CN112862856A - Method, device and equipment for identifying illegal vehicle and computer readable storage medium - Google Patents
Method, device and equipment for identifying illegal vehicle and computer readable storage medium Download PDFInfo
- Publication number
- CN112862856A CN112862856A CN201911183772.2A CN201911183772A CN112862856A CN 112862856 A CN112862856 A CN 112862856A CN 201911183772 A CN201911183772 A CN 201911183772A CN 112862856 A CN112862856 A CN 112862856A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- image
- identifying
- pictures
- illegal
- 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.)
- Pending
Links
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
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Abstract
The embodiment of the application provides an identification method, an identification device, identification equipment and a computer readable storage medium of an illegal parking vehicle, which are used for improving the identification efficiency of the illegal parking vehicle. The method for identifying the illegal vehicle comprises the following steps: acquiring an image shot by a camera arranged on a vehicle; identifying a delinquent region in an image; identifying vehicles in the parking violation area in the image; carrying out vehicle tracking on the vehicle in the image to obtain a motion track of wheels of the vehicle; calculating a detection value of the motion track, wherein the detection value is used for indicating the change amplitude of the motion track; when the detection value is less than the threshold value, the vehicle is identified as an illicit vehicle.
Description
Technical Field
The present application relates to the field of image recognition technologies, and in particular, to a method, an apparatus, a device, and a computer-readable storage medium for recognizing an illegal vehicle.
Background
Along with the increase of vehicles, the burden on road traffic is increased, wherein the vehicle parking violation not only easily causes road congestion, but also improves traffic safety risks, and therefore the management of the vehicle parking violation has important significance on the management of the road traffic.
The traditional on-site manual law enforcement mode is included in the capturing mode of the illegal parking vehicle, the labor cost is high, the coverage range is limited, and the traditional manual law enforcement mode is gradually replaced by the fixed-point video monitoring nowadays.
In fixed-point video monitoring, when the motion state of a vehicle is estimated and judged based on the posture of a camera so as to judge whether the vehicle is in an illegal parking state, complex data processing amount is needed, and the recognition efficiency of the illegal parking vehicle is low.
Disclosure of Invention
The embodiment of the application provides an identification method, an identification device, identification equipment and a computer readable storage medium of an illegal parking vehicle, which are used for improving the identification efficiency of the illegal parking vehicle.
In a first aspect, an embodiment of the present application provides a method for identifying an illegal vehicle, where the method includes:
acquiring an image shot by a camera arranged on a vehicle;
identifying a delinquent region in an image;
identifying vehicles in the parking violation area in the image;
carrying out vehicle tracking on the vehicle in the image to obtain a motion track of wheels of the vehicle;
calculating a detection value of the motion track, wherein the detection value is used for indicating the change amplitude of the motion track;
when the detection value is less than the threshold value, the vehicle is identified as an illicit vehicle.
In some embodiments, vehicle tracking the vehicle in the image, and obtaining the motion track of the wheel of the vehicle includes:
selecting N frames of pictures containing vehicles from the images;
detecting N frames of pictures in a preset detection range based on an angular point detection algorithm to obtain a plurality of angular points of the wheel;
and calculating the optical flow of the angular points based on an optical flow algorithm to be used as the motion trail of the wheels of the vehicle.
In some embodiments, the image comprises a plurality of frames of pictures, and extracting N frames of pictures containing the vehicle from the image in the image comprises:
sequentially detecting the overlapping degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image through a vehicle detection frame;
detecting the correlation degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image by a correlation filtering algorithm in sequence;
and selecting the adjacent pictures of the two frames before and after the image with the overlapping degree and the correlation degree meeting the requirements as N frames of pictures.
In some embodiments, the method further comprises:
generating illegal parking event information corresponding to the illegal parking vehicles according to the N frames of pictures;
and outputting the violation event information.
In some embodiments, calculating the detection value of the motion trajectory includes:
calculating a track central point of the motion track;
and calculating the track variance of the motion track according to the track central point, and taking the track variance as a detection value.
In some embodiments, identifying a vehicle in the parking violation area in the image comprises:
and identifying the vehicle from the image based on a target detection network, wherein the target detection network is obtained by training vehicle information added with vehicle labels.
In some embodiments, acquiring an image captured by a camera provided in a vehicle includes:
and receiving the image shot by the camera transmitted by the camera based on the network connection with the camera arranged on the vehicle.
In a second aspect, an embodiment of the present application further provides an apparatus for identifying an illegal vehicle, where the apparatus includes:
the device comprises an acquisition unit, a processing unit and a display unit, wherein the acquisition unit is used for acquiring an image shot by a camera arranged on a vehicle;
an identification unit for identifying a delinquent region in an image;
an identification unit further configured to identify a vehicle in the parking violation area in the image;
the tracking unit is used for carrying out vehicle tracking on the vehicle in the image to obtain the motion trail of the wheels of the vehicle;
the calculating unit is used for calculating a detection value of the motion track, and the detection value is used for indicating the change amplitude of the motion track;
and the identification unit is also used for identifying the vehicle as an illegal vehicle when the detection value is smaller than the threshold value.
