CN113743316A - Vehicle jamming behavior identification method, system and device based on target detection - Google Patents
Vehicle jamming behavior identification method, system and device based on target detection Download PDFInfo
- Publication number
- CN113743316A CN113743316A CN202111042394.3A CN202111042394A CN113743316A CN 113743316 A CN113743316 A CN 113743316A CN 202111042394 A CN202111042394 A CN 202111042394A CN 113743316 A CN113743316 A CN 113743316A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- center point
- target
- straight lines
- distance
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a vehicle jamming behavior identification method, a system and a device based on target detection, which are characterized by comprising the steps of obtaining a monitoring image or video; detecting the image or the video through a target detection algorithm to obtain detection results of all targets, and obtaining vehicle information in the detection results of all targets; the method and the system can simply, quickly, accurately and efficiently identify the vehicle congestion behavior at the urban intersection.
Description
Technical Field
The invention relates to the field of intelligent traffic, in particular to a vehicle jamming behavior identification method, system and device based on target detection.
Background
The traffic jam behavior at urban intersections is a common illegal phenomenon, and traffic jam and even traffic accidents can be caused in serious conditions. The traffic road driving rule stipulates that the vehicle owner who acts with congestion carries out fine and deduction.
In recent years, traffic police departments increase punishment on congestion behaviors at urban traffic intersections, and generally adopt modes of unmanned aerial vehicle aerial photography, electronic police snapshot, road surface investigation and the like, and although the schemes are effective, the consumption of manpower and financial resources is huge.
In the technical field, when vehicle jamming behavior recognition is carried out, the mainstream method is to process the whole image by a traditional image processing method according to information of a monitoring video stream, extract corresponding vehicle information and further predict vehicle running tracks.
The deep learning result which can be rapidly developed according to the big data is quite abundant, and the target detection algorithm based on the deep learning can effectively detect the target in the image or the video. However, if abnormal behaviors such as vehicle jam in an image or video are to be directly recognized or a lane line detection method is to be used, a large number of image samples are required for training, and the quality of the samples directly affects the accuracy of the detection result. Therefore, a method for efficiently detecting the vehicle congestion behavior recognition at the urban intersection by using a deep learning-based target detection algorithm is needed.
Disclosure of Invention
The invention aims to provide a vehicle jamming behavior identification method, a vehicle jamming behavior identification system and a vehicle jamming behavior identification device based on target detection, and aims to solve the problem of vehicle jamming behavior identification.
The invention provides a vehicle jamming behavior recognition method based on target detection, which comprises the following steps:
s1, acquiring a monitoring image or video;
s2, detecting the image or the video through a target detection algorithm to obtain detection results of all targets, and acquiring vehicle information in the detection results of all targets;
and S3, judging whether the vehicle has the vehicle jamming behavior according to the vehicle information, and marking the jamming behavior.
The invention also provides a vehicle jamming behavior recognition system based on target detection, which comprises:
an acquisition module: acquiring a monitoring image or video;
a detection module: detecting the image or the video through a target detection algorithm to obtain detection results of all targets, and obtaining vehicle information in the detection results of all targets;
a marking module: and judging whether the vehicle has a vehicle jamming behavior or not according to the vehicle information, and marking the jamming behavior.
The embodiment of the invention also provides a vehicle jamming behavior recognition device based on target detection, which comprises: a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program implementing the steps of the above method when executed by the processor.
The embodiment of the invention also provides a computer readable storage medium, wherein an implementation program for information transmission is stored on the computer readable storage medium, and the implementation program realizes the steps of the method when being executed by a processor.
