CN110738857B - Vehicle violation evidence obtaining method, device and equipment - Google Patents
Vehicle violation evidence obtaining method, device and equipment Download PDFInfo
- Publication number
- CN110738857B CN110738857B CN201810800074.1A CN201810800074A CN110738857B CN 110738857 B CN110738857 B CN 110738857B CN 201810800074 A CN201810800074 A CN 201810800074A CN 110738857 B CN110738857 B CN 110738857B
- Authority
- CN
- China
- Prior art keywords
- vehicle target
- violation
- license plate
- vehicle
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the invention provides a vehicle violation evidence obtaining method, a device and equipment, wherein the method comprises the following steps: presetting a license plate acquisition region, and acquiring a detailed image containing a license plate of a vehicle after the vehicle enters the region; in addition, whether the vehicle has the violation behaviors or not is judged through the monitoring image, and if the violation behaviors exist, the monitoring image containing the violation behaviors and the detail image containing the license plate of the vehicle are output; therefore, the scheme realizes evidence obtaining of the vehicle with the violation behaviors.
Description
Technical Field
The invention relates to the technical field of intelligent traffic, in particular to a vehicle violation evidence obtaining method, device and equipment.
Background
At present, vehicles are more and more, traffic pressure is more and more, and various violation behaviors of the vehicles are more and more common, such as illegal lane changing, illegal turning, illegal parking, jamming, retrograde motion and the like. These vehicle violations seriously affect travel efficiency and pose a great threat to life and property safety.
Disclosure of Invention
The embodiment of the invention aims to provide a method, a device and equipment for obtaining evidence of vehicle violation, so as to obtain evidence of vehicle violation.
In order to achieve the above object, an embodiment of the present invention provides a vehicle violation forensics method, including:
acquiring a monitoring image, wherein the monitoring image comprises one or more vehicle targets;
judging whether the vehicle target has violation behaviors or not aiming at each vehicle target;
if the license plate exists, acquiring a detailed image of the license plate containing the violation vehicle target; wherein, the vehicle object of violating the regulations is the vehicle object of judging that there is the violation, and the detail image is: the method comprises the steps of acquiring an image after a vehicle target enters a preset area;
and outputting the acquired detail image and a monitoring image containing the violation behaviors.
Optionally, the determining whether the vehicle target has a violation, includes:
determining the attribute of a lane where the vehicle target is located by analyzing the monitoring image;
and judging whether the vehicle target is matched with the attribute of the lane where the vehicle target is located, if not, indicating that the vehicle target has violation behaviors.
Optionally, the determining whether the vehicle target has a violation, includes:
determining lane line parameters around the vehicle target by analyzing the monitoring image;
tracking the vehicle target in the monitoring image to obtain the track of the vehicle target;
and judging whether the vehicle target has violation behaviors or not based on the track of the vehicle target and the lane line parameters around the vehicle target.
Optionally, after the obtaining of the detail image of the license plate containing the violation vehicle target, the method further includes:
determining the license plate area of the violation vehicle target according to the size and the position of the tracking frame of the violation vehicle target in the acquired detail image;
identifying the license plate area to obtain the license plate number of the violation vehicle target;
and outputting the license plate number of the violation vehicle target and alarm information aiming at the violation vehicle target.
Optionally, the determining, in the obtained detail image, a license plate area of the violation vehicle target according to the size and the position of the tracking frame of the violation vehicle target includes:
determining a first vertex coordinate of a license plate area according to a first vertex coordinate of a tracking frame of the violation vehicle target and the height of the tracking frame in the acquired detail image;
determining the width of the tracking frame as the width of the license plate area, and determining the product of the height of the tracking frame and a preset value as the height of the license plate area, wherein the preset value is less than 1;
and determining the license plate area of the violation vehicle target in the acquired detail image according to the determined first vertex coordinate and the width and height of the license plate area.
Optionally, the obtaining of the detailed image of the license plate containing the violation vehicle target includes:
acquiring a plurality of detailed images of license plates containing violation vehicle targets;
the method further comprises the following steps:
carrying out license plate recognition on the acquired multiple detail images;
and if a plurality of license plate numbers are obtained through identification, determining the license plate number with the highest confidence coefficient as the license plate number of the violation vehicle target.
In order to achieve the above object, an embodiment of the present invention further provides a device for obtaining evidence of vehicle violation, including:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a monitoring image, and the monitoring image comprises one or more vehicle targets;
the judging module is used for judging whether the vehicle target has violation behaviors or not aiming at each vehicle target; if the current time slot exists, triggering a second acquisition module;
the second acquisition module is used for acquiring a detailed image of a license plate containing a violation vehicle target; wherein, the vehicle object of violating the regulations is the vehicle object of judging that there is the violation, and the detail image is: the method comprises the steps of acquiring an image after a vehicle target enters a preset area;
and the output module is used for outputting the acquired detail image and the monitoring image containing the violation behaviors.
Optionally, the determining module is specifically configured to:
determining the attribute of a lane where the vehicle target is located by analyzing the monitoring image;
and judging whether the vehicle target is matched with the attribute of the lane where the vehicle target is located, if not, indicating that the vehicle target has violation behaviors.
Optionally, the determining module is specifically configured to:
determining lane line parameters around the vehicle target by analyzing the monitoring image;
tracking the vehicle target in the monitoring image to obtain the track of the vehicle target;
and judging whether the vehicle target has violation behaviors or not based on the track of the vehicle target and the lane line parameters around the vehicle target.
Optionally, the apparatus further comprises:
the determining module is used for determining the license plate area of the violation vehicle target according to the size and the position of the tracking frame of the violation vehicle target in the acquired detail image;
the first identification module is used for identifying the license plate area to obtain the license plate number of the violation vehicle target;
the output module is also used for outputting the license plate number of the violation vehicle target and the alarm information aiming at the violation vehicle target.
Optionally, the determining module is specifically configured to:
determining a first vertex coordinate of a license plate area according to a first vertex coordinate of a tracking frame of the violation vehicle target and the height of the tracking frame in the acquired detail image;
determining the width of the tracking frame as the width of the license plate area, and determining the product of the height of the tracking frame and a preset value as the height of the license plate area, wherein the preset value is less than 1;
and determining the license plate area of the violation vehicle target in the acquired detail image according to the determined first vertex coordinate and the width and height of the license plate area.
