CN113593250A - Illegal parking detection system based on visual identification - Google Patents
Illegal parking detection system based on visual identification Download PDFInfo
- Publication number
- CN113593250A CN113593250A CN202110786710.1A CN202110786710A CN113593250A CN 113593250 A CN113593250 A CN 113593250A CN 202110786710 A CN202110786710 A CN 202110786710A CN 113593250 A CN113593250 A CN 113593250A
- Authority
- CN
- China
- Prior art keywords
- module
- illegal parking
- information
- detection
- traffic
- 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
- 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
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/933—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of aircraft or spacecraft
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
- G05D1/106—Change initiated in response to external conditions, e.g. avoidance of elevated terrain or of no-fly zones
- G05D1/1064—Change initiated in response to external conditions, e.g. avoidance of elevated terrain or of no-fly zones specially adapted for avoiding collisions with other aircraft
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
The invention provides a system for detecting illegal parking based on visual identification, which comprises: the system comprises a traffic interconnection detection network and a violation event monitoring platform, wherein the traffic interconnection detection network is connected with the violation event monitoring platform in a wireless communication mode; the traffic interconnection detection network comprises a plurality of road parking detection devices and a plurality of unmanned aerial vehicle auxiliary detection devices, wherein the road parking detection devices are arranged in an illegal parking monitoring area of an urban road and are used for detecting illegal parking information of vehicles and giving a warning when the vehicles are illegal to park, and the unmanned aerial vehicle auxiliary detection devices are used for routing inspection on a traffic trunk road and a traffic branch road with more traffic flow; the traffic violation event monitoring platform is used for analyzing and storing the detection result of the illegal parking according to the area and issuing illegal parking information to the traffic police management terminal so as to process the illegal parking event in time.
Description
Technical Field
The invention relates to the technical field of machine vision, in particular to a system for detecting illegal parking based on visual identification.
Background
With the improvement of living standards of people and the development of transportation industry, automobiles are more and more popular as outgoing and transportation tools, and although the automobiles bring convenience for outgoing, roads are gradually crowded due to the increase of vehicles, and traffic problems occur frequently, such as illegal parking. The motor vehicle illegal parking is an important factor causing traffic jam, how to improve the monitoring means and the event processing efficiency is an important way for relieving traffic jam caused by illegal parking, most of the existing parking monitoring systems are independent detection systems, and the events cannot be intensively controlled and processed in a distributed manner, so that the event processing efficiency is lower, the detection range is small, and the detection is not accurate and comprehensive enough.
In conclusion, the provision of the system for detecting the illegal parking based on the visual identification, which can improve the traffic violation processing efficiency, has a wide detection range, and is more accurate and comprehensive, is a problem that needs to be solved urgently by technical personnel in the field.
Disclosure of Invention
In view of the above-mentioned problems and needs, the present solution provides a system for detecting illegal parking based on visual recognition, which can solve the above technical problems by adopting the following technical solutions.
In order to achieve the purpose, the invention provides the following technical scheme: a system for detecting illegal parking based on visual identification, comprising: the system comprises a traffic interconnection detection network and a violation event monitoring platform, wherein the traffic interconnection detection network is connected with the violation event monitoring platform in a wireless communication mode;
the traffic interconnection detection network comprises a plurality of road parking detection devices and a plurality of unmanned aerial vehicle auxiliary detection devices, wherein the road parking detection devices are arranged in an illegal parking monitoring area of an urban road and are used for detecting illegal parking information of vehicles and giving a warning when the vehicles are illegal to park, and the unmanned aerial vehicle auxiliary detection devices are used for routing inspection on a traffic trunk road and a traffic branch road with more traffic flow;
the illegal event monitoring platform is used for analyzing and storing illegal parking detection results according to regions and issuing illegal parking information to the traffic police management terminal so as to timely process illegal parking events.