In some embodiments, the tracking unit is specifically configured to:
selecting N frames of pictures containing vehicles from the images;
detecting N frames of pictures in a preset detection range based on an angular point detection algorithm to obtain a plurality of angular points of the wheel;
and calculating the optical flow of the angular points based on an optical flow algorithm to be used as the motion trail of the wheels of the vehicle.
In some embodiments, the tracking unit is specifically configured to:
sequentially detecting the overlapping degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image through a vehicle detection frame;
detecting the correlation degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image by a correlation filtering algorithm in sequence;
and selecting the adjacent pictures of the two frames before and after the image with the overlapping degree and the correlation degree meeting the requirements as N frames of pictures.
In some embodiments, the apparatus further comprises:
the generating unit is used for generating illegal parking event information corresponding to illegal parking vehicles according to the N frames of pictures;
and the output unit is used for outputting the violation event information.
In some embodiments, the computing unit is specifically configured to:
calculating a track central point of the motion track;
and calculating the track variance of the motion track according to the track central point, and taking the track variance as a detection value.
In some embodiments, the identification unit is specifically configured to:
and identifying the vehicle from the image based on a target detection network, wherein the target detection network is obtained by training vehicle information added with vehicle labels.
In some embodiments, the obtaining unit is specifically configured to:
and receiving the image shot by the camera transmitted by the camera based on the network connection with the camera arranged on the vehicle.
In a third aspect, an embodiment of the present application further provides a processing device, which includes a processor and a memory, where the memory stores a computer program, and the processor executes, when calling the computer program in the memory, any one of the steps in the method for identifying an illegal vehicle provided by the embodiment of the present application.
In a fourth aspect, the present application further provides a computer-readable storage medium, where a plurality of instructions are stored, and the instructions are adapted to be loaded by a processor to perform the steps in any of the methods for identifying an illegal vehicle provided by the present application.
From the above, the present application has the following advantageous effects:
compared with the prior art in which the motion state of the vehicle can be judged through camera-based attitude estimation, complex data processing amount is needed, the calculation cost and the time cost are higher, and in the process of identifying the illegal parking vehicle, the motion track of the wheels of the vehicle is directly tracked, and the motion state of the vehicle is judged according to the detection value of the motion track of the wheels, so that the illegal parking vehicle can be effectively identified, the data processing amount can be reduced, the illegal parking vehicle can be conveniently identified, and the identification efficiency of the illegal parking vehicle is greatly improved.
In addition, the camera is arranged on the vehicle to acquire the images of the vehicles violating the parking, so that the characteristics of the vehicles running on different traffic roads can be utilized, a large number of images of the vehicles violating the parking can be acquired conveniently, and the coverage range of the vehicles violating the parking is enlarged; the cameras are convenient to deploy on the vehicles, the cost is low, high deployment cost required for deploying the cameras at fixed points in the prior art can be avoided, and meanwhile, as the distance between the vehicles is easy to be shortened, a clearer image can be obtained through shooting, so that the flexibility in application is obviously improved, and the automatic snapshot of the illegal parking vehicles is convenient to popularize.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of the method for identifying an illegal vehicle according to the present application;
FIG. 2 is a schematic flow chart of step S104 of the method for identifying an illegal vehicle according to the present application;
FIG. 3 is a schematic flow chart of step S201 in the method for identifying an illegal vehicle according to the present application;
FIG. 4 is a schematic flow chart of step S105 of the method for identifying an illegal vehicle according to the present application;
FIG. 5 is a schematic diagram of an exemplary embodiment of an apparatus for recognizing an illegal vehicle;
FIG. 6 is a schematic diagram of a processing apparatus according to the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. 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 application.
In the description that follows, specific embodiments of the present application will be described with reference to steps and symbols executed by one or more computers, unless otherwise indicated. Accordingly, these steps and operations will be referred to, several times, as being performed by a computer, the computer performing operations involving a processing unit of the computer in electronic signals representing data in a structured form. This operation transforms the data or maintains it at locations in the computer's memory system, which may be reconfigured or otherwise altered in a manner well known to those skilled in the art. The data maintains a data structure that is a physical location of the memory that has particular characteristics defined by the data format. However, while the principles of the application have been described in language specific to above, it is not intended to be limited to the specific form set forth herein, and it will be recognized by those of ordinary skill in the art that various of the steps and operations described below may be implemented in hardware.
The principles of the present application may be employed in numerous other general-purpose or special-purpose computing, communication environments or configurations. Examples of well known computing systems, environments, and configurations that may be suitable for use with the application include, but are not limited to, hand-held telephones, personal computers, servers, multiprocessor systems, microcomputer-based systems, mainframe-based computers, and distributed computing environments that include any of the above systems or devices.
The terms "first", "second", and "third", etc. in this application are used to distinguish between different objects and not to describe a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions.
First, before the embodiments of the present application are described, the relevant contents of the present application about the application background will be described.