Optionally, the second obtaining module is specifically configured to: acquiring a plurality of detailed images of license plates containing violation vehicle targets;
the device further comprises:
the second recognition module is used for recognizing the license plate of the acquired multiple detailed images; and if a plurality of license plate numbers are obtained through identification, determining the license plate number with the highest confidence coefficient as the license plate number of the violation vehicle target.
In order to achieve the above object, an embodiment of the present invention further provides an electronic device, including a processor and a memory;
a memory for storing a computer program;
and the processor is used for realizing any vehicle violation forensics method when executing the program stored in the memory.
In order to achieve the above object, an embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements any one of the above vehicle violation forensics methods.
In order to achieve the above object, an embodiment of the present invention further provides a chip, including a processor and a memory;
a memory for storing a computer program;
and the processor is used for realizing any vehicle violation forensics method when executing the program stored in the memory.
In the embodiment of the invention, a license plate acquisition area is preset, and after a vehicle enters the area, a detailed image containing the license plate of the vehicle is acquired; in addition, whether the vehicle has the violation behaviors or not is judged through the monitoring image, and if the violation behaviors exist, the monitoring image containing the violation behaviors and the detail image containing the license plate of the vehicle are output; therefore, the scheme realizes evidence obtaining of the vehicle with the violation behaviors.
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, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1a is a schematic diagram of a framework according to an embodiment of the present invention;
fig. 1b is a schematic view of an application scenario provided in the embodiment of the present invention;
FIG. 1c is a schematic diagram of target detection according to an embodiment of the present invention;
FIG. 1d is a schematic diagram of target tracking according to an embodiment of the present invention;
fig. 1e is a schematic view of analysis of a violation behavior according to an embodiment of the present invention;
fig. 1f is a schematic diagram of license plate recognition according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a vehicle violation forensics method according to an embodiment of the invention;
fig. 3 is a schematic structural diagram of a vehicle violation forensics device according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, 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 invention.
The terms of the present invention are explained as follows:
violation evidence obtaining: and carrying out snapshot and evidence collection on the violation behaviors of the vehicle targets in the road area.
Multi-frame identification: the license plate recognition method is mainly used for license plate recognition, and obtains the optimal license plate number by combining the license plate number recognized by the current frame and the license plate number recognized by the historical frame.
Deep Learning (Deep Learning): from the research of artificial neural networks, the multilayer perceptron with multiple hidden layers is a deep learning structure. Deep learning forms a more abstract class or feature of high-level representation properties by combining low-level features to discover a distributed feature representation of the data.
Cnn (volumetric Neural network): a convolutional neural network.
R-CNN (regions with CNN features): applying CNN to the field of target detection.
Yolo (young only look once): a fast real-time target detection method.
Ssd (single Shot MultiBox detector): a target detection method using feature fusion.
Camshift (continuous Adaptive Mean-SHIFT algorithm): the basic idea is to perform Meanshift operation on all frames of the video image, and to use the result of the previous frame as the initial value of Search Window of the Meanshift algorithm of the next frame, and so on.
The invention concept of the invention is as follows:
as shown in fig. 1a, an embodiment of the present invention may include five modules: the system comprises a parameter input module 101, a target detection module 102, a target tracking module 103, a violation analysis module 104 and a license plate recognition module 105.
Specifically, various road parameters may be input through the parameter input module 101. The road parameters may include: lane position, lane attributes, lane line parameters, license plate collection area, and the like. Lane attributes such as motorways, non-motorways, sidewalks, etc. Lane line parameters such as double yellow lines (for separating traffic lines running in opposite directions), white solid lines (for separating motor vehicles and non-motor vehicles running in the same direction), diversion lines (indicating that the vehicle must run along a specified route, and must not be pressed or run over the line), and the like, which are not listed. The license plate collection area can be as shown in fig. 1b, and in this area, the license plate of the vehicle can be collected more clearly.
According to the road parameters, a lane area, lane attributes, lane line parameters and a license plate collection area can be identified in the monitored image. Alternatively, whether parking of the vehicle is permitted for each zone, the traveling direction of each zone, whether lane changing of the vehicle is permitted for each zone, whether turning of the vehicle is permitted for each zone, and the like may be set according to the lane attribute and the lane line parameter, and these set contents may also belong to the road parameter.
The deep network model may be trained in advance in the target detection module 102. Specifically, road traffic image samples in different time periods, different weather and different scenes can be sorted, and the number of the image samples can be more than 20 ten thousand; and calibrating the targets such as motor vehicles, pedestrians, non-motor vehicles, roadblocks and the like in the image sample. As shown in fig. 1c, the deep network model is obtained by training using the image samples, the calibrated data, and network structures such as R-CNN, YOLO, or SSD. For example, based on the caffe environment, the image samples and the calibrated data are input into network structures such as R-CNN, YOLO, or SSD, and are iteratively adjusted 100 ten thousand times to achieve convergence, so as to obtain a deep network model.
After the deep network model is obtained, the collected monitoring image can be input into the deep network model, and a detection frame is output, wherein the detection frame is the detected vehicle target.
The target tracking module 103 tracks the vehicle target detected by the target detection module 102. As shown in fig. 1d, the target tracking module 103 may comprise a preprocessing unit, a matching unit, a tracking unit.
The preprocessing unit may filter the vehicle target detected by the target detection module 102, for example, filter a target with abnormal dimensions, and the specific filtering condition is not limited.
The matching unit can match the vehicle target in the current frame image with the vehicle target in the tracking queue, specifically, match the vehicle target, can calculate the coincidence degree of the detection frame of the vehicle target, if the coincidence degree is greater than a threshold value, it indicates that the matching is successful, that is, the vehicle target in the current frame image and the vehicle target in the tracking queue are the same vehicle target, and records the vehicle target information in the current frame image, such as the ID and the position of the vehicle target, that is, the position of the vehicle target is updated, thereby realizing the tracking of the vehicle target.
If the degree of coincidence is less than the threshold, a matching failure is indicated, in which case the target tracking unit may be triggered. For example, the target tracking unit may perform target tracking by using a camshift algorithm, specifically, may initialize the size and position of a search window, and adaptively adjust the position and size of the search window according to a detection frame obtained from a previous frame, thereby locating the position of the vehicle target of the current frame. The vehicle target information, such as the ID and the position of the vehicle target, is recorded, that is, the position of the vehicle target is updated, so that the tracking of the vehicle target is realized. If the target tracking unit fails to track the vehicle target, the vehicle target is represented as a newly appeared target, and the vehicle target can be newly built and the newly built vehicle target information can be recorded.