Furthermore, each road parking detection device comprises a pan-tilt camera, a control module, an illegal parking judgment module, a remote communication module and an intelligent alarm module;
the pan-tilt camera is used for acquiring video image data of a monitoring area, decoding the video image data through a decoder after acquiring the video image data, converting a color space from a YUV format into an RGB format, and sending the processed video image data to the illegal parking judgment module;
the control module is used for controlling the rotation and the zooming of the pan-tilt, the control module controls the motion of the pan-tilt camera to selectively track the non-motor vehicles in a monitored area after receiving the control signal of the illegal parking determination module, judges whether the distance between the target position and the center of the view field exceeds a threshold range or not, calculates a rotation control parameter if the distance exceeds a set threshold, sends a command of starting the motion and stopping the motion to the control pan-tilt according to the rotation control parameter, judges whether the lens needs zooming if the distance does not exceed the set threshold, calculates the magnification if the lens needs zooming, sends a command of starting the zooming and stopping to the control pan-tilt, and otherwise, keeps the tracking state;
the illegal parking determination module is used for detecting and identifying non-motor vehicles in a monitoring area and judging tracking, border crossing and illegal parking actions of the identified detection vehicles;
the remote communication module is used for sending illegal parking information to the illegal event monitoring platform, and the illegal parking information comprises an area address, a vehicle type and tracking video data;
the intelligent alarm module is used for carrying out on-site reminding and comprises an audible and visual alarm.
Further, the illegal parking determination module comprises a vehicle detection unit, an out-of-range judgment unit and an illegal parking judgment unit;
the vehicle detection unit detects a motion area by adopting a frame difference method, then performs HOG feature extraction on a detection target, obtains a classification result through an SVM classifier by adopting an SVM-based classification algorithm, and judges whether the detection target is a vehicle or not;
the boundary crossing judging unit reads video frame information, detects a moving area by using a background difference method, and simultaneously obtains the position of a boundary area, wherein the step of detecting the moving area comprises the steps of extracting a foreground image of a video image by using the background difference method, obtaining a foreground image of a rough moving vehicle, removing most of shadows according to a set threshold value to obtain a more accurate moving vehicle foreground image, drawing the contour of the moving vehicle, deleting a rectangle with a small area, taking the middle point of the bottom edge in the contour of the moving vehicle as a moving target position, and comparing the moving target position with the set position of the boundary area to judge whether boundary crossing occurs or not;
the illegal parking judging unit tracks the vehicle by adopting a target tracking method based on fDSST algorithm, and judges that the vehicle behavior is illegal parking when the fact that the target stays in the out-of-range area for more than a certain time length is detected.
Furthermore, when the position of the moving target is compared with the position of the set boundary area, the monitoring rectangular area and the position of the moving target are subjected to perspective transformation to obtain a new position, so that the object in the image shot by the camera is prevented from being distorted, the new position is taken as the position of the moving target to be compared with the position of the set boundary area, and the image is projected to a new plane through the perspective transformation through a perspective matrix.
Further, the extraction process of the HOG feature extraction includes: acquiring a detection window, carrying out graying processing on an image, and carrying out color space standardization on an input image by adopting a Gamma correction method, so that the influence caused by local shadow and illumination change of the image is reduced, and the interference of noise is inhibited; capturing contour information, and calculating the gradient of each pixel of the image, including the size and the direction; dividing the image blocks into small units, counting the gradient histogram of each unit, and then comparing and normalizing each overlapped image block to obtain the HOG feature vector.
Further, the target tracking method specifically includes: combining the gray scale of a sample image and the HOG characteristics of the sample as the multi-dimensional characteristics of an input sample f, obtaining ideal output through training, determining a new target position by using a two-dimensional position filter in a new frame of image, then obtaining candidate blocks with different scales under a one-dimensional scale correlation filter by taking the current central position as a central point, and finally finding out the optimal matching scale.