The executing main body of the method for identifying the illegal vehicle can be the identification device of the illegal vehicle provided by the application, or the server Equipment, the physical host, the vehicle-mounted terminal or the User Equipment (UE) and other processing Equipment integrated with the identification device of the illegal vehicle, wherein the identification device of the illegal vehicle can be realized in a hardware or software mode, and the UE can be a terminal Equipment such as a smart phone, a tablet computer, a notebook computer, a palm computer, a desktop computer or a Personal Digital Assistant (PDA).
The processing equipment can adopt a working mode of independent operation or a working mode of equipment cluster, and the method for identifying the illegal vehicle provided by the application can obviously improve the flexibility and the identification efficiency of automatic snapshot of the illegal vehicle. In addition, a system for identifying the illegal parking vehicles can be built based on a plurality of processing devices, and the system is used for online management of sharing, synchronization and the like of the related information of the illegal parking vehicles.
Next, the method for identifying an illegal vehicle provided by the present application will be described.
Referring to fig. 1, fig. 1 shows a schematic flow chart of the method for identifying an illegal vehicle provided by the present application, and the method for identifying an illegal vehicle provided by the present application may specifically include the following steps:
step S101, acquiring an image shot by a camera arranged on a vehicle;
in practical application, the processing equipment applying the method for identifying the illegal vehicle can directly comprise a camera of the vehicle on hardware, locally stores an image shot by the camera and can directly read the image in the equipment; or the processing equipment can also establish network connection with a camera of the vehicle and acquire an image shot by the camera on line from the camera according to the network connection; alternatively, the processing device may also read out the image captured by the camera from a related storage medium storing the image captured by the camera of the vehicle, and the specific acquisition mode is not limited herein.
The vehicle provided with the camera may be a logistics vehicle of a logistics company as an example. The commodity circulation vehicle carries out the vehicle that the commodity circulation was transported for the commodity circulation company, can be the freight train or the electric vehicle of different models, is taking a photograph the camera in the commodity circulation vehicle after, and usable commodity circulation vehicle is applicable to different traffic routes and longer travel time's characteristics, gathers the image of a large amount of vehicles of violating the law conveniently, enlarges the coverage of taking a candid photograph the vehicle of violating the law. In addition, the camera is shot in the logistics vehicle, and unified management can be conveniently carried out through a logistics company.
The camera can shoot images according to a preset shooting mode, for example, shooting height, shooting direction or shooting distance can be set, the specific shooting mode can be adjusted according to the camera, and the camera is not limited specifically. The images shot by the camera are composed of multiple frames of pictures, and videos can be composed through time lines.
It is to be understood that the description of the related actions in the following description is omitted for the sake of convenience and brevity, and is not directly described.
Step S102, identifying an illegal parking area in an image;
after an image obtained by shooting through a camera of a vehicle is obtained, information such as a parking prohibition identifier, a long-time parking prohibition identifier, a time-limited section (long) parking, a special parking space, a diversion line, a bus lane or a non-motor vehicle lane in the image can be identified, and then an illegal parking area in the image can be identified.
Step S103, identifying vehicles in the illegal parking area in the image;
after the illegal parking areas in the image are identified, vehicles in the illegal parking areas in the image are continuously identified again, and vehicles in the illegal parking areas can be identified by identifying information such as vehicle body outlines, license plates or vehicle types and the like in the image.
It is to be understood that, at this time, although the vehicle is in the parking violation area, the parking violation event does not necessarily occur, and therefore, it is necessary to perform dynamic identification to identify whether the vehicle is staying in the parking violation area or traveling through the parking violation area.
Step S104, carrying out vehicle tracking on the vehicle in the image to obtain the motion track of the wheels of the vehicle;
in the process of dynamically identifying the vehicle, the vehicle is tracked in the image to track the motion track of the vehicle, and it can be understood that during tracking, a plurality of dense tracking points are identified and the motion track of the tracking points is tracked, in the process, the wheel can be determined according to the wheel image characteristics of the wheel, for example, the wheel contour or the position of the wheel on the vehicle body, and the wheel of the vehicle is tracked, so that the motion track of the wheel of the vehicle can be tracked.
Step S105, calculating a detection value of the motion track;
wherein the detection value is used for indicating the change amplitude of the motion trail.
After the motion trail of the wheel is obtained, a detection value indicating the variation amplitude of the motion trail can be calculated through a related calculation formula.
And step S106, when the detection value is smaller than the threshold value, identifying the vehicle as an illegal parking vehicle.
When the calculated detection value is smaller than a preset threshold value, the current vehicle can be identified as an illegal vehicle, and an illegal parking event is determined to be generated by the vehicle.
Compared with the prior art that the motion state of the vehicle can be judged through the posture estimation based on the camera, the complex data processing amount is needed, and the calculation cost and the time cost are high, in the process of identifying the illegal parking vehicle, the motion state of the vehicle is judged directly through tracking the motion track of the wheels of the vehicle and the detection value of the motion track of the wheels, so that the illegal parking vehicle can be effectively identified, the data processing amount can be reduced, the illegal parking vehicle can be conveniently identified, and the identification efficiency of the illegal parking vehicle is greatly improved.