By the matching unit or the target tracking unit, the trajectory of the vehicle target, which corresponds to the ID, can be obtained.
The target tracking module 103 sends the obtained vehicle target ID and track to the violation analysis module 104, and in addition, the violation analysis module 104 also obtains the road parameters in the parameter input module 101; in this way, the violation analysis module 104 may determine whether the violation is present in the vehicle target based on the road parameters and the trajectory of the vehicle target.
By way of example, violations may include:
the illegal act: the behaviour of the vehicle is not parked as specified.
And (3) reverse behavior: a behavior in which the vehicle travels in a direction opposite to the prescribed direction.
Line pressing action: and pressing the behaviors of a white solid line, a double yellow line, a virtual solid line and a real dotted line during the running process of the vehicle.
Lane change behavior: the vehicle does not follow the guidance and acts of illegal lane change.
Turning around behavior: and forbidding the illegal turning behavior at the turning intersection.
Emergency lane occupation/non-occupation behavior: illegal emergency lane or non-motor lane occupation of the vehicle.
And (3) plugging detection: and (5) taking the other lanes to change lane and queue for overtaking.
Detecting a traffic accident: a traffic accident occurs in a designated area.
For example, if the road parameters set the position of the area where the vehicle cannot be parked, and the vehicle target is determined to be parked in the area according to the track of the vehicle target, the vehicle target has violation behaviors. If the driving direction of a certain area is set in the road parameters, and the track of the vehicle target is inconsistent with the driving direction, the vehicle target has violation behaviors. If the position of a certain lane line is set in the road parameters, and the track of the vehicle target is crossed with the position of the lane line, the vehicle target has violation behaviors. If the road parameters set the area which can not change the lane, and the lane change of the vehicle target in the area is determined according to the track of the vehicle target, the vehicle target has violation behaviors. If the intersection area which can not turn around is set in the road parameters, and the vehicle target turns around in the intersection area according to the track of the vehicle target, the vehicle target has violation behaviors. And the like, and various ways for judging whether the vehicle target has the violation behaviors are provided on the basis of the preset road parameters and the track of the vehicle target, which are not repeated one by one.
In addition, in the target detection module 102, the pedestrians, the roadblocks, and the like are calibrated, so that the trained deep network model can also identify the targets of the pedestrians, the roadblocks, and the like. Therefore, the violation analysis module 104 can also analyze whether the pedestrian has violation behaviors, and can also detect the conditions of roadblocks, construction and the like. For example, if a certain lane attribute is set as a motor lane in the road parameters, and the track of a certain pedestrian appears in the motor lane, it is determined that the pedestrian has violation behavior.
As shown in fig. 1e, on one hand, the violation behavior analysis is performed according to the ID and the track of the vehicle target and the preset road parameters, and on the other hand, the image acquisition is performed on the vehicle target entering the license plate acquisition area to obtain a detailed image including the license plate. For the purpose of distinguishing description, in the present embodiment, an image capable of clearly reflecting a license plate is referred to as a detail image, and an image having a large field of view and capable of reflecting a vehicle violation behavior is referred to as a monitoring image.
Referring to fig. 1b and 1e, after entering the monitoring area, the uplink vehicle passes through the license plate acquisition area first, and after the downlink vehicle passes through the monitoring area, the downlink vehicle passes through the license plate acquisition area finally. Therefore, when the vehicle target is judged to have the violation behaviors, the vehicle target may enter the license plate acquisition area or may not enter the license plate acquisition area. And if the vehicle target enters the license plate acquisition area, acquiring the detail image of the license plate containing the vehicle target, and outputting the detail image of the license plate containing the vehicle target, the monitoring image containing the violation behaviors and violation alarm information. If the vehicle target does not enter the license plate acquisition area, firstly caching the monitoring image containing the violation behaviors and the violation alarm information, waiting for the vehicle target to enter the license plate acquisition area, acquiring the detail image containing the license plate of the vehicle target, and then outputting the detail image containing the license plate of the vehicle target, the cached monitoring image containing the violation behaviors and the violation alarm information.
The license plate recognition module 105 can recognize the detail image of the license plate containing the vehicle target to obtain the license plate number, and can further output the license plate number of the violation vehicle on the basis of outputting the contents. Specifically, as shown in fig. 1f, the license plate number is obtained through license plate positioning, Character segmentation, and Character Recognition (OCR).
For example, for the same vehicle target, if multiple detailed images of the vehicle target are collected in a license plate collection area, the multiple detailed images can be respectively identified, and if multiple license plates are obtained, the license plate number with the highest confidence coefficient is determined as the license plate number of the vehicle target.
Therefore, on the first aspect, the scheme provides a scheme for obtaining evidence of the vehicle violating the regulations, and can detect various violations and obtain the evidence; in the scheme, a vehicle target is tracked, the license plate is collected in a preset area, the vehicle target is large, and the tracking distance is long; in the scheme, a deep network model is adopted for target detection, and a scheme of multi-frame recognition is adopted for license plate recognition, so that the accuracy is high.
Based on the same inventive concept, the embodiment of the invention provides a vehicle violation evidence obtaining method, a vehicle violation evidence obtaining device and vehicle violation evidence obtaining equipment. First, a vehicle violation forensics method provided by the embodiment of the invention is explained in detail below.
Fig. 2 is a schematic flow chart of a vehicle violation forensics method provided by an embodiment of the present invention, including:
s201: and acquiring a monitoring image, wherein the monitoring image comprises one or more vehicle targets.
As described above, the monitor image in the present embodiment may be an image having a large field of view. Referring to fig. 1b, when a vehicle object enters the monitoring range of the monitoring device (the whole image in fig. 1b belongs to the monitoring range), the vehicle object is included in the monitoring image collected by the monitoring device. One or more vehicle targets may be included in the monitored image.
S202: and judging whether the vehicle target has violation behaviors or not aiming at each vehicle target. If so, S203-S204 are performed.
As described above, road parameters, such as lane position, lane attribute, lane line parameter, license plate collection area, and the like, may be set in advance. The lane position can be understood as a position parameter of a lane area; lane attributes such as motorways, non-motorways, sidewalks, etc.; lane line parameters such as double yellow lines (for separating traffic lines running in opposite directions), white solid lines (for separating motor vehicles and non-motor vehicles running in the same direction), diversion lines (indicating that the vehicle must run along a specified route, and must not be pressed or run over the line), and the like, which are not listed. The license plate collection area can be as shown in fig. 1b, and in this area, the license plate of the vehicle can be collected more clearly.