Furthermore, each unmanned aerial vehicle auxiliary detection device comprises an unmanned aerial vehicle control panel, an IMU inertia measurement unit and a second illegal parking judgment unit;
the unmanned aerial vehicle control panel comprises a power driving module, a positioning module, a radar obstacle avoidance module and a flight control module, the flight control module controls the power driving module to drive a power motor according to set cruising route information and current flying speed, flying direction and attitude information of the unmanned aerial vehicle detected by an IMU inertial measurement unit, so as to control flying angle and speed, and meanwhile, automatic obstacle avoidance flight is carried out when an obstacle is detected through the radar obstacle avoidance module, and the positioning module is used for acquiring real-time spatial position information;
the second illegal parking judgment unit comprises a map transmission device and a ground analysis module, the map transmission device transmits image data acquired by the high-precision digital imaging equipment to the ground analysis module, and the ground analysis module adopts a SSD-based vehicle illegal parking judgment model to perform illegal parking analysis.
Furthermore, the method comprises the steps of performing feature extraction on an input traffic monitoring picture by using a target feature extraction network based on the SSD, converting two full-connection layers of the VGG16 network into convolution layers, then adding four convolution layers to obtain an SSD network structure, performing convolution on an image through each layer of the network to generate feature maps with different sizes, performing Softmax classification and position regression simultaneously by using a plurality of feature maps, and performing Softmax classification and position regression according to the feature maps
Calculating the size of the target detection frame, wherein UiDenotes the size, U, of the ith feature mapminIs the size of the minimum feature map, UmaxM is the number of the feature maps, when the SSD-based target feature extraction network is subjected to model training,setting the target detection frame to different aspect ratios; classifying and identifying the extracted feature maps to obtain an identification result, then adopting a non-maximum suppression algorithm to obtain an optimal target detection frame, and sequencing all the target detection frames according to the confidence degree of the obtained target detection frames; and then calculating the areas of all the prediction frames, calculating IoU of the target detection frame with the highest confidence coefficient and each remaining candidate frame, deleting IoU candidate frames with values larger than the threshold value according to the set threshold value, outputting the final detection result of the vehicle, and judging that the behavior of the vehicle is illegal parking if the target vehicle stays in the out-of-range area for more than a certain time length.
Further, the violation event monitoring platform comprises a violation information management module and an information issuing module, the violation information management module comprises a structured storage module, an information inquiry module and a notification reminding module, the structured storage module is used for storing the video information of the illegal parking and the geographical position information and the time information of the illegal parking in different structures, the information inquiry module is used for inquiring the processed events and the unprocessed events by the manager through the inquiry list, the information query module comprises a query list and an event marking module, the query list comprises geographic position information and time information, the event marking module is used for automatically updating the event processing state according to feedback information sent by the traffic police management terminal, and the notification reminding module is used for marking and reminding unprocessed events in a certain period; and the information issuing module sends the illegal parking information to the traffic police management terminal corresponding to the management area according to the illegal area address information.
According to the technical scheme, the invention has the beneficial effects that: the traffic violation processing system can improve the traffic violation processing efficiency, has a wide detection range, is more accurate and comprehensive, and better assists traffic management personnel in carrying out violation recording and event processing.
In addition to the above objects, features and advantages, preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings so that the features and advantages of the present invention can be easily understood.
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 of the present invention or the prior art will be briefly described, wherein the drawings are only used for illustrating some embodiments of the present invention and do not limit all embodiments of the present invention thereto.
Fig. 1 is a schematic view of the composition structure of the system for detecting illegal parking based on visual identification.
Fig. 2 is a schematic view of the structure of each road parking detection device according to the present invention.
Fig. 3 is a schematic view of the composition structure of each auxiliary detection device of the unmanned aerial vehicle.
Fig. 4 is a schematic diagram illustrating specific steps of the road parking analysis process in this embodiment.
Fig. 5 is a schematic diagram of specific steps of an unmanned aerial vehicle violation analysis process in this embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of specific embodiments of the present invention. Like reference symbols in the various drawings indicate like elements. It should be noted that the described embodiments are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention without any inventive step, are within the scope of protection of the invention.
