiring information of target object identification and mark information identification at a critical point of adjacent cameras in the moving direction, and performing accurate information link matching;
s4c, if the information link matching beyond the boundary point is successful, namely the object identification at the boundary point acquires the coordinate information of the target object, linking the object tracks of the front camera and the back camera; otherwise, the link fails;
s5, if the target object moves out of a camera monitoring area and no other camera monitoring area is adjacent to the camera monitoring area or no critical point connection information is matched, namely the step S4c link fails, regularly matching the mark information of which the mark information identification accuracy rate of all objects in the cell is less than 80%, and if the matching is successful, considering the moving track information of the same target object and combining the moving track information; otherwise, the matching fails and no processing is performed.
2. The target object trajectory analysis method of claim 1, wherein: in step S1, when the target object enters the community, the certainty information authentication is performed to generate target object marker information and initial coordinate information [ [ x ] ]0min,y0min],[x0max,y0max]]"is: when a vehicle enters a community, a license plate is shot under a snapshot camera at the entrance and exit of a brake, and license plate information and vehicle body coordinate information [ x ] corresponding to the snapshot camera of the vehicle body at the brake are generated0min,y0min],[x0max,y0max]]。
3. The target object trajectory analysis method of claim 1, wherein: in step S1, when the target object enters the community, the certainty information authentication is performed to generate target object marker information and initial coordinate information [ [ x ] ]0min,y0min],[x0max,y0max]]"is: when a pedestrian enters a community, a camera at an entrance and an exit of the community carries out face snapshot to generate face information and corresponding pedestrian coordinate information [ [ x ] of the camera at the entrance and the exit of the community0min,y0min],[x0max,y0max]]。
4. The target object trajectory analysis method of claim 1, wherein: in the step S3, the video frame capturing frequency is determined by the number of frames per second x shot by the camera, the moving speed of the target object y m/S, the length of the target object a m, and the crossing rate threshold b: maximum interval acquisition frequency n = (a × b)/(y/x).
5. The target object trajectory analysis method of claim 1, wherein: the specific method for calculating the intersection rates of the two frame coordinate sets and the target type object frame in the same frame video through the two rectangle intersection rate algorithm in the step S3c is as follows:
a. calculating two rectangular areas s1 and s 2;
b. if (x)imin≥xjmax) or (x)imax≤xjmin) or (y)imin≥yjmax) or (y)imax≤yjmin), the intersection area of the two rectangles is considered to be 0, and the intersection rate is also 0; otherwise, turning to the step c;
c. if two rectangles are overlapped, the overlapped area must be a rectangle, and then the point pair set ([ x ]imin,yimin],[ximin,yimax],[ximin,yjmin],[ximin,yjmax],[ximax,yimin],[ximax,yimax],[ximax,yjmin],[ximax,yjmax],[xjmin,yimin],[xjmin,yimax],[ximin,yjmin],[xjmin,yjmax],[xjmax,yimin],[xjmax,yimax],[xjmax,yjmin],[xjmax,yjmax]) Totally 16 points are used for judging whether boundaries are included in the two matrixes simultaneously, namely if one point P is included, ximin≤xP≤ximax and yimin≤yP ≤yimax and xjmin≤xP≤xjmax and yjmin≤yP≤yjmax, then record the point into the point set points _ set of the rewrite area;
d. removing the duplicate of the point set points _ set to obtain four points which are four point coordinates of an overlapped region rectangle;
e. calculating the area s of the overlapping area according to the coordinates of the four points of the rectangle of the overlapping area;
f. calculating the crossing rate W = (s/s1+ s/s2)/2 of the overlapping region area; and the crossing rate W of the areas of the overlapping areas is the crossing rate of the target type object frames.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010290588.4A CN111489380B (en) | 2020-04-14 | 2020-04-14 | Target object track analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010290588.4A CN111489380B (en) | 2020-04-14 | 2020-04-14 | Target object track analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111489380A CN111489380A (en) | 2020-08-04 |
| CN111489380B true CN111489380B (en) | 2022-04-12 |
Family
ID=71812756
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010290588.4A Active CN111489380B (en) | 2020-04-14 | 2020-04-14 | Target object track analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111489380B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114220077A (en) * | 2022-02-21 | 2022-03-22 | 金叶仪器(山东)有限公司 | Method for realizing object quantity statistics and moving direction monitoring based on monitoring equipment |
| CN115018433B (en) * | 2022-08-10 | 2022-11-29 | 四川港投新通道物流产业投资集团有限公司 | Wine supply chain monitoring method, device, equipment and medium |