According to the road parameters, a lane area, lane attributes, lane line parameters and a license plate collection area can be identified in the monitored image. Alternatively, whether parking of the vehicle is permitted for each zone, the traveling direction of each zone, whether lane changing of the vehicle is permitted for each zone, whether turning of the vehicle is permitted for each zone, and the like may be set according to the lane attribute and the lane line parameter, and these set contents may also belong to the road parameter.
By way of example, violations may include:
the illegal act: the behaviour of the vehicle is not parked as specified.
And (3) reverse behavior: a behavior in which the vehicle travels in a direction opposite to the prescribed direction.
Line pressing action: and pressing the behaviors of a white solid line, a double yellow line, a virtual solid line and a real dotted line during the running process of the vehicle.
Lane change behavior: the vehicle does not follow the guidance and acts of illegal lane change.
Turning around behavior: and forbidding the illegal turning behavior at the turning intersection.
Emergency lane occupation/non-occupation behavior: illegal emergency lane or non-motor lane occupation of the vehicle.
And (3) plugging detection: and (5) taking the other lanes to change lane and queue for overtaking.
Detecting a traffic accident: a traffic accident occurs in a designated area.
In one embodiment, determining whether the vehicle target has violation may include: determining the attribute of a lane where the vehicle target is located by analyzing the monitoring image; and judging whether the vehicle target is matched with the attribute of the lane where the vehicle target is located, if not, indicating that the vehicle target has violation behaviors.
It will be appreciated that if a vehicle target is present on a non-motorized lane, or on a sidewalk, it is indicative of a violation of the vehicle target. In the embodiment, the lane attribute of the vehicle target is determined; if the determined lane attribute is a motor lane, namely the vehicle target is matched with the lane attribute where the vehicle target is located, the vehicle target does not have violation behaviors; if the determined lane attribute is a non-motor lane or a sidewalk, namely the vehicle target is not matched with the lane attribute where the vehicle target is located, violation behaviors exist in the vehicle target.
Or in other scenes, the lane attributes may further include a passenger car lane, a cargo lane, and the like. For example, if the attribute of the lane where the vehicle target is located is determined to be a passenger car lane, and the vehicle target is analyzed to be a large truck, it indicates that the vehicle target is not matched with the attribute of the lane where the vehicle target is located, and the vehicle target has violation behaviors.
In another embodiment, the determining whether the vehicle target has violation behavior may include: determining lane line parameters around the vehicle target by analyzing the monitoring image; tracking the vehicle target in the monitoring image to obtain the track of the vehicle target; and judging whether the vehicle target has violation behaviors or not based on the track of the vehicle target and the lane line parameters around the vehicle target.
Specifically, the vehicle target can be tracked in the monitored image through a target tracking algorithm to obtain the track of the vehicle target. And then judging whether the vehicle target has violation behaviors or not based on the track and the lane line parameters.
For example, if the lane line parameter around the vehicle object indicates that the vehicle cannot be parked in the area, and it is determined from the trajectory of the vehicle object that the vehicle object is parked in the area, the vehicle object has illegal parking behavior. If the driving direction indicated by the lane line parameters around the vehicle object does not coincide with the trajectory of the vehicle object, the vehicle object exhibits retrograde behavior. If the lane line parameters around the vehicle target indicate that the vehicle cannot line, and the track of the vehicle target is crossed with the position of the lane line, the vehicle target has line pressing behavior. And if the lane line parameters around the vehicle target indicate that lane changing cannot be performed in the area, and the lane changing of the vehicle target in the area is determined according to the track of the vehicle target, the vehicle target has illegal lane changing behaviors. If the lane line parameters around the vehicle target indicate that the intersection area can not turn around, and the vehicle target is determined to turn around in the intersection area according to the track of the vehicle target, the vehicle target has illegal turning behavior.
In addition, as described above, in one case, it is also possible to set whether or not parking of the vehicle is permitted for each zone, the traveling direction of each zone, whether or not lane change of the vehicle is permitted for each zone, whether or not the vehicle is permitted for each zone to turn around, and the like, according to the lane attribute, the lane line parameter, and the like, and these set contents may also belong to the road parameter.
In this case, if the location of the area where the vehicle cannot be parked is set in the road parameter and it is determined that the vehicle target is parked in the area based on the trajectory of the vehicle target, there is an illegal parking behavior of the vehicle target. If the driving direction of a certain area is set in the road parameters, and the track of the vehicle target is inconsistent with the driving direction, the vehicle target has reverse behavior. If the position of a certain lane line is set in the road parameters, and the track of the vehicle target is crossed with the position of the lane line, the vehicle target has line pressing behavior. If the road parameter sets up the area that can not change the lane, and confirm the vehicle goal has changed the lane in this area according to the orbit of the vehicle goal, then the illegal lane change behavior of this vehicle goal exists. If the intersection area which can not turn around is set in the road parameters, and the vehicle target turns around in the intersection area according to the track of the vehicle target, the vehicle target has illegal turning around behaviors. And the like, and various ways for judging whether the vehicle target has the violation behaviors are provided on the basis of the road parameters and the track of the vehicle target, and are not repeated one by one.
S203: and acquiring a detailed image of the license plate containing the violation vehicle target.
Wherein, the vehicle object of violating the regulations is the vehicle object of judging that there is the violation, and the detail image is: the images are collected when the vehicle target enters a preset area.
The preset area may also be referred to as a license plate collection area. Referring to fig. 1b, the preset area may be an area close to the monitoring device, and after the vehicle target enters the preset area, the monitoring device may clearly acquire the license plate of the vehicle target.
In one implementation mode, a plurality of monitoring devices can be arranged in a scene, some monitoring devices are used for collecting monitoring videos, each frame of image in the monitoring videos is the monitoring image, other monitoring devices can be a snapshot machine, and after a vehicle target enters a preset area, the snapshot machine takes a snapshot of the vehicle target to obtain a detailed image containing a license plate.
Or, in another embodiment, as shown in fig. 1b, only one monitoring device may be provided, and the monitoring device may collect a monitoring video and capture a vehicle target entering a preset area to obtain a detailed image. Or, the monitoring equipment only collects monitoring videos and does not additionally capture images. The electronic device (executing subject) obtains a detailed image of a license plate containing a vehicle object in the monitoring video.