By adopting the embodiment of the invention, the vehicle jam behavior at the urban intersection can be simply, quickly, accurately and efficiently identified.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a vehicle congestion behavior recognition method based on object detection according to an embodiment of the present invention, as shown in fig. 1,
FIG. 2 is a detailed flowchart of a vehicle jamming behavior identification method based on object detection according to an embodiment of the present invention;
FIG. 3 is a schematic illustration of monitoring inputs for a vehicle jamming behavior recognition method based on object detection in accordance with an embodiment of the present invention;
fig. 4 is a schematic view of a hough transform curve of a vehicle jamming behavior identification method based on target detection according to an embodiment of the present invention;
fig. 5 is a schematic diagram of the result of suppressing the direction of a non-lane of a vehicle congestion behavior recognition method based on object detection according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a vehicle jamming photograph of a vehicle jamming behavior recognition method based on target detection according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a vehicle jamming behavior recognition system based on object detection according to an embodiment of the present invention;
fig. 8 is a schematic diagram of a vehicle jamming behavior recognition device based on object detection according to an embodiment of the present invention.
Description of reference numerals:
710: an acquisition module; 720: a detection module; 730: a marking module; 80, a memory; and 82, a processor.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments, and it should be understood 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.
In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate orientations and positional relationships based on those shown in the drawings, and are used only for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be considered as limiting the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise. Furthermore, the terms "mounted," "connected," and "connected" are to be construed broadly and may, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Method embodiment
According to an embodiment of the present invention, a vehicle congestion behavior identification method based on target detection is provided, fig. 1 is a flowchart of the vehicle congestion behavior identification method based on target detection according to the embodiment of the present invention, and as shown in fig. 1, the method specifically includes:
s1, acquiring a monitoring image or video;
s2, detecting the image or the video through a target detection algorithm to obtain detection results of all targets, and acquiring vehicle information in the detection results of all targets;
s2 specifically includes: and detecting the image or the video through a target detection algorithm to obtain the position information and the category labels of all targets, and obtaining the position information of the vehicle category according to the position information and the category labels of all the targets.
And S3, judging whether the vehicle has the vehicle jamming behavior according to the vehicle information, and marking the jamming behavior.
S3 specifically includes:
determining a vehicle center point according to the vehicle position information, performing straight line fitting on the vehicle center point through Hough transform, processing the fitted straight line, calculating the distance from the vehicle center point to the processed straight line, judging whether a vehicle is jammed according to the distance, and marking the jammed vehicle if the jammed vehicle is jammed.
Determining a vehicle center point according to vehicle position information, performing straight line fitting on the vehicle center point through Hough transform, processing a fitted straight line, calculating the distance from the vehicle center point to the processed straight line, judging whether a vehicle jamming behavior exists according to the distance, and if the vehicle jamming behavior exists, labeling the vehicle jamming specifically comprises: determining a vehicle target center point according to vehicle position information, performing straight line fitting on the vehicle center point through Hough transform to obtain a voting result of a parameter variable value interval in a parameter space, performing non-maximum value suppression operation on the voting result, deleting a close straight line, updating the voting result, selecting K pair parameters with the highest vote number in the voting result, calculating a formula of K straight lines of the K pair parameters under an original coordinate system, performing non-lane direction suppression on the K straight lines, deleting straight lines which do not accord with a lane line direction to obtain K straight lines which finally accord with the lane line direction, calculating the distance from the vehicle center point to the K straight lines according to the formula of the K straight lines to obtain K distance values, judging whether the minimum value of the K distance values of each vehicle is in a threshold value interval, and if the minimum value is in the threshold value interval, judging that the target does not run in the lane, and (4) carrying out a jamming action and marking the vehicle target.
According to the method, the specific implementation method is as follows:
fig. 2 is a detailed flowchart of a vehicle congestion behavior recognition method based on object detection according to an embodiment of the present invention, as shown in fig. 2,
reading a monitoring image or video of an urban intersection and sending the monitoring image or video into a target detection network;
step two, detecting the image or video read in the step one by using a target detection algorithm to obtain information such as the position and the category of a target in the image or the video; the target detection network can effectively detect various types of targets, and the identification of the congestion behavior only needs to identify the traffic type targets, so that the detection result is screened, and the target types needing congestion identification are reserved.