The invention provides the system for detecting the illegal parking based on the visual identification, which can improve the traffic violation processing efficiency, has a wide detection range and is more accurate and comprehensive. As shown in fig. 1 to 4, the system includes: the traffic interconnection detection network is connected with the violation event monitoring platform in a wireless communication mode. The traffic interconnection detection network comprises a plurality of road parking detection devices and a plurality of unmanned aerial vehicle auxiliary detection devices, the plurality of road parking detection devices are arranged in illegal parking monitoring areas of urban roads and used for detecting illegal parking information of vehicles and warning when the vehicles are illegal to park, and the plurality of unmanned aerial vehicle auxiliary detection devices are used for routing inspection on traffic main roads and traffic branches with more traffic flow.
In the system, each road parking detection device comprises a pan-tilt camera, a control module, an illegal parking judgment module, a remote communication module and an intelligent alarm module; the pan-tilt camera is used for acquiring video image data of a monitoring area, decoding the video image data through a decoder after acquiring the video image data, converting a color space from a YUV format into an RGB format, and sending the processed video image data to the illegal parking judgment module; the control module is used for controlling the rotation and the zooming of the pan-tilt, the control module controls the motion of the pan-tilt camera to selectively track the non-motor vehicles in a monitored area after receiving the control signal of the illegal parking determination module, judges whether the distance between the target position and the center of the view field exceeds a threshold range or not, calculates a rotation control parameter if the distance exceeds a set threshold, sends a command of starting the motion and stopping the motion to the control pan-tilt according to the rotation control parameter, judges whether the lens needs zooming if the distance does not exceed the set threshold, calculates the magnification if the lens needs zooming, sends a command of starting the zooming and stopping to the control pan-tilt, and otherwise, keeps the tracking state; the illegal parking determination module is used for detecting and identifying non-motor vehicles in a monitoring area and judging tracking, border crossing and illegal parking actions of the identified detection vehicles; the remote communication module is used for sending illegal parking information to the illegal event monitoring platform, and the illegal parking information comprises an area address, a vehicle type and tracking video data; the intelligent alarm module is used for carrying out on-site reminding and comprises an audible and visual alarm. In this embodiment, the remote communication module includes a dedicated traffic communication network device, an ethernet module, a WIFI module, and the like, and the alarm module may also be other reminding devices, such as a light, a voice, or an LED electronic notice board.
The illegal parking judgment module comprises a vehicle detection unit, a border crossing judgment unit and an illegal parking judgment unit; as shown in fig. 4, the road parking analysis process is as follows: a. the vehicle detection unit detects a motion area by adopting a frame difference method, then performs HOG feature extraction on a detection target, obtains a classification result through an SVM classifier by adopting an SVM-based classification algorithm, and judges whether the detection target is a vehicle or not; b. the method comprises the steps that a boundary crossing judging unit reads video frame information, detects a motion area by using a background difference method and simultaneously obtains the position of a boundary area, the detection of the motion area comprises the steps of extracting a foreground image of a video image by using the background difference method, obtaining a foreground image of a rough motion vehicle, removing most of shadows according to a set threshold value to obtain a more accurate foreground image of the motion vehicle, drawing the outline of the motion vehicle, deleting a rectangle with a small area, and taking the middle point of the bottom edge in the outline of the motion vehicle as a motion target position; c. comparing the position of the moving target with the position of a set boundary area, and judging whether boundary crossing occurs or not; d. the illegal parking judging unit tracks the vehicle by adopting a target tracking method based on fDSST algorithm, and judges that the vehicle behavior is illegal parking when the fact that the target stays in the out-of-range area for more than a certain time length is detected. Wherein the extraction process of the HOG feature extraction comprises the following steps: acquiring a detection window, carrying out graying processing on an image, and carrying out color space standardization on an input image by adopting a Gamma correction method, so that the influence caused by local shadow and illumination change of the image is reduced, and the interference of noise is inhibited; capturing contour information, and calculating the gradient of each pixel of the image, including the size and the direction; dividing the image blocks into small units, counting the gradient histogram of each unit, and then comparing and normalizing each overlapped image block to obtain the HOG feature vector. The target tracking method specifically comprises the following steps: combining the gray scale of a sample image and the HOG characteristics of the sample as the multi-dimensional characteristics of an input sample f, obtaining ideal output through training, determining a new target position by using a two-dimensional position filter in a new frame of image, then obtaining candidate blocks with different scales under a one-dimensional scale correlation filter by taking the current central position as a central point, and finally finding out the optimal matching scale. And when the position of the moving target is compared with the position of the set boundary area, the monitoring rectangular area and the position of the moving target are subjected to perspective transformation to obtain a new position, so that the object in the image shot by the camera is prevented from being distorted, the new position is taken as the position of the moving target to be compared with the position of the set boundary area, and the image is projected to a new plane through the perspective transformation through a perspective matrix.