In one case, for each vehicle target entering a preset area, acquiring a detail image containing the vehicle target; thus, it is desirable to determine a detail image that contains the violation vehicle target in the captured detail image.
For example, the violation vehicle target can be tracked in the monitoring image, the time for acquiring the detail image of the violation vehicle target is determined, and the detail image containing the violation vehicle target is determined according to the determined time. Or the violation vehicle target in the monitoring image and the vehicle target in the detail image can be compared in similarity, and the detail image containing the violation vehicle target is determined according to the comparison result.
And after S203, the detailed image containing the license plate of the violation vehicle target can be identified to obtain the license plate number of the violation vehicle target. As an implementation mode, the license plate area of the violation vehicle target can be determined according to the size and the position of the tracking frame of the violation vehicle target in the acquired detail image; and identifying the license plate area to obtain the license plate number of the violation vehicle target.
In the embodiment, the license plate area is determined firstly, only the license plate area is identified when the license plate is identified, the whole image is not required to be identified, and the identification efficiency is improved.
For example, determining the license plate region may include:
determining a first vertex coordinate of a license plate area according to a first vertex coordinate of a tracking frame of the violation vehicle target and the height of the tracking frame in the acquired detail image; determining the width of the tracking frame as the width of the license plate area, and determining the product of the height of the tracking frame and a preset value as the height of the license plate area, wherein the preset value is less than 1; and determining the license plate area of the violation vehicle target in the acquired detail image according to the determined first vertex coordinate and the width and height of the license plate area.
Specifically, the following equation may be used to adjust the tracking frame of the vehicle target:
pr_rect.x＝obj_rect.x，
pr_rect.y＝obj_rect.y+0.5*obj_rect.h，
pr_rect.w＝obj_rect.w，
pr_rect.h＝0.6*obj_rect.h。
wherein obj _ rect represents a tracking frame, pr _ rect represents a license plate region, (obj _ rect. x, obj _ rect. y) represents a coordinate value of the upper left corner of the tracking frame, namely the first vertex coordinate of the tracking frame in a detail image, obj _ rect. h represents a height value of the tracking frame, obj _ rect. w represents a width value of the tracking frame, (pr _ rect. x, pr _ rect. y) represents a coordinate value of the upper left corner of the license plate region, namely the first vertex coordinate of the license plate region in the detail image, pr _ rect. h represents a height value of the license plate region, and pr _ rect. w represents a width value of the license plate region; and the coordinate values of the tracking frame and the license plate region use the upper left corner of the tracking frame as an origin. The preset value in the above equation is 0.6, or may be other values less than 1, and is not limited specifically.
As shown in fig. 1f, license plate identification can be performed by license plate positioning, character segmentation, and character identification to obtain a license plate number. Specifically, image features such as a hopping frequency feature, a color feature, and an HOG (Histogram of Oriented gradients) feature may be extracted from the license plate region, and the license plate may be located according to the features. And then searching the left and right boundaries of each character in the positioned area, and segmenting each character in the positioned area. Then, the size and brightness of the divided characters are normalized, the characters after the normalization processing are sent to a classifier, the classifier outputs the recognition result of each character, and the recognition results are combined into a character string, wherein the character string is the license plate number.
Under one condition, a scheme of multi-frame identification can be adopted, for example, a plurality of detailed images of license plates containing the violation vehicle target can be acquired aiming at each violation vehicle target, and license plate identification is carried out on the plurality of acquired detailed images; and if a plurality of license plate numbers are obtained through identification, determining the license plate number with the highest confidence coefficient as the license plate number of the violation vehicle target. Therefore, the license plate recognition accuracy can be improved.
S204: and outputting the acquired detail image and a monitoring image containing the violation behavior.
In one case, the monitoring images containing the violation behaviors can comprise three monitoring images before violation, during violation and after violation, and the whole vehicle violation process can be intuitively reflected by outputting the three monitoring images.
In addition, a detailed image containing the license plate of the vehicle target is output, so that evidence collection is more convincing.
Referring to fig. 1b, after entering the monitoring area, the uplink vehicle passes through the license plate acquisition area first, and after the downlink vehicle passes through the monitoring area, the downlink vehicle passes through the license plate acquisition area finally. Therefore, when the vehicle target is judged to have the violation behaviors, the vehicle target may enter the license plate acquisition area or may not enter the license plate acquisition area. And if the vehicle target enters the license plate acquisition area, acquiring the detail image of the license plate containing the vehicle target, and outputting the detail image of the license plate containing the vehicle target and the monitoring image containing the violation behaviors. If the vehicle target does not enter the license plate acquisition area, firstly caching the monitoring image containing the violation behaviors, waiting for the vehicle target to enter the license plate acquisition area, acquiring the detail image containing the license plate of the vehicle target, and then outputting the detail image containing the license plate of the vehicle target and the cached monitoring image containing the violation behaviors.
However, in some cases, if the violation vehicle target does not enter the preset area, the detailed image of the license plate containing the violation vehicle target is not acquired, and only the monitoring image containing the violation behavior can be output.
In one embodiment, the license plate number of the violation vehicle target can be output after the license plate number of the violation vehicle target is obtained.
In one embodiment, if the determination result at S202 is yes, violation alarm information of the vehicle target may be output. For example, the violation alarm information may be only information prompting the occurrence of a violation, such as "monitoring device a finds a violation"; or the violation alarm information can also comprise the type information of the violation behaviors, such as 'the monitoring device A finds the illegal lane change'; and the specific content of the violation alarm information is not limited.
By applying the embodiment shown in fig. 2 of the invention, a license plate acquisition area is preset, and after a vehicle enters the area, a detailed image containing the license plate of the vehicle is acquired; in addition, whether the vehicle has the violation behaviors or not is judged through the monitoring image, and if the violation behaviors exist, the monitoring image containing the violation behaviors and the detail image containing the license plate of the vehicle are output; therefore, the scheme realizes evidence obtaining of the vehicle with the violation behaviors.