In addition, the data processing amount of the illegal parking vehicle identification is greatly reduced, and the identification device or the processing equipment applying the illegal parking vehicle identification method provided by the application can be allowed to sink to the processing equipment with limited data processing capacity, such as the vehicle-mounted terminal and the UE, so that the flexibility and the practicability are higher.
In addition, the camera is arranged on the vehicle to acquire the images of the vehicles violating the parking, so that the characteristics that the vehicles can run on different traffic roads can be utilized, a large number of images of the vehicles violating the parking can be acquired conveniently, and the coverage range of the vehicles violating the parking is enlarged; the cameras are convenient to deploy on the vehicles, the cost is low, high deployment cost required for deploying the cameras at fixed points in the prior art can be avoided, and meanwhile, as the distance between the vehicles is easy to be shortened, a clearer image can be obtained through shooting, so that the flexibility in application is obviously improved, and the automatic snapshot of the illegal parking vehicles is convenient to popularize.
Continuing to refer to fig. 2, fig. 2 shows another schematic flow chart of the method for identifying an illegal vehicle provided by the present application, and in some embodiments, the above-mentioned diagram corresponds to step S104 of the embodiment, and specifically may include the following steps:
step S201, selecting N frames of pictures containing vehicles from the images;
after identifying a vehicle in the parking violation area in the image, a picture of a plurality of frames including the vehicle may be selected from the image, where N may be used to represent the number of selected pictures.
Wherein, all pictures containing the vehicle can be used as the object for executing vehicle tracking; alternatively, the pictures may be selected from all the pictures including the vehicle in a selection manner such as a relevant selection ratio, a number of selected pictures, or a number of interval frames.
Step S202, detecting N frames of pictures in a preset detection range based on an angular point detection algorithm to obtain a plurality of angular points of the wheel;
and calling a preset angular point Detection algorithm Commer Detection, and based on the preset Detection range of the picture, performing angular point Detection on the wheels by the N frames of pictures to obtain angular points of a plurality of wheels.
The angular point is a specific image feature point, such as an intersection point of two or more edges, a point at which an image gradient rate or a gradient change rate reaches a threshold value, a point at which an object edge is discontinuous, or a pixel point corresponding to a local maximum of a first-order derivative (i.e., a gradient of a gray level), and the like.
The corner detection algorithm can be particularly the corner detection based on a gray-scale image, the corner detection based on the gray-scale image can be divided into 3 types based on gradient, template or template gradient combination, the existence of the corner is judged by calculating the curvature of the edge based on the gradient method, the size of the corner calculation value is not only related to the edge strength, but also related to the change rate of the edge direction, and the noise ratio of the method is more sensitive to noise than the corner detection method based on the template; the template-based method mainly considers the gray scale change of pixel neighborhood points, namely the change of image brightness, and defines points with enough brightness contrast with the neighborhood points as angular points, such as a Kitchen-Rosenfeld angular point detection algorithm, a Harris angular point detection algorithm, a KLT angular point detection algorithm and a SUSAN angular point detection algorithm; a Canny edge mapping image is obtained through convolution of a Gaussian template and an image based on a template gradient combination method, then the vector potential is obtained through calculation of the gradient and an edge vector, and whether the image is a saddle point or not is judged through calculation of Gaussian curvature and average curvature of the vector potential, and the image is supposed to correspond to an angular point of the image.
Or, the corner detection algorithm may also be a type of algorithm such as corner detection based on a binary image or corner detection based on a contour curve, and the like, and the method based on a binary image may regard the original image as a polygon, so that the corners of the polygon are necessarily on the extension line of the skeleton, the maximum disc radius of the skeleton point corresponding to the corner point should tend to 0, and the point where the maximum disc in the skeleton is 0 is detected, i.e. the corner point; the method based on the contour curve can directly calculate the discrete curve, or can perform piecewise fitting on the original curve by a function, and then calculate the curvature extreme value of the curve according to the fitted curve piecewise equation to obtain the position of the corner point.
In step S203, an optical flow at the corner is calculated based on an optical flow algorithm as a motion trajectory.
And then, a preset Optical Flow algorithm Optical Flow is called, the Optical flows of the angular points are calculated and used as the motion trail of the wheel, and the Optical Flow algorithm is suitable for motion tracking and has higher detection precision.
Taking the motion track of one corner point in two pictures as an example, the coordinate of the corner point in the current picture is (x)i，yi) The angular point displacement is (Deltax)i，Δyi) The motion locus of the corner point is ((x)i，yi)，(xi+Δxi，yi+Δyi) The motion track of the corner point in the N frames of pictures is calculated in turn and is recorded as (P)i) I ∈ (0, N), which is used for the calculation of the subsequent detection value.