As shown in fig. 3, each unmanned aerial vehicle auxiliary detection device includes an unmanned aerial vehicle control panel, an IMU inertial measurement unit, and a second illegal parking determination unit; the unmanned aerial vehicle control panel comprises a power driving module, a positioning module, a radar obstacle avoidance module and a flight control module, the flight control module controls the power driving module to drive a power motor according to set cruising route information and current flying speed, flying direction and attitude information of the unmanned aerial vehicle detected by an IMU inertial measurement unit, so as to control flying angle and speed, and meanwhile, automatic obstacle avoidance flight is carried out when an obstacle is detected through the radar obstacle avoidance module, and the positioning module is used for acquiring real-time spatial position information; the second illegal parking judgment unit comprises a map transmission device and a ground analysis module, the map transmission device transmits image data acquired by the high-precision digital imaging equipment to the ground analysis module, and the ground analysis module adopts a SSD-based vehicle illegal parking judgment model to perform illegal parking analysis.
As shown in fig. 5, the unmanned plane violation analysis process is as follows: a. the method comprises the steps of utilizing a target feature extraction network based on the SSD to carry out feature extraction on an input traffic monitoring picture, changing two full-connection layers of a VGG16 network into convolution layers, then adding four convolution layers to obtain an SSD network structure, carrying out convolution on an image through each layer of network to generate feature maps with different sizes, utilizing a plurality of feature maps to carry out Softmax classification and position regression simultaneously, and carrying out Softmax classification and position regression according to the feature maps
Calculating the size of the target detection frame, wherein UiDenotes the size, U, of the ith feature mapminIs the size of the minimum feature map, UmaxSetting a target detection frame to be different in aspect ratio when the SSD-based target feature extraction network is subjected to model training, wherein m is the number of feature maps and is the size of the maximum feature map; b. to pairClassifying and identifying the extracted feature maps to obtain an identification result; c. then, obtaining an optimal target detection frame by adopting a non-maximum suppression algorithm, and sequencing all the target detection frames according to the confidence degree of the obtained target detection frames; d. then calculating the areas of all the prediction frames, calculating IoU of the target detection frame with the highest confidence coefficient and each remaining candidate frame, deleting IoU candidate frames with values larger than the threshold value according to the set threshold value, and outputting the final detection result of the vehicle; e. and if the target vehicle stays in the border crossing area for more than a certain time, judging that the vehicle behavior is illegal parking.