In the embodiment of the invention, based on a deep learning theory, the video processing technology is utilized, and the track of the vehicle target and the configured road parameters are combined, so that the vehicle violation evidence obtaining method and the device are provided, and the innovation points are as follows:
(1) by combining a multi-target tracking technology and a multi-frame license plate recognition scheme, license plate recognition and evidence collection in a global monitoring range (large monitoring range and long tracking distance) can be realized, and remote/near distance violation behaviors are monitored; the accuracy is further improved while the license plate recognition rate is greatly improved, and the method has extremely important significance for road traffic supervision departments;
(2) the method has the advantages that the high-performance intelligent chip is used as a platform, the deep learning technology is used as a basis, the real-time detection of road targets such as motor vehicles/non-motor vehicles/pedestrians/roadblocks and the like is realized, the target detection rate is obviously improved, and the capture rate and the efficiency of events are high; the method has strong portability and expandability, and has strong reference significance for development and expansion of other functions;
(3) the scheme integrates various violation behaviors, can simultaneously support real-time detection and evidence collection of more than ten violation behaviors such as illegal lane changing, illegal turning, illegal parking, illegal line pressing, emergency lane occupation and retrograde driving, and has simple and intelligent product form and convenient subsequent maintenance and development;
(4) the scheme can simultaneously consider the scenes of crossroads, urban roads, high-speed roads, tunnels and the like, and has wide application scenes and strong adaptability.
Corresponding to the above method embodiment, an embodiment of the present invention further provides a vehicle violation forensics device, as shown in fig. 3, including:
a first obtaining module 301, configured to obtain a monitoring image, where the monitoring image includes one or more vehicle targets;
the judging module 302 is used for judging whether the vehicle target has violation behaviors or not aiming at each vehicle target; if the current time slot exists, triggering a second acquisition module;
the second obtaining module 303 is configured to obtain a detailed image of a license plate including a violation vehicle target; wherein, the vehicle object of violating the regulations is the vehicle object of judging that there is the violation, and the detail image is: the method comprises the steps of acquiring an image after a vehicle target enters a preset area;
and the output module 304 is used for outputting the acquired detail image and the monitoring image containing the violation behaviors.
As an implementation manner, the determining module 302 may specifically be configured to:
determining the attribute of a lane where the vehicle target is located by analyzing the monitoring image;
and judging whether the vehicle target is matched with the attribute of the lane where the vehicle target is located, if not, indicating that the vehicle target has violation behaviors.
As an implementation manner, the determining module 302 may specifically be configured to:
determining lane line parameters around the vehicle target by analyzing the monitoring image;
tracking the vehicle target in the monitoring image to obtain the track of the vehicle target;
and judging whether the vehicle target has violation behaviors or not based on the track of the vehicle target and the lane line parameters around the vehicle target.
As an embodiment, the apparatus may further include: a determination module and a first identification module (not shown in the figures), wherein,
the determining module is used for determining the license plate area of the violation vehicle target according to the size and the position of the tracking frame of the violation vehicle target in the acquired detail image;
the first identification module is used for identifying the license plate area to obtain the license plate number of the violation vehicle target;
and the output module 304 is further used for outputting the license plate number of the violation vehicle target and alarm information aiming at the violation vehicle target.
As an embodiment, the determining module may be specifically configured to:
determining a first vertex coordinate of a license plate area according to a first vertex coordinate of a tracking frame of the violation vehicle target and the height of the tracking frame in the acquired detail image; determining the width of the tracking frame as the width of the license plate area, and determining the product of the height of the tracking frame and a preset value as the height of the license plate area, wherein the preset value is less than 1; and determining the license plate area of the violation vehicle target in the acquired detail image according to the determined first vertex coordinate and the width and height of the license plate area.
As an embodiment, the second obtaining module 303 may be specifically configured to: acquiring a plurality of detailed images of license plates containing violation vehicle targets; the device further comprises:
the second recognition module (not shown in the figure) is used for carrying out license plate recognition on the acquired multiple detailed images; and if a plurality of license plate numbers are obtained through identification, determining the license plate number with the highest confidence coefficient as the license plate number of the violation vehicle target.
By applying the embodiment shown in fig. 3 of the invention, a license plate acquisition area is preset, and after a vehicle enters the area, a detailed image containing the license plate of the vehicle is acquired; in addition, whether the vehicle has the violation behaviors or not is judged through the monitoring image, and if the violation behaviors exist, the monitoring image containing the violation behaviors and the detail image containing the license plate of the vehicle are output; therefore, the scheme realizes evidence obtaining of the vehicle with the violation behaviors.
The determination module 302 in the embodiment of fig. 3 may be understood as the violation analysis module 104 in fig. 1a, and the first recognition module and/or the second recognition module may be understood as the license plate recognition module 105 in fig. 1 a.
An embodiment of the present invention further provides an electronic device, as shown in fig. 4, including a processor 401 and a memory 402,
a memory 402 for storing a computer program;
the processor 401, when executing the program stored in the memory 402, implements any of the vehicle violation forensics methods described above.
The Memory mentioned in the above electronic device may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
The embodiment of the invention also provides a chip, which comprises a processor and a memory; a memory for storing a computer program; and the processor is used for realizing any vehicle violation forensics method when executing the program stored in the memory.
The embodiment of the invention also provides a computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and when the computer program is executed by a processor, any one of the vehicle violation forensics methods is provided.
The embodiment of the invention also provides a chip, which comprises a processor and a memory;
a memory for storing a computer program;
and the processor is used for realizing any vehicle violation forensics method when executing the program stored in the memory.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the embodiment of the vehicle violation forensics device shown in fig. 3, the embodiment of the electronic equipment shown in fig. 4, the embodiment of the computer-readable storage medium and the embodiment of the chip, since the embodiments are basically similar to the embodiment of the vehicle violation forensics method shown in fig. 1 a-2, the description is relatively simple, and relevant points can be referred to as part of the description of the embodiment of the vehicle violation forensics method shown in fig. 1 a-2.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (15)
1. A vehicle violation forensics method, comprising:
acquiring a monitoring image, wherein the monitoring image comprises one or more vehicle targets;
judging whether the vehicle target has violation behaviors or not aiming at each vehicle target;
if the license plate exists, acquiring a detailed image of the license plate containing the violation vehicle target; wherein, the vehicle object of violating the regulations is the vehicle object of judging that there is the violation, and the detail image is: when a vehicle target enters a preset area, acquiring an image containing a license plate of the vehicle target; outputting the obtained detail image and a monitoring image containing the violation behaviors;
the step of obtaining the detail image of the license plate containing the violation vehicle target comprises the following steps:
comparing the similarity of the violation vehicle target with the vehicle target in the detail image, and acquiring the detail image of the license plate containing the violation vehicle target according to the comparison result;
the monitoring image is an image which has a large view field and can reflect the violation behaviors of the vehicle.