Further, in some embodiments, referring to another flow diagram of the method for identifying an illegal parking vehicle shown in fig. 3, step S201 in the embodiment of fig. 2 may specifically include the following steps:
step S301, sequentially detecting the overlapping degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame through a vehicle detection frame from the image;
in the process of selecting the picture containing the vehicle in the illegal parking area, vehicle tracking can be carried out to identify the picture containing the vehicle.
Specifically, the vehicle detection frame can be used for detecting the overlapping degree of the vehicles contained in the front and rear frames of pictures of each group, so as to obtain the overlapping degree of the vehicles contained in the front and rear frames of pictures of the plurality of groups.
The detection frame can be understood as a rectangular outline selected from a plurality of coordinate points in the picture, and the picture information in the rectangular outline is a tracking range including the vehicle and is used for detecting the overlapping degree of the vehicles included in the front and rear frames of pictures. And the overlapping degree is used for representing the overlapping degree of the vehicles contained in the front and rear frames of pictures. Step S302, detecting the correlation degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the images sequentially through a correlation filtering algorithm;
besides the overlapping degree, the correlation degree detection can be carried out on the vehicles contained in the front and rear frames of pictures of each group, so that the correlation degrees of the vehicles contained in the front and rear frames of pictures of the plurality of groups are obtained.
The related filtering algorithm may be a circular Structure of Tracking-by-Detection with Kernel (CSK) algorithm, and the CSK algorithm has the advantages of high learning capability and processing speed, and can be conveniently and accurately used for detecting the correlation of the vehicle.
It should be understood that the execution sequence of the steps S301 and S302 is not limited, that is, the step S301 may be executed, and then the step S302 may be executed; alternatively, step S302 may be executed first, and step S301 may be executed; alternatively, step S301 and step S302 may be executed simultaneously, and the method is not limited to this.
Step S303, select two adjacent pictures of different front and back frames with overlapping degree and correlation degree meeting the requirement as N frames of pictures.
It is easily understood that the requirement for the degree of overlap is above the overlap threshold and the requirement for the degree of correlation is above the correlation threshold.
After the overlapping degrees and the correlation degrees of the multiple groups of front and rear frames of pictures are obtained, pictures which meet the requirements can be screened out according to the preset overlapping degrees and the preset correlation degrees, so that even if the overlapping degrees are high, the lower correlation degree can be judged to be unmatched, and the vehicle is judged to exit, so that the vehicles in the front and rear pictures can be ensured to be the same vehicle, and the vehicle tracking effect is achieved.
Of course, in the process of this screening, all the pictures (i.e., N frames of pictures) including the vehicle may also be used as the object for performing vehicle tracking; alternatively, the N-frame images may be selected from all the images including the vehicle in a selection manner such as a relevant selection ratio, a number of selected images, or a number of interval frames.
Further, in some embodiments, since the pictures are composed of pixels, in practical applications, the preset detection range mentioned in step S202 of the embodiment in fig. 2 may be specifically the pixels with the number of pixel rows from the total number of 2/3 pixels and the number of pixel columns from the total number of 3/4 pixels in each picture, and since the parking violation region and the parking violation vehicle usually appear at the lower right of the vehicle during the operation of the vehicle, the corner detection may be performed on the lower right region of the picture according to the detection range, so that the effective detection of the corner can be greatly improved, the data processing amount thereof can be reduced, and the detection efficiency of the corner detection can be improved.
With reference to fig. 4, still another flow chart of the method for identifying an illegal parking vehicle according to the present application is shown, in some embodiments, step S105 in the embodiment corresponding to fig. 1 may specifically include the following steps:
step S401, calculating a track central point of a motion track;
in some embodiments, the manner related to calculating the center point of the trajectory may be specifically implemented by a first formula, where the first formula is:
After obtaining the movement trajectory of the wheels of the vehicle by means of the optical flow method, i.e. a plurality of sets (P)i) I e (0, N), the center point between different corner points, i.e. the center point of the track, can be calculated first.
Step S402, calculating the track variance of the motion track according to the track central point, and taking the track variance as a detection value;
in some embodiments, calculating the trajectory variance according to the trajectory center point may specifically be implemented by a second formula, where the second formula is:
After the track center points between the different corner points are obtained through calculation, the track variance of the motion track of the wheel can be calculated according to the track center points, and the track variance can be used for indicating the variation amplitude of the motion track of the wheel.
It is easy to understand that when the vehicle is stationary, the trajectory lines of different angular points of the wheel are more consistent, the trajectory center point and the trajectory line are also more consistent, and thus the trajectory variance is smaller, so that the vehicle can be judged to stay in an illegal parking area; when the vehicle is in a moving state, the curvature of the track lines of different corner points of the wheel is large, the track center point is inconsistent with the track lines, and thus the track variance is large, so that the vehicle can be judged to pass through an illegal parking area.
In some embodiments, step S103 in the embodiment corresponding to fig. 1 may specifically include the following steps:
and identifying the vehicle from the image based on a target detection network, wherein the target detection network is trained by the vehicle information added with the vehicle label.