In the embodiment, the coverage of violation detection and the detection accuracy are improved in a mode of combining road detection and unmanned aerial vehicle detection, meanwhile, violation information monitored by the whole network is centrally processed and monitored through a violation event monitoring platform, the conditions of roads in various areas are mastered in real time, and the violation event monitoring platform is used for analyzing violation parking detection results according to the areas and storing data, and issuing violation parking information to a traffic police management terminal so as to timely process violation parking events. The violation event monitoring platform comprises a violation information management module and an information issuing module, the violation information management module comprises a structured storage module, an information inquiry module and a notification reminding module, the structured storage module is used for storing the video information of the illegal parking and the geographical position information and the time information of the illegal parking in different structures, the information inquiry module is used for inquiring the processed events and the unprocessed events by the manager through the inquiry list, the information query module comprises a query list and an event marking module, the query list comprises geographic position information and time information, the event marking module is used for automatically updating the event processing state according to feedback information sent by the traffic police management terminal, and the notification reminding module is used for marking and reminding unprocessed events in a certain period; and the information issuing module sends the illegal parking information to the traffic police management terminal corresponding to the management area according to the illegal area address information. In this embodiment, traffic management personnel accessible traffic police management terminal in time obtains the concrete position and the handling condition of parking violating the regulations, improves incident treatment effeciency, has handled this act of violating the regulations through the feedback simultaneously, in time records incident treatment process, prevents to miss the condition of examining.
It should be noted that the described embodiments of the invention are only preferred ways of implementing the invention, and that all obvious modifications, which are within the scope of the invention, are all included in the present general inventive concept.
Claims (9)
1. A system for detecting illegal parking based on visual identification is characterized by comprising: the system comprises a traffic interconnection detection network and a violation event monitoring platform, wherein the traffic interconnection detection network is connected with the violation event monitoring platform in a wireless communication mode;
the traffic interconnection detection network comprises a plurality of road parking detection devices and a plurality of unmanned aerial vehicle auxiliary detection devices, wherein the road parking detection devices are arranged in an illegal parking monitoring area of an urban road and are used for detecting illegal parking information of vehicles and giving a warning when the vehicles are illegal to park, and the unmanned aerial vehicle auxiliary detection devices are used for routing inspection on a traffic trunk road and a traffic branch road with more traffic flow;
the illegal event monitoring platform is used for analyzing and storing illegal parking detection results according to regions and issuing illegal parking information to the traffic police management terminal so as to timely process illegal parking events.
2. The system for detecting illegal parking based on visual identification as claimed in claim 1, wherein each road parking detection device comprises a pan-tilt camera, a control module, an illegal parking judgment module, a remote communication module and an intelligent alarm module;
the pan-tilt camera is used for acquiring video image data of a monitoring area, decoding the video image data through a decoder after acquiring the video image data, converting a color space from a YUV format into an RGB format, and sending the processed video image data to the illegal parking judgment module;
the control module is used for controlling the rotation and the zooming of the pan-tilt, the control module controls the motion of the pan-tilt camera to selectively track the non-motor vehicles in a monitored area after receiving the control signal of the illegal parking determination module, judges whether the distance between the target position and the center of the view field exceeds a threshold range or not, calculates a rotation control parameter if the distance exceeds a set threshold, sends a command of starting the motion and stopping the motion to the control pan-tilt according to the rotation control parameter, judges whether the lens needs zooming if the distance does not exceed the set threshold, calculates the magnification if the lens needs zooming, sends a command of starting the zooming and stopping to the control pan-tilt, and otherwise, keeps the tracking state;
the illegal parking determination module is used for detecting and identifying non-motor vehicles in a monitoring area and judging tracking, border crossing and illegal parking actions of the identified detection vehicles;
the remote communication module is used for sending illegal parking information to the illegal event monitoring platform, and the illegal parking information comprises an area address, a vehicle type and tracking video data;
the intelligent alarm module is used for carrying out on-site reminding and comprises an audible and visual alarm.