2. The method of claim 1 wherein said determining whether the vehicle target is violating includes:
determining the attribute of a lane where the vehicle target is located by analyzing the monitoring image;
and judging whether the vehicle target is matched with the attribute of the lane where the vehicle target is located, if not, indicating that the vehicle target has violation behaviors.
3. The method of claim 1 wherein said determining whether the vehicle target is violating includes:
determining lane line parameters around the vehicle target by analyzing the monitoring image;
tracking the vehicle target in the monitoring image to obtain the track of the vehicle target;
and judging whether the vehicle target has violation behaviors or not based on the track of the vehicle target and the lane line parameters around the vehicle target.
4. The method of claim 1, after said obtaining a detailed image of a license plate containing the violation vehicle target, further comprising:
determining the license plate area of the violation vehicle target according to the size and the position of the tracking frame of the violation vehicle target in the acquired detail image;
identifying the license plate area to obtain the license plate number of the violation vehicle target;
and outputting the license plate number of the violation vehicle target and alarm information aiming at the violation vehicle target.
5. The method of claim 4 wherein determining the license plate area of the violation vehicle target based on the size and location of the tracking frame of the violation vehicle target in the acquired detail image comprises:
determining a first vertex coordinate of a license plate area according to a first vertex coordinate of a tracking frame of the violation vehicle target and the height of the tracking frame in the acquired detail image;
determining the width of the tracking frame as the width of the license plate area, and determining the product of the height of the tracking frame and a preset value as the height of the license plate area, wherein the preset value is less than 1;
and determining the license plate area of the violation vehicle target in the acquired detail image according to the determined first vertex coordinate and the width and height of the license plate area.
6. The method of claim 1, wherein said obtaining a detailed image of a license plate containing a violation vehicle target comprises:
acquiring a plurality of detailed images of license plates containing violation vehicle targets;
the method further comprises the following steps:
carrying out license plate recognition on the acquired multiple detail images;
and if a plurality of license plate numbers are obtained through identification, determining the license plate number with the highest confidence coefficient as the license plate number of the violation vehicle target.
7. A vehicle violation forensics device, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a monitoring image, and the monitoring image comprises one or more vehicle targets;
the judging module is used for judging whether the vehicle target has violation behaviors or not aiming at each vehicle target; if the current time slot exists, triggering a second acquisition module;
the second acquisition module is used for acquiring a detailed image of a license plate containing a violation vehicle target; wherein, the vehicle object of violating the regulations is the vehicle object of judging that there is the violation, and the detail image is: when a vehicle target enters a preset area, acquiring an image containing a license plate of the vehicle target;
the output module is used for outputting the acquired detail image and the monitoring image containing the violation behaviors;
the second obtaining module is specifically configured to: comparing the similarity of the violation vehicle target with the vehicle target in the detail image, and acquiring the detail image of the license plate containing the violation vehicle target according to the comparison result;
the monitoring image is an image which has a large view field and can reflect the violation behaviors of the vehicle.
8. The apparatus of claim 7, wherein the determining module is specifically configured to:
determining the attribute of a lane where the vehicle target is located by analyzing the monitoring image;
and judging whether the vehicle target is matched with the attribute of the lane where the vehicle target is located, if not, indicating that the vehicle target has violation behaviors.
9. The apparatus of claim 7, wherein the determining module is specifically configured to:
determining lane line parameters around the vehicle target by analyzing the monitoring image;
tracking the vehicle target in the monitoring image to obtain the track of the vehicle target;
and judging whether the vehicle target has violation behaviors or not based on the track of the vehicle target and the lane line parameters around the vehicle target.
10. The apparatus of claim 7, further comprising:
the determining module is used for determining the license plate area of the violation vehicle target according to the size and the position of the tracking frame of the violation vehicle target in the acquired detail image;
the first identification module is used for identifying the license plate area to obtain the license plate number of the violation vehicle target;
the output module is also used for outputting the license plate number of the violation vehicle target and the alarm information aiming at the violation vehicle target.
11. The apparatus of claim 10, wherein the determining module is specifically configured to:
determining a first vertex coordinate of a license plate area according to a first vertex coordinate of a tracking frame of the violation vehicle target and the height of the tracking frame in the acquired detail image;
determining the width of the tracking frame as the width of the license plate area, and determining the product of the height of the tracking frame and a preset value as the height of the license plate area, wherein the preset value is less than 1;
and determining the license plate area of the violation vehicle target in the acquired detail image according to the determined first vertex coordinate and the width and height of the license plate area.
12. The apparatus of claim 7, wherein the second obtaining module is specifically configured to: acquiring a plurality of detailed images of license plates containing violation vehicle targets;
the device further comprises:
the second recognition module is used for recognizing the license plate of the acquired multiple detailed images; and if a plurality of license plate numbers are obtained through identification, determining the license plate number with the highest confidence coefficient as the license plate number of the violation vehicle target.
13. An electronic device comprising a processor and a memory;
a memory for storing a computer program;
a processor for implementing the method steps of any of claims 1-6 when executing a program stored in the memory.
14. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of claims 1 to 6.