It can be understood that the target detection network can be obtained by training vehicle information added with vehicle labels in advance, so that when vehicles in the parking violation area are identified, the model can be directly called to identify the vehicles, and specific information of the vehicles can be obtained.
Taking the target detection network YOLOv3 as an example, the target detection network YOLOv3 comprises a single-stage end-to-end algorithm, and has the advantages of high accuracy and high calculation speed. Of course, other detection algorithms may be used to identify the vehicle besides the target detection network YOLOv3, and the specific details are not limited herein.
The method comprises the steps of preparing pictures of the vehicle under the conditions of different angles, different colors, different environments and the like for different vehicles, adding vehicle labels such as 'small SUV-popular T-Cross armor-golden' for the pictures, taking the pictures with the vehicle labels as vehicle information, and training a target detection network, so that after the pictures containing the vehicles are input into the target detection network, the target detection network can identify the vehicles in the pictures and the specific information thereof, and for example, the target detection network can output 'the vehicles detected by the current pictures are small SUV-popular T-Cross armor-golden'.
In some embodiments, the method for identifying an illegal vehicle may further include, after identifying the illegal vehicle, the following steps:
generating illegal parking event information corresponding to the illegal parking vehicles according to the N frames of pictures;
and outputting the violation event information.
It can be understood that after the vehicle is determined to be illegal, illegal event information can be generated according to the N frames of pictures of the vehicle, and in the illegal event information, the related images when the illegal vehicle causes illegal events are carried, that is, the effective pictures of the illegal vehicle in the N frames of pictures provide data support for recording of the illegal events and illegal management.
Specifically, the output may be output to a related UE, an illegal parking reporting platform, an illegal parking reporting system, or the like, which is not limited herein.
In order to better implement the method for identifying the illegal vehicle, the embodiment of the application also provides a device for identifying the illegal vehicle.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an identification device for an illegal vehicle according to the present application, wherein the identification device for an illegal vehicle specifically includes the following structure:
an obtaining unit 501, configured to obtain an image captured by a camera provided in a vehicle;
an identifying unit 502 for identifying an illegal region in an image;
an identification unit 502, further configured to identify a vehicle in the parking violation area in the image;
a tracking unit 503, configured to perform vehicle tracking on the vehicle in the image to obtain a motion track of a wheel of the vehicle;
a calculating unit 504, configured to calculate a detection value of the motion trajectory, where the detection value is used to indicate a variation amplitude of the motion trajectory;
the identifying unit 502 is further configured to identify the vehicle as an illegal vehicle when the detection value is smaller than the threshold value.
In some embodiments, the tracking unit 503 is specifically configured to:
selecting N frames of pictures containing vehicles from the images;
detecting N frames of pictures in a preset detection range based on an angular point detection algorithm to obtain a plurality of angular points of the wheel;
and calculating the optical flow of the corner points based on an optical flow algorithm to serve as the motion trail.
In some embodiments, the tracking unit 503 is specifically configured to:
sequentially detecting the overlapping degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image through a vehicle detection frame;
detecting the correlation degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image by a correlation filtering algorithm in sequence;
and selecting the adjacent pictures of the two frames before and after the image with the overlapping degree and the correlation degree meeting the requirements as N frames of pictures.
In some embodiments, the apparatus further comprises:
a generating unit 505, configured to generate illegal parking event information corresponding to the illegal parking vehicle according to the N frames of pictures;
an output unit 506, configured to output the violation event information.
In some embodiments, the calculating unit 504 is specifically configured to:
calculating a track central point of the motion track;
and calculating the track variance of the motion track according to the track central point, and taking the track variance as a detection value.
In some embodiments, the identifying unit 502 is specifically configured to:
and identifying the vehicle from the image based on a target detection network, wherein the target detection network is obtained by training vehicle information added with vehicle labels.
In some embodiments, the obtaining unit 501 is specifically configured to:
and receiving the image shot by the camera transmitted by the camera based on the network connection with the camera arranged on the vehicle.
The present application further provides a processing device, and referring to fig. 6, fig. 6 shows a schematic structural diagram of the processing device of the present application, and specifically, the processing device of the present application includes a processor 601, where the processor 601 is configured to execute a computer program stored in a memory 602 to implement the steps of the method for identifying an illegal vehicle according to any of the embodiments corresponding to fig. 1 to fig. 4; alternatively, the processor 601 is configured to implement the functions of the units in the corresponding embodiment of fig. 5 when executing the computer program stored in the memory 602.
Illustratively, a computer program may be partitioned into one or more modules/units, which are stored in the memory 602 and executed by the processor 601 to accomplish the present application. One or more modules/units may be a series of computer program instruction segments capable of performing certain functions, the instruction segments being used to describe the execution of a computer program in a computer device.
The processing device may include, but is not limited to, a processor 601, a memory 602. Those skilled in the art will appreciate that the illustration is merely an example of a processing device and does not constitute a limitation of the processing device and may include more or less components than those illustrated, or combine certain components, or different components, for example, the processing device may also include an input output device, a network access device, a bus, etc., through which the processor 601, the memory 602, the input output device, the network access device, etc., are connected.