Step three, acquiring traffic target information in a target detection result;
fourthly, carrying out Hough transformation on the central point of the traffic class target; the method specifically comprises the following steps: and replacing each target by a central point, and carrying out Hough transform on the central point to obtain a plurality of fitted straight lines.
And step five, generating a plurality of similar straight lines around one lane by Hough transform, and therefore adopting non-maximum value suppression to suppress the similar straight lines.
Carrying out non-maximum value inhibition on the result of Hough transform, deleting the close straight lines around the straight line in the lane direction, and updating the result;
step six, selecting K pairs of extreme values with the highest votes in the Hough transformation results, and calculating corresponding expressions of the K pairs of extreme values under rectangular coordinates;
step seven, performing non-lane direction inhibition on the result of the step six, and deleting the straight lines which do not accord with the lane direction to obtain k straight lines which accord with the lane direction;
and step eight, calculating the distance from each target to k straight lines, judging whether the minimum distance in the k distances of each target is in the distance interval, if so, judging that the target is in the jamming behavior, and marking the jamming behavior.
Straight lines which do not accord with the lane direction but have high votes appear in the fitting result of Hough transform, and the distance between adjacent points on the straight lines which accord with the lane direction is small according to the observation of traffic images and videos, so that non-lane direction inhibition is adopted, the distance between the adjacent points on each straight line is calculated, and the straight lines which do not accord with the lane direction are deleted. And calculating the distance between each target and each lane direction straight line, and if the target is located in the distance interval, indicating that the vehicle does not run in a normal lane and a congestion behavior is occurring, and marking and intercepting the vehicle.
Fig. 3 is a schematic monitoring input diagram of a vehicle congestion behavior identification method based on object detection according to an embodiment of the present invention, as shown in fig. 3: the method for identifying the vehicle congestion behavior at the urban intersection by target detection specifically comprises the following steps:
reading a monitoring image or video of an urban intersection;
step two, detecting the image or video read in the step one through a target detection algorithm, wherein the detection result is information such as the position and the type of a target in the image or the video;
step three, carrying out category screening on the detection result generated in the step two, and extracting the target category of the jamming behavior, wherein the category comprises the following steps: automobiles, buses, trucks;
fig. 4 is a schematic view of a hough transform curve of a vehicle congestion behavior identification method based on target detection in the embodiment of the present invention, as shown in fig. 4:
step four, performing straight line fitting on the target center points of the categories by using Hough transform to obtain voting results of parameter variable value intervals in parameter space;
step five, carrying out non-maximum suppression operation on the voting result, deleting the similar straight line, and updating the voting result;
step six, selecting K pairs of parameters with the highest vote number in the voting result, and calculating a formula of K straight lines under the corresponding original coordinate system;
step seven, restraining the K straight lines in the non-lane direction, deleting the straight lines which do not accord with the lane line direction, and obtaining the K straight lines which finally accord with the lane line direction;
step eight, calculating the distance from the center point of each target to k straight lines, wherein each target has k distance values, judging whether the minimum value of the k distance values of each target is within a threshold interval, if so, determining that the target does not run in a lane and is in a jamming behavior, framing the vehicle, marking 'warning', and capturing the target.
Moreover, the method for detecting the vehicle congestion behavior at the urban intersection comprises the following steps: the method comprises the steps of detecting urban intersection traffic images or videos by using a deep learning-based target detection algorithm, and acquiring information such as positions and types of targets in the images or videos.
Then, the detected objects are subjected to category screening. The existing target detection algorithm can detect dozens of types of targets, and only three types of vehicles, buses and trucks are used for carrying out the congestion behavior, so that the congestion behavior only needs to be identified for the three types of targets of the vehicles, the buses and the trucks.
And, Hough transform is performed on the center point of the target object. And obtaining the lane information of the urban intersection. The number of lanes at the urban intersection is more than one, and fitting of a plurality of straight lines in the same image is achieved through Hough transformation.
Furthermore, the result of the hough transform is subjected to non-maximum suppression. The Hough transform generates a plurality of similar straight lines in the same lane direction, and the non-maximum value of the Hough transform result is restrained, so that finally, each straight line does not have similarity with the adjacent straight lines.