3. The system for detecting illegal parking based on visual identification according to claim 2 is characterized in that the illegal parking judging module comprises a vehicle detecting unit, a border-crossing judging unit and an illegal parking judging unit;
the vehicle detection unit detects a motion area by adopting a frame difference method, then performs HOG feature extraction on a detection target, obtains a classification result through an SVM classifier by adopting an SVM-based classification algorithm, and judges whether the detection target is a vehicle or not;
the boundary crossing judging unit reads video frame information, detects a moving area by using a background difference method, and simultaneously obtains the position of a boundary area, wherein the step of detecting the moving area comprises the steps of extracting a foreground image of a video image by using the background difference method, obtaining a foreground image of a rough moving vehicle, removing most of shadows according to a set threshold value to obtain a more accurate moving vehicle foreground image, drawing the contour of the moving vehicle, deleting a rectangle with a small area, taking the middle point of the bottom edge in the contour of the moving vehicle as a moving target position, and comparing the moving target position with the set position of the boundary area to judge whether boundary crossing occurs or not;
the illegal parking judging unit tracks the vehicle by adopting a target tracking method based on fDSST algorithm, and judges that the vehicle behavior is illegal parking when the fact that the target stays in the out-of-range area for more than a certain time length is detected.
4. The system of claim 3, wherein when comparing the moving target position with the set boundary area position, the monitoring rectangular area and the moving target position are subjected to perspective transformation to obtain a new position, so as to avoid distortion of an object in an image captured by the camera, and then the new position is used as the moving target position to be compared with the set boundary area position, and the perspective transformation projects the image to a new plane through a perspective matrix.
5. The system of claim 4 wherein the HOG feature extraction process includes: acquiring a detection window, carrying out graying processing on an image, and carrying out color space standardization on an input image by adopting a Gamma correction method, so that the influence caused by local shadow and illumination change of the image is reduced, and the interference of noise is inhibited; capturing contour information, and calculating the gradient of each pixel of the image, including the size and the direction; dividing the image blocks into small units, counting the gradient histogram of each unit, and then comparing and normalizing each overlapped image block to obtain the HOG feature vector.
6. The system for detecting illegal parking based on visual identification according to claim 3 is characterized in that the target tracking method specifically comprises the following steps: combining the gray scale of a sample image and the HOG characteristics of the sample as the multi-dimensional characteristics of an input sample f, obtaining ideal output through training, determining a new target position by using a two-dimensional position filter in a new frame of image, then obtaining candidate blocks with different scales under a one-dimensional scale correlation filter by taking the current central position as a central point, and finally finding out the optimal matching scale.
7. The visual identification-based illegal parking detection system according to claim 1, wherein each unmanned aerial vehicle auxiliary detection device comprises an unmanned aerial vehicle control board, an IMU inertial measurement unit and a second illegal parking judgment unit;
the unmanned aerial vehicle control panel comprises a power driving module, a positioning module, a radar obstacle avoidance module and a flight control module, the flight control module controls the power driving module to drive a power motor according to set cruising route information and current flying speed, flying direction and attitude information of the unmanned aerial vehicle detected by an IMU inertial measurement unit, so as to control flying angle and speed, and meanwhile, automatic obstacle avoidance flight is carried out when an obstacle is detected through the radar obstacle avoidance module, and the positioning module is used for acquiring real-time spatial position information;
the second illegal parking judgment unit comprises a map transmission device and a ground analysis module, the map transmission device transmits image data acquired by the high-precision digital imaging equipment to the ground analysis module, and the ground analysis module adopts a SSD-based vehicle illegal parking judgment model to perform illegal parking analysis.