| CN115601686B (en) * | 2022-12-09 | 2023-04-11 | 浙江莲荷科技有限公司 | Method, device and system for confirming delivery of articles |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106846374A (en) * | 2016-12-21 | 2017-06-13 | 大连海事大学 | The track calculating method of vehicle under multi-cam scene |
| CN107483889A (en) * | 2017-08-24 | 2017-12-15 | 北京融通智慧科技有限公司 | The tunnel monitoring system of wisdom building site control platform |
| CN108875588A (en) * | 2018-05-25 | 2018-11-23 | 武汉大学 | Across camera pedestrian detection tracking based on deep learning |
| CN110378931A (en) * | 2019-07-10 | 2019-10-25 | 成都数之联科技有限公司 | A kind of pedestrian target motion track acquisition methods and system based on multi-cam |
| CN110428443A (en) * | 2019-08-11 | 2019-11-08 | 上海天诚比集科技有限公司 | A kind of intelligence community Vehicle tracing method |
| CN110619657A (en) * | 2019-08-15 | 2019-12-27 | 青岛文达通科技股份有限公司 | Multi-camera linkage multi-target tracking method and system for smart community |
| CN110619277A (en) * | 2019-08-15 | 2019-12-27 | 青岛文达通科技股份有限公司 | Multi-community intelligent deployment and control method and system |
-
2020
- 2020-04-14 CN CN202010290588.4A patent/CN111489380B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106846374A (en) * | 2016-12-21 | 2017-06-13 | 大连海事大学 | The track calculating method of vehicle under multi-cam scene |
| CN107483889A (en) * | 2017-08-24 | 2017-12-15 | 北京融通智慧科技有限公司 | The tunnel monitoring system of wisdom building site control platform |
| CN108875588A (en) * | 2018-05-25 | 2018-11-23 | 武汉大学 | Across camera pedestrian detection tracking based on deep learning |
| CN110378931A (en) * | 2019-07-10 | 2019-10-25 | 成都数之联科技有限公司 | A kind of pedestrian target motion track acquisition methods and system based on multi-cam |
| CN110428443A (en) * | 2019-08-11 | 2019-11-08 | 上海天诚比集科技有限公司 | A kind of intelligence community Vehicle tracing method |
| CN110619657A (en) * | 2019-08-15 | 2019-12-27 | 青岛文达通科技股份有限公司 | Multi-camera linkage multi-target tracking method and system for smart community |
| CN110619277A (en) * | 2019-08-15 | 2019-12-27 | 青岛文达通科技股份有限公司 | Multi-community intelligent deployment and control method and system |
Non-Patent Citations (1)
| Title |
|---|
| 基于深度学习的多摄像头协同目标跟踪方法研究;陈渠;《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅰ辑》;20190915;论文全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111489380A (en) | 2020-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111489380B (en) | Target object track analysis method | |
| CN108062349B (en) | Video monitoring method and system based on video structured data and deep learning | |
| CN108053427B (en) | Improved multi-target tracking method, system and device based on KCF and Kalman | |
| CN108052859B (en) | Abnormal behavior detection method, system and device based on clustering optical flow characteristics | |
| KR102122859B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| Bloisi et al. | Argos—A video surveillance system for boat traffic monitoring in Venice | |
| CN103986910A (en) | Method and system for passenger flow statistics based on cameras with intelligent analysis function | |
| CN111626275B (en) | Abnormal parking detection method based on intelligent video analysis | |
| CN106558224B (en) | A kind of traffic intelligent monitoring and managing method based on computer vision | |
| CN101303727A (en) | Intelligent management method based on video human number Stat. and system thereof | |
| CN111738240A (en) | Region monitoring method, device, equipment and storage medium | |
| CN111325048B (en) | Personnel gathering detection method and device | |
| Chang et al. | Video analytics in smart transportation for the AIC'18 challenge | |
| CN102855508B (en) | Opening type campus anti-following system | |
| KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN104200466A (en) | Early warning method and camera | |
| CN108986472A (en) | One kind turns around vehicle monitoring method and device | |
| CN112733598A (en) | Vehicle law violation determination method and device, computer equipment and storage medium | |
| CN112084892B (en) | Road abnormal event detection management device and method thereof | |
| CN113903008A (en) | Ramp exit vehicle violation identification method based on deep learning and trajectory tracking | |
| Huang et al. | A real-time and color-based computer vision for traffic monitoring system | |
| Mehboob et al. | Trajectory based vehicle counting and anomalous event visualization in smart cities | |
| WO2018209470A1 (en) | License plate identification method and system | |
| CN105227918A (en) | A kind of intelligent control method and device | |
| CN112562315B (en) | Method, terminal and storage medium for acquiring traffic flow information |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231205 Address after: No.2252, Shaling Road, Shatou, Jiangsu Province Patentee after: JIANGSU TC SMART SYSTEMS GROUP Co.,Ltd. Address before: Room 904, building 2, No. 618, Guangxing Road, Songjiang District, Shanghai 201613 Patentee before: SHANGHAI TIANCHENG BIJI TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: No.2252, Shaling Road, Shatou, Jiangsu Province Patentee after: Jiangsu Tiancheng Intelligent Group Co.,Ltd. Country or region after: China Address before: No.2252, Shaling Road, Shatou, Jiangsu Province Patentee before: JIANGSU TC SMART SYSTEMS GROUP Co.,Ltd. Country or region before: China |