Then, the result is subjected to the non-lane direction suppression. Due to the regularity of the lanes and the regularity of the queuing of vehicles in the lanes, the hough transform produces a straight line with a direction different from the lanes. In order to eliminate such straight lines, through observation of traffic scene images and videos, it is noticed that the distance between adjacent points on the straight lines which accord with the lane direction is small, and the distance between adjacent points on the straight lines which do not accord with the lane direction is large, so whether the straight lines are restrained or not is judged by calculating the average value of the distances between the adjacent points on each straight line, and if the average distance between the adjacent points on the straight lines is higher than a threshold value, the straight lines are judged to not accord with the lane line direction, and the straight lines are deleted.
And, whether the target vehicle has a congestion behavior is determined by calculating whether the distance between the object of each target category and the lane belongs to a distance threshold section.
And when the target vehicle is judged to be jammed, the vehicle is subjected to frame selection and marked with 'warning', and the target vehicle is captured.
Step one, reading a monitoring image or video of the urban intersection.
The input of the target detection algorithm is an image or a video, and as shown in fig. 3, the invention takes the monitoring image of the urban intersection as the input of the target detection.
And step two, detecting the image or the video in the step one by using a target detection algorithm.
The output results of the object detection are position information, confidence and category labels of the object in the image or video, and the representation examples are as follows:
””
Predicate＝[pred_boxes:Boxes(tensor([[x1,y1,x2,y2],[x1,y1,x2,y2]])),Score:tensor([0.99,0.98]),pred_classes:tensor([2,4])]
formula 1;
wherein [ x1, y1, x2, y2] represents the top left and bottom right coordinates of the regression box for an object, 0.99 represents the confidence of the object, and 2 represents the class label of the object.
And step three, acquiring traffic target information in the target detection result.
The invention uses the trained target detection algorithm, and can effectively detect various types of targets. When the target center point is used for Hough transform fitting of a lane direction straight line, only traffic targets are needed, and only the traffic targets are needed to be identified when congestion behaviors are detected, so that the non-traffic targets belong to noise points. However, the re-labeling of the data set only containing the traffic targets and the training are tedious in workload, time-consuming and labor-consuming, so that the categories are screened on the original detection result, and only the information of the traffic targets, such as automobiles, trucks and buses, is reserved.
And step four, carrying out Hough transform on the central points of the class targets.
And calculating the coordinate of the center point of the target according to the upper left corner coordinate and the lower right corner coordinate of the target in the target detection result. Because the number of the lane lines in the traffic image or the video is more than one, the common least square method, gradient descent method and the like cannot be used for the straight line where the lane lines are fitted in the traffic image. Therefore, hough transform is performed on the center point of the target, a corresponding curve of the target under the parameter coordinate system is drawn as shown in fig. 3, and a voting matrix of the corresponding parameter is calculated.
And step five, performing non-maximum suppression on the result of Hough transform, suppressing a similar straight line around the straight line, and updating the voting result.
Since the similar straight lines can obtain similar votes, when the hough transform result is drawn to the original image, a plurality of straight lines are obtained at the same lane position. And (3) carrying out non-maximum value inhibition on the parameters, respectively setting similar ranges for the parameters, only keeping a pair of parameters with the highest votes in the ranges, returning the votes of the other parameter pairs to zero, and updating the voting matrix of the corresponding parameters under the parameter coordinate system.
And step six, selecting the K pairs of extreme values with the highest votes in the voting results, and calculating K expressions under corresponding rectangular coordinates.
Fig. 5 is a schematic diagram of the result of suppressing the direction of a non-lane in the vehicle congestion behavior recognition method based on the target detection according to the embodiment of the present invention, as shown in fig. 5:
for the voting result, a plurality of straight lines represented by parameter pairs with low votes do not accord with expectations, so that the straight lines are subjected to a hard threshold operation, the voting result of the corresponding parameter pairs with the votes below a certain threshold is zeroed, and the voting matrix under the parameter coordinate system is updated.