15. A chip comprising a processor and a memory;
a memory for storing a computer program;
a processor for implementing the method steps of any of claims 1-6 when executing a program stored in the memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810800074.1A CN110738857B (en) | 2018-07-18 | 2018-07-18 | Vehicle violation evidence obtaining method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810800074.1A CN110738857B (en) | 2018-07-18 | 2018-07-18 | Vehicle violation evidence obtaining method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110738857A CN110738857A (en) | 2020-01-31 |
| CN110738857B true CN110738857B (en) | 2022-01-21 |
Family
ID=69234069
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810800074.1A Active CN110738857B (en) | 2018-07-18 | 2018-07-18 | Vehicle violation evidence obtaining method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110738857B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111680633A (en) * | 2020-06-10 | 2020-09-18 | 上海眼控科技股份有限公司 | Vehicle violation identification method and device, computer equipment and storage medium |
| CN111815959B (en) * | 2020-06-19 | 2021-11-16 | 浙江大华技术股份有限公司 | Vehicle violation detection method and device and computer readable storage medium |
| CN112071058A (en) * | 2020-08-14 | 2020-12-11 | 深延科技(北京)有限公司 | Road traffic monitoring and vehicle abnormity, contraband and fire detection method and system based on deep learning |
| CN112102615B (en) * | 2020-08-28 | 2022-03-25 | 浙江大华技术股份有限公司 | Traffic accident detection method, electronic device, and storage medium |
| CN112446902A (en) * | 2020-11-24 | 2021-03-05 | 浙江大华技术股份有限公司 | Method and device for determining abnormality of target vehicle, storage medium, and electronic device |
| CN112597945A (en) * | 2020-12-29 | 2021-04-02 | 上海眼控科技股份有限公司 | Vehicle detection method and device |
| CN112947446A (en) * | 2021-02-07 | 2021-06-11 | 启迪云控(上海)汽车科技有限公司 | Intelligent networking application scene automatic identification method, device, medium and equipment based on fully-known visual angle and feature extraction |
| CN113269042B (en) * | 2021-04-25 | 2024-03-29 | 安徽银徽科技有限公司 | Intelligent traffic management method and system based on driving vehicle violation identification |
| CN113380047A (en) * | 2021-06-22 | 2021-09-10 | 重庆盛海科技发展有限公司 | Illegal parking detection method based on traditional camera |
| CN113487878A (en) * | 2021-07-12 | 2021-10-08 | 重庆长安新能源汽车科技有限公司 | Motor vehicle illegal line pressing running detection method and system |
| CN114415708A (en) * | 2022-01-24 | 2022-04-29 | 上海复亚智能科技有限公司 | Road self-inspection method and device, unmanned aerial vehicle and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101964145A (en) * | 2009-07-23 | 2011-02-02 | 北京中星微电子有限公司 | Automatic license plate recognition method and system |
| CN102081844A (en) * | 2011-01-25 | 2011-06-01 | 华中科技大学 | Traffic video behavior analyzing and alarming server |
| CN105913000A (en) * | 2016-04-05 | 2016-08-31 | 浙江工业大学 | Vehicle tracking method based on video analysis |
| CN106991820A (en) * | 2016-01-20 | 2017-07-28 | 中兴通讯股份有限公司 | Violation vehicle processing method and processing device |
| CN108091142A (en) * | 2017-12-12 | 2018-05-29 | 公安部交通管理科学研究所 | For vehicle illegal activities Tracking Recognition under highway large scene and the method captured automatically |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101436348B (en) * | 2008-12-22 | 2014-04-23 | 北京中星微电子有限公司 | Electronic policeman system and information identification method applied to the same |
| US20120148092A1 (en) * | 2010-12-09 | 2012-06-14 | Gorilla Technology Inc. | Automatic traffic violation detection system and method of the same |
| CN105336170B (en) * | 2014-08-08 | 2017-09-19 | 杭州海康威视系统技术有限公司 | Traffic violation supervising method and its device |
| JP6459808B2 (en) * | 2015-07-08 | 2019-01-30 | オムロン株式会社 | Image processing apparatus and traffic violation management system provided with the same |
| CN105654734A (en) * | 2016-03-15 | 2016-06-08 | 深圳市佳信捷技术股份有限公司 | Data processing method and device for vehicle violations |
| CN106600977B (en) * | 2017-02-13 | 2020-09-15 | 深圳英飞拓科技股份有限公司 | Multi-feature recognition-based illegal parking detection method and system |
-
2018
- 2018-07-18 CN CN201810800074.1A patent/CN110738857B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101964145A (en) * | 2009-07-23 | 2011-02-02 | 北京中星微电子有限公司 | Automatic license plate recognition method and system |
| CN102081844A (en) * | 2011-01-25 | 2011-06-01 | 华中科技大学 | Traffic video behavior analyzing and alarming server |
| CN106991820A (en) * | 2016-01-20 | 2017-07-28 | 中兴通讯股份有限公司 | Violation vehicle processing method and processing device |
| CN105913000A (en) * | 2016-04-05 | 2016-08-31 | 浙江工业大学 | Vehicle tracking method based on video analysis |
| CN108091142A (en) * | 2017-12-12 | 2018-05-29 | 公安部交通管理科学研究所 | For vehicle illegal activities Tracking Recognition under highway large scene and the method captured automatically |
Non-Patent Citations (1)
| Title |
|---|
| 基于车辆检测与跟踪的交通事件检测方法研究;高炎;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20180415(第4期);第26-29页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110738857A (en) | 2020-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110738857B (en) | Vehicle violation evidence obtaining method, device and equipment | |
| US9679203B2 (en) | Traffic violation detection | |
| US6442474B1 (en) | Vision-based method and apparatus for monitoring vehicular traffic events | |
| KR20200071799A (en) | object recognition and counting method using deep learning artificial intelligence technology | |
| KR102122850B1 (en) | Solution for analysis road and recognition vehicle license plate employing deep-learning | |
| KR102282800B1 (en) | Method for trackig multi target employing ridar and camera | |
| CN107705577A (en) | A kind of real-time detection method and system based on lane line demarcation vehicle peccancy lane change | |
| Kejriwal et al. | Vehicle detection and counting using deep learning basedYOLO and deep SORT algorithm for urban traffic management system | |
| Kadim et al. | Real-time deep-learning based traffic volume count for high-traffic urban arterial roads | |
| Dinh et al. | Development of a tracking-based system for automated traffic data collection for roundabouts | |
| Al Nasim et al. | An automated approach for the recognition of bengali license plates | |
| Ng et al. | Outdoor illegal parking detection system using convolutional neural network on Raspberry Pi | |
| CN116311166A (en) | Traffic obstacle recognition method and device and electronic equipment | |
| Bravi et al. | Detection of stop sign violations from dashcam data | |
| Ramadan et al. | Real-time Automated License Plate Recognition and Tracking of Runaway Vehicles | |
| CN111627224A (en) | Vehicle speed abnormality detection method, device, equipment and storage medium | |
| Neto et al. | Computer-vision-based surveillance of intelligent transportation systems | |
| CN112686136B (en) | Object detection method, device and system | |
| KR20220089217A (en) | Two-wheeled vehicle control apparatus and method | |
| CN114693722A (en) | Vehicle driving behavior detection method, detection device and detection equipment | |
| Mokha et al. | A review of computer vision system for the vehicle identification and classification from online and offline videos | |
| CN113850112A (en) | Road condition identification method and system based on twin neural network | |
| CN105205500A (en) | Vehicle detecting method based on multi-target tracking and cascade classifier combination | |
| Manguri et al. | A Review of Computer Vision–Based Traffic Controlling and Monitoring | |
| Gao et al. | Crossing road monitoring system based on adaptive decision for illegal situation |
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 |