The Processor 601 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor being the control center for the processing device and the various interfaces and lines connecting the various parts of the overall processing device.
The memory 602 may be used for storing computer programs and/or modules, and the processor 601 may implement various functions of the computer apparatus by executing or executing the computer programs and/or modules stored in the memory 602 and calling data stored in the memory 602. The memory 602 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, video data, etc.) created according to the use of the processing device, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described identification device for an illegal vehicle, the processing device and the corresponding units thereof may refer to the description of the identification method for an illegal vehicle in any embodiment corresponding to fig. 1 to 4, and are not repeated herein.
It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor.
For this reason, an embodiment of the present application provides a computer-readable storage medium, where a plurality of instructions are stored, where the instructions can be loaded by a processor to execute steps in the method for identifying an illegal vehicle in any embodiment corresponding to fig. 1 to 4 in the present application, and specific operations may refer to descriptions of the method for identifying an illegal vehicle in any embodiment corresponding to fig. 1 to 4, which are not repeated herein.
Wherein the computer-readable storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.
Since the instructions stored in the computer-readable storage medium can execute the steps in the method for identifying an illegal vehicle according to any embodiment of fig. 1 to 4, the beneficial effects that can be achieved by the method for identifying an illegal vehicle according to any embodiment of fig. 1 to 4 can be achieved, which are described in detail in the foregoing description and are not repeated herein.
The method, the apparatus, the processing device and the computer-readable storage medium for recognizing an illegal vehicle provided by the present application are described in detail above, and specific examples are applied herein to illustrate the principles and embodiments of the present application, and the description of the above embodiments is only used to help understand the method and the core ideas of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (10)
1. A method for identifying an offending vehicle, the method comprising:
acquiring an image shot by a camera arranged on a vehicle;
identifying a delinquent region in the image;
identifying vehicles in the image that are in a parking violation area;
carrying out vehicle tracking on the vehicle in the image to obtain a motion track of wheels of the vehicle;
calculating a detection value of the motion trail, wherein the detection value is used for indicating the change amplitude of the motion trail;
when the detection value is less than a threshold value, the vehicle is identified as an illicit vehicle.
2. The method of claim 1, wherein the vehicle tracking the vehicle in the image, obtaining a motion trajectory of a wheel of the vehicle comprises:
selecting N frames of pictures containing the vehicle from the image;
detecting the N frames of pictures in a preset detection range based on an angular point detection algorithm to obtain a plurality of angular points of the wheel;
and calculating the optical flow of the angular points based on an optical flow algorithm to be used as the motion trail of the wheels of the vehicle.
3. The method of claim 2, wherein the image comprises a plurality of frames of pictures, and wherein the selecting the N frames of pictures from the image that include the vehicle comprises:
sequentially detecting the overlapping degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image through a vehicle detection frame;
detecting the correlation degree of the vehicles contained in the adjacent pictures of the front frame and the rear frame from the image by a correlation filtering algorithm in sequence;
and selecting the adjacent pictures of the two frames before and after the image with the overlapping degree and the correlation degree meeting the requirements as the N frames of pictures.
4. The method of claim 2, wherein after identifying the vehicle as a parking violation vehicle, the method further comprises:
generating illegal parking event information corresponding to the illegal parking vehicles according to the N frames of pictures;
and outputting the illegal event information.
5. The method of claim 1, wherein the calculating the detection value of the motion trajectory comprises:
calculating a track central point of the motion track;
and calculating the track variance of the motion track according to the track central point, and taking the track variance as the detection value.
6. The method of claim 1, wherein the identifying vehicles in the parking violation area in the image comprises:
and identifying the vehicle from the image based on a target detection network, wherein the target detection network is obtained by training vehicle information added with vehicle labels.
7. The method of claim 1, wherein the obtaining the image captured by the camera disposed on the vehicle comprises:
and receiving the image shot by the camera transmitted by the camera based on the network connection with the camera arranged on the vehicle.
8. An apparatus for identifying a parked vehicle, the apparatus comprising:
the device comprises an acquisition unit, a processing unit and a display unit, wherein the acquisition unit is used for acquiring an image shot by a camera arranged on a vehicle;
an identification unit configured to identify a violation area in the image;
the identification unit is further used for identifying the vehicles in the illegal parking areas in the images;
the tracking unit is used for carrying out vehicle tracking on the vehicle in the image to obtain the motion track of the wheels of the vehicle;
a calculation unit configured to calculate a detection value of the motion trajectory, the detection value indicating a variation amplitude of the motion trajectory;
the identification unit is further used for identifying the vehicle as an illegal vehicle when the detection value is smaller than a threshold value.
9. A processing device comprising a processor and a memory, the memory having a computer program stored therein, the processor executing the method of identifying an offending vehicle as claimed in any one of claims 1 to 7 when the processor calls the computer program in the memory.