And seventhly, performing non-lane direction inhibition on the K straight lines, deleting the straight lines which do not accord with the lane line direction to obtain the K straight lines which accord with the lane line direction, and drawing the K straight lines to the input image as shown in fig. 5.
Due to the regularity of the lanes and the regularity of the queuing of vehicles in the lanes, the hough transform produces a line in a direction different from the lanes, such as a line perpendicular to the lane line. In order to suppress such straight lines, the result is suppressed in the non-lane direction, and through observation of images and videos of traffic scenes, the distance between adjacent points on the straight lines which accord with the lane direction is small, and the distance between adjacent points on the straight lines which do not accord with the lane direction is large, so that whether the straight lines are suppressed or not is judged by calculating the average value of the distances between the adjacent points on each straight line, and if the average distance between the adjacent points on the straight lines is higher than a threshold value, the straight lines are judged to be not in the lane line direction, and the straight lines are suppressed.
And step eight, calculating the distance from each target to k straight lines, judging whether the minimum distance in the k distances of each target is in the distance interval, and if so, judging that the target is in the jamming behavior.
Calculating the distance between each target and k straight lines, wherein if the distance is greater than the upper bound of the distance interval, it indicates that the distance from the target to another lane is calculated, if the distance is less than the lower bound of the distance interval, it indicates that the target is in a traffic jam behavior, and if the distance is within the distance interval, it indicates that the target is in a traffic jam behavior, and the vehicle is framed and labeled with "warning" to represent that the vehicle is a traffic jam vehicle, fig. 6 is a vehicle traffic jam photographing schematic diagram of a vehicle traffic jam behavior recognition method based on target detection according to an embodiment of the present invention, as shown in fig. 6:
and capturing the vehicles with the detected plugging behavior, and generating an alarm to realize the recognition of the plugging behavior of the vehicles at the urban intersection.
The traffic image or video is subjected to early-stage processing by using a target detection algorithm based on deep learning, a detection result is used as input of vehicle congestion behavior recognition, straight line fitting is carried out on target vehicles by using Hough transform, and post-processing such as non-maximum value suppression, non-lane direction suppression and the like is used for assisting in Hough transform. And finally, calculating the distance from each vehicle to a straight line in the lane direction, and judging that the vehicle is performing the congestion behavior when the distance is within a certain distance interval. The method adopts the target detection based on deep learning, ensures the detection precision and speed, processes the target detection result by combining the traditional Hough transform method and the practical situation, can effectively detect the traffic jam behavior of the intersection on the premise of not retraining labeled data and the like, and is suitable for identifying the traffic jam behavior of the urban traffic intersection.
System embodiment
According to an embodiment of the present invention, a vehicle congestion behavior recognition system based on object detection is provided, and fig. 7 is a schematic diagram of the vehicle congestion behavior recognition system based on object detection according to the embodiment of the present invention, as shown in fig. 7, specifically including:
the obtaining module 710: acquiring a monitoring image or video;
the detection module 720: detecting the image or the video through a target detection algorithm to obtain detection results of all targets, and obtaining vehicle information in the detection results of all targets;
the detection module 720 is specifically configured to:
and detecting the image or the video through a target detection algorithm to obtain the position information and the category labels of all targets, and obtaining the position information of the vehicle category according to the position information and the category labels of all the targets.
The marking module 730: and judging whether the vehicle has a vehicle jamming behavior or not according to the vehicle information, and marking the jamming behavior.
The marking module 730 is specifically configured to:
determining a vehicle center point according to the vehicle position information, performing straight line fitting on the vehicle center point through Hough transform, processing the fitted straight line, calculating the distance from the vehicle center point to the processed straight line, judging whether a vehicle is jammed according to the distance, and marking the jammed vehicle if the jammed vehicle is jammed.
Determining a vehicle target center point according to vehicle position information, performing straight line fitting on the vehicle center point through Hough transform to obtain a voting result of a parameter variable value interval in a parameter space, performing non-maximum value suppression operation on the voting result, deleting a close straight line, updating the voting result, selecting K pair parameters with the highest vote number in the voting result, calculating a formula of K straight lines of the K pair parameters under an original coordinate system, performing non-lane direction suppression on the K straight lines, deleting straight lines which do not accord with a lane line direction to obtain K straight lines which finally accord with the lane line direction, calculating the distance from the vehicle center point to the K straight lines according to the formula of the K straight lines to obtain K distance values, judging whether the minimum value of the K distance values of each vehicle is in a threshold value interval, and if the minimum value is in the threshold value interval, judging that the target does not run in the lane, and (4) carrying out a jamming action and marking the vehicle target.
The embodiment of the present invention is a system embodiment corresponding to the above method embodiment, and specific operations of each module may be understood with reference to the description of the method embodiment, which is not described herein again.
Apparatus embodiment one
An embodiment of the present invention provides a vehicle jamming behavior recognition apparatus based on target detection, as shown in fig. 8, including: a memory 80, a processor 82 and a computer program stored on the memory 80 and executable on the processor 82, the computer program, when executed by the processor, implementing the steps of the above-described method embodiments.
Device embodiment II
The embodiment of the present invention provides a computer-readable storage medium, on which an implementation program for information transmission is stored, and when being executed by the processor 82, the implementation program implements the steps in the above method embodiments.
The computer-readable storage medium of this embodiment includes, but is not limited to: ROM, RAM, magnetic or optical disks, and the like.
It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; however, these modifications or alternative technical solutions of the embodiments of the present invention do not depart from the scope of the present invention.
Claims (10)
1. A vehicle jamming behavior recognition method based on target detection is characterized by comprising the following steps,
s1, acquiring a monitoring image or video;
s2, detecting the image or the video through a target detection algorithm to obtain detection results of all targets, and acquiring vehicle information in the detection results of all targets;
and S3, judging whether the vehicle has the vehicle jamming behavior according to the vehicle information, and marking the jamming behavior.
2. The method according to claim 1, wherein the S2 specifically includes: and detecting the image or the video through a target detection algorithm to obtain the position information and the category labels of all targets, and obtaining the position information of the vehicle category according to the position information and the category labels of all the targets.
3. The method according to claim 2, wherein the S3 specifically includes:
determining a vehicle center point according to the vehicle position information, performing straight line fitting on the vehicle center point through Hough transform, processing the fitted straight line, calculating the distance from the vehicle center point to the processed straight line, judging whether a vehicle is jammed according to the distance, and marking the jammed vehicle if the jammed vehicle is jammed.
4. The method according to claim 3, wherein the determining a vehicle center point according to the vehicle position information, fitting a straight line to the vehicle center point through Hough transform, processing the fitted straight line, calculating a distance from the vehicle center point to the processed straight line, and determining whether a congestion behavior exists according to the distance, and if the congestion behavior exists, labeling the congested vehicle specifically includes: determining a vehicle target center point according to vehicle position information, performing straight line fitting on the vehicle center point through Hough transform to obtain a voting result of a parameter variable value interval in a parameter space, performing non-maximum value suppression operation on the voting result, deleting a close straight line, updating the voting result, selecting K pair parameters with the highest vote number in the voting result, calculating a formula of K straight lines of the K pair parameters under an original coordinate system, performing non-lane direction suppression on the K straight lines, deleting straight lines which do not accord with a lane line direction to obtain K straight lines which finally accord with the lane line direction, calculating the distance from the vehicle center point to the K straight lines according to the formula of the K straight lines to obtain K distance values, judging whether the minimum value of the K distance values of each vehicle is in a threshold value interval, and if the minimum value is in the threshold value interval, judging that the target does not run in the lane, and (4) carrying out a jamming action and marking the vehicle target.
5. A vehicle jamming behavior recognition system based on object detection, comprising:
an acquisition module: acquiring a monitoring image or video;
a detection module: detecting the image or the video through a target detection algorithm to obtain detection results of all targets, and obtaining vehicle information in the detection results of all targets;
a marking module: and judging whether the vehicle has a vehicle jamming behavior or not according to the vehicle information, and marking the jamming behavior.