10. A computer readable storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the method for identifying a parked vehicle according to any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911183772.2A CN112862856A (en) | 2019-11-27 | 2019-11-27 | Method, device and equipment for identifying illegal vehicle and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911183772.2A CN112862856A (en) | 2019-11-27 | 2019-11-27 | Method, device and equipment for identifying illegal vehicle and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112862856A true CN112862856A (en) | 2021-05-28 |
Family
ID=75984878
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911183772.2A Pending CN112862856A (en) | 2019-11-27 | 2019-11-27 | Method, device and equipment for identifying illegal vehicle and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112862856A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113506386A (en) * | 2021-06-30 | 2021-10-15 | 深圳市路卓科技有限公司 | Monitoring method for illegal parking, automobile data recorder, program and storage medium |
| CN114820692A (en) * | 2022-06-29 | 2022-07-29 | 珠海视熙科技有限公司 | State analysis method and device of tracking target, storage medium and terminal |
| CN115272954A (en) * | 2022-06-29 | 2022-11-01 | 珠海视熙科技有限公司 | Passenger flow statistics device and intelligent terminal |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108280847A (en) * | 2018-01-18 | 2018-07-13 | 维森软件技术（上海）有限公司 | A kind of vehicle movement track method of estimation |
| CN109766867A (en) * | 2019-01-22 | 2019-05-17 | 长沙智能驾驶研究院有限公司 | Travel condition of vehicle determines method, apparatus, computer equipment and storage medium |
| CN110378276A (en) * | 2019-07-16 | 2019-10-25 | 顺丰科技有限公司 | Vehicle-state acquisition methods, device, equipment and storage medium |
-
2019
- 2019-11-27 CN CN201911183772.2A patent/CN112862856A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108280847A (en) * | 2018-01-18 | 2018-07-13 | 维森软件技术（上海）有限公司 | A kind of vehicle movement track method of estimation |
| CN109766867A (en) * | 2019-01-22 | 2019-05-17 | 长沙智能驾驶研究院有限公司 | Travel condition of vehicle determines method, apparatus, computer equipment and storage medium |
| CN110378276A (en) * | 2019-07-16 | 2019-10-25 | 顺丰科技有限公司 | Vehicle-state acquisition methods, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| XIAOJING SONG ET AL.: "A Vision Based Wheel Slip Estimation Technique for Mining Vehicles", 《IFAC PROCEEDINGS VOLUMES》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113506386A (en) * | 2021-06-30 | 2021-10-15 | 深圳市路卓科技有限公司 | Monitoring method for illegal parking, automobile data recorder, program and storage medium |
| CN114820692A (en) * | 2022-06-29 | 2022-07-29 | 珠海视熙科技有限公司 | State analysis method and device of tracking target, storage medium and terminal |
| CN115272954A (en) * | 2022-06-29 | 2022-11-01 | 珠海视熙科技有限公司 | Passenger flow statistics device and intelligent terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10261574B2 (en) | Real-time detection system for parked vehicles | |
| Wu et al. | Lane-mark extraction for automobiles under complex conditions | |
| EP3806064B1 (en) | Method and apparatus for detecting parking space usage condition, electronic device, and storage medium | |
| WO2018187632A1 (en) | Deep learning methods for estimating density and/or flow of objects, and related methods and software | |
| US9760800B2 (en) | Method and system to detect objects using block based histogram of oriented gradients | |
| CN108877269B (en) | Intersection vehicle state detection and V2X broadcasting method | |
| CN112862856A (en) | Method, device and equipment for identifying illegal vehicle and computer readable storage medium | |
| US20180129883A1 (en) | Detection method and apparatus of a status of a parking lot and electronic equipment | |
| CN112101305B (en) | Multi-path image processing method and device and electronic equipment | |
| CN110929655B (en) | Lane line identification method in driving process, terminal device and storage medium | |
| CN110414514B (en) | Image processing method and device | |
| CN110781823B (en) | Screen recording detection method and device, readable medium and electronic equipment | |
| CN111582032A (en) | Pedestrian detection method and device, terminal equipment and storage medium | |
| CN109102026B (en) | Vehicle image detection method, device and system | |
| CN112991349B (en) | Image processing method, device, equipment and storage medium | |
| Khongprasongsiri et al. | A hardware implementation for real-time lane detection using high-level synthesis | |
| Ashraf et al. | HVD-net: a hybrid vehicle detection network for vision-based vehicle tracking and speed estimation | |
| CN114120259A (en) | Empty parking space identification method and system, computer equipment and storage medium | |
| CN109740502B (en) | Road quality detection method and device | |
| KR101888959B1 (en) | Method and system for recognizing vehicle plate in multi-lane | |
| Pun et al. | A real-time detector for parked vehicles based on hybrid background modeling | |
| CN112101139A (en) | Human shape detection method, device, equipment and storage medium | |
| CN113869123A (en) | Crowd-based event detection method and related device | |
| CN113147746A (en) | Method and device for detecting ramp parking space | |
| CN105809971A (en) | Vehicle tracking method and device |
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 |