6. The system of claim 5, wherein the detection module is specifically configured to:
and detecting the image or the video through a target detection algorithm to obtain the position information and the category labels of all targets, and obtaining the position information of the vehicle category according to the position information and the category labels of all the targets.
7. The system of claim 6, wherein the tagging module is specifically configured to:
determining a vehicle center point according to the vehicle position information, performing straight line fitting on the vehicle center point through Hough transform, processing the fitted straight line, calculating the distance from the vehicle center point to the processed straight line, judging whether a vehicle is jammed according to the distance, and marking the jammed vehicle if the jammed vehicle is jammed.
8. The system of claim 7, wherein the tagging module is specifically configured to:
determining a vehicle target center point according to vehicle position information, performing straight line fitting on the vehicle center point through Hough transform to obtain a voting result of a parameter variable value interval in a parameter space, performing non-maximum value suppression operation on the voting result, deleting a close straight line, updating the voting result, selecting K pair parameters with the highest vote number in the voting result, calculating a formula of K straight lines of the K pair parameters under an original coordinate system, performing non-lane direction suppression on the K straight lines, deleting straight lines which do not accord with a lane line direction to obtain K straight lines which finally accord with the lane line direction, calculating the distance from the vehicle center point to the K straight lines according to the formula of the K straight lines to obtain K distance values, judging whether the minimum value of the K distance values of each vehicle is in a threshold value interval, and if the minimum value is in the threshold value interval, judging that the target does not run in the lane, and (4) carrying out a jamming action and marking the vehicle target.
9. A vehicle jamming behavior recognition apparatus based on object detection, characterized by comprising: memory, a processor and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the steps of the object detection based vehicle jamming behavior recognition method according to any of claims 1 to 4.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon an information transfer-implementing program, which when executed by a processor implements the steps of the object detection-based vehicle congestion behavior recognition method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111042394.3A CN113743316B (en) | 2021-09-07 | 2021-09-07 | Vehicle plugging behavior identification method, system and device based on target detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111042394.3A CN113743316B (en) | 2021-09-07 | 2021-09-07 | Vehicle plugging behavior identification method, system and device based on target detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113743316A true CN113743316A (en) | 2021-12-03 |
| CN113743316B CN113743316B (en) | 2023-09-19 |
Family
ID=78736396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111042394.3A Active CN113743316B (en) | 2021-09-07 | 2021-09-07 | Vehicle plugging behavior identification method, system and device based on target detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113743316B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116052440A (en) * | 2023-02-03 | 2023-05-02 | 深圳市旗扬特种装备技术工程有限公司 | Vehicle intention plug behavior identification method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004173195A (en) * | 2002-11-22 | 2004-06-17 | Fuji Heavy Ind Ltd | Device and method for monitoring vehicle |
| CN109949578A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of vehicle crimping based on deep learning |
| CN111398924A (en) * | 2020-04-29 | 2020-07-10 | 上海英恒电子有限公司 | Radar installation angle calibration method and system |
| CN111523464A (en) * | 2020-04-23 | 2020-08-11 | 上海眼控科技股份有限公司 | Method and device for detecting illegal lane change of vehicle |
| CN112163543A (en) * | 2020-10-12 | 2021-01-01 | 北京易华录信息技术股份有限公司 | Method and system for detecting illegal lane occupation of vehicle |
| CN112820115A (en) * | 2021-01-05 | 2021-05-18 | 四川铁投信息技术产业投资有限公司 | Method for identifying steering state of running vehicle |
-
2021
- 2021-09-07 CN CN202111042394.3A patent/CN113743316B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004173195A (en) * | 2002-11-22 | 2004-06-17 | Fuji Heavy Ind Ltd | Device and method for monitoring vehicle |
| CN109949578A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of vehicle crimping based on deep learning |
| CN111523464A (en) * | 2020-04-23 | 2020-08-11 | 上海眼控科技股份有限公司 | Method and device for detecting illegal lane change of vehicle |
| CN111398924A (en) * | 2020-04-29 | 2020-07-10 | 上海英恒电子有限公司 | Radar installation angle calibration method and system |
| CN112163543A (en) * | 2020-10-12 | 2021-01-01 | 北京易华录信息技术股份有限公司 | Method and system for detecting illegal lane occupation of vehicle |
| CN112820115A (en) * | 2021-01-05 | 2021-05-18 | 四川铁投信息技术产业投资有限公司 | Method for identifying steering state of running vehicle |
Non-Patent Citations (4)
| Title |
|---|
| DUAN T D等: "Building an automatic vehicle license plate recognition system", 《PROC. INT. CONF. COMPUT. SCI. RIVF.》 * |
| V. S. DEV等: "Steering angle estimation for autonomous vehic", 《INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING, COMMUNICATIONS AND INFORMATICS (ICACCI)》 * |
| 唐悦: "基于视频监控的危险特征检测技术研究", 《中国优秀硕士学位论文全文数据库：工程科技Ⅱ辑》 * |
| 孙悦: "基于单目视觉的车道边缘检测及中心线提取算法研究", 《汽车实用技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116052440A (en) * | 2023-02-03 | 2023-05-02 | 深圳市旗扬特种装备技术工程有限公司 | Vehicle intention plug behavior identification method, device, equipment and storage medium |
| CN116052440B (en) * | 2023-02-03 | 2024-02-02 | 深圳市旗扬特种装备技术工程有限公司 | Vehicle intention plug behavior identification method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113743316B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11455805B2 (en) | Method and apparatus for detecting parking space usage condition, electronic device, and storage medium | |
| CN109087510B (en) | Traffic monitoring method and device | |
| CN111382704B (en) | Vehicle line pressing violation judging method and device based on deep learning and storage medium | |
| CN111898491B (en) | Identification method and device for reverse driving of vehicle and electronic equipment | |
| CN110956169A (en) | License plate recognition method and device and electronic equipment | |
| CN110909692A (en) | Abnormal license plate recognition method and device, computer storage medium and electronic equipment | |
| CN109344886B (en) | Occlusion number plate distinguishing method based on convolutional neural network | |
| CN111274926B (en) | Image data screening method, device, computer equipment and storage medium | |
| CN111524350B (en) | Method, system, terminal device and medium for detecting abnormal driving condition of vehicle and road cooperation | |
| CN111967396A (en) | Processing method, device and equipment for obstacle detection and storage medium | |
| CN113011331A (en) | Method and device for detecting whether motor vehicle gives way to pedestrians, electronic equipment and medium | |
| CN113158922A (en) | Traffic flow statistical method, device and equipment based on YOLO neural network | |
| CN112699711B (en) | Lane line detection method and device, storage medium and electronic equipment | |
| Lee | Neural network approach to identify model of vehicles | |
| CN113743316B (en) | Vehicle plugging behavior identification method, system and device based on target detection | |
| CN113435350A (en) | Traffic marking detection method, device, equipment and medium | |
| CN114724119B (en) | Lane line extraction method, lane line detection device, and storage medium | |
| CN116863711A (en) | Lane flow detection method, device, equipment and medium based on highway monitoring | |
| CN111160183A (en) | Method and device for detecting red light running of vehicle | |
| Kodwani et al. | Automatic license plate recognition in real time videos using visual surveillance techniques | |
| CN111627224A (en) | Vehicle speed abnormality detection method, device, equipment and storage medium | |
| KR102435435B1 (en) | System for searching numbers of vehicle and pedestrian based on artificial intelligence | |
| CN107993446A (en) | A kind of traffic prohibition parking area domain parking offense monitoring device | |
| CN114119953A (en) | Method for quickly positioning and correcting license plate, storage medium and equipment | |
| CN112597924A (en) | Electric bicycle track tracking method, camera device and server |
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 |