8. The system of claim 7, wherein the incoming traffic monitoring images are feature extracted using an SSD-based target feature extraction network, two fully-connected layers of the VGG16 network are replaced with convolutional layers, four convolutional layers are added to obtain an SSD network structure, after the images are convolved through each layer of the network, feature maps of different sizes are generated, Softmax classification and position regression can be simultaneously performed using a plurality of feature maps, and the method is based on the principle that the incoming traffic monitoring images are feature extracted using an SSD-based target feature extraction network
Calculating the size of the target detection frame, wherein UiDenotes the size, U, of the ith feature mapminIs the size of the minimum feature map, UmaxIs the size of the maximum feature map, m is the number of feature maps, in the pair of the baseSetting a target detection box to different aspect ratios when a target feature extraction network of the SSD carries out model training; classifying and identifying the extracted feature maps to obtain an identification result, then adopting a non-maximum suppression algorithm to obtain an optimal target detection frame, and sequencing all the target detection frames according to the confidence degree of the obtained target detection frames; and then calculating the areas of all the prediction frames, calculating IoU of the target detection frame with the highest confidence coefficient and each remaining candidate frame, deleting IoU candidate frames with values larger than the threshold value according to the set threshold value, outputting the final detection result of the vehicle, and judging that the behavior of the vehicle is illegal parking if the target vehicle stays in the out-of-range area for more than a certain time length.
9. The system for detecting illegal parking based on visual identification as claimed in claim 1, wherein the platform for monitoring illegal event comprises a violation information management module and an information issuing module, the violation information management module comprises a structured storage module, an information inquiry module and a notice reminding module, the structured storage module is used for storing video information of illegal parking and geographical location information and time information of illegal parking in different structures, the information inquiry module is used for an administrator to inquire processed events and unprocessed events through an inquiry list, the information inquiry module comprises an inquiry list and an event marking module, the inquiry list comprises geographical location information and time information, the event marking module is used for automatically updating the event processing state according to the feedback information sent by the traffic police management terminal, the notification reminding module is used for marking and reminding unprocessed events in a certain period; and the information issuing module sends the illegal parking information to the traffic police management terminal corresponding to the management area according to the illegal area address information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110786710.1A CN113593250A (en) | 2021-07-12 | 2021-07-12 | Illegal parking detection system based on visual identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110786710.1A CN113593250A (en) | 2021-07-12 | 2021-07-12 | Illegal parking detection system based on visual identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113593250A true CN113593250A (en) | 2021-11-02 |
Family
ID=78247105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110786710.1A Pending CN113593250A (en) | 2021-07-12 | 2021-07-12 | Illegal parking detection system based on visual identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113593250A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114267182A (en) * | 2021-12-29 | 2022-04-01 | 湖南凌翔磁浮科技有限责任公司 | Violation reminding robot system and violation reminding method |
| CN114512005A (en) * | 2022-02-15 | 2022-05-17 | 复亚智能科技(太仓)有限公司 | Road self-inspection method and device, unmanned aerial vehicle and storage medium |
| CN114648748A (en) * | 2022-05-23 | 2022-06-21 | 科大天工智能装备技术（天津）有限公司 | Motor vehicle illegal parking intelligent identification method and system based on deep learning |
| CN114664093A (en) * | 2022-05-20 | 2022-06-24 | 湖南第一师范学院 | Traffic control system based on computer vision |
| CN115082903A (en) * | 2022-08-24 | 2022-09-20 | 深圳市万物云科技有限公司 | Non-motor vehicle illegal parking identification method and device, computer equipment and storage medium |
| CN115188091A (en) * | 2022-07-13 | 2022-10-14 | 国网江苏省电力有限公司泰州供电分公司 | Unmanned aerial vehicle grid inspection system and method integrating power transmission and transformation equipment |
| CN116264037A (en) * | 2023-05-17 | 2023-06-16 | 山东科技大学 | Illegal parking detection system and method based on intelligent network vehicle connection |
| CN117037501A (en) * | 2023-10-10 | 2023-11-10 | 成都创一博通科技有限公司 | Urban parking management method and management system based on artificial intelligence |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389988A (en) * | 2015-12-07 | 2016-03-09 | 北京航空航天大学 | Multi-unmanned aerial vehicle cooperation highway intelligent inspection system |
| CN106741895A (en) * | 2016-12-28 | 2017-05-31 | 合肥工业大学 | The operating method of inspection unmanned plane violating the regulations and inspection unmanned plane violating the regulations |
| CN107273880A (en) * | 2017-07-31 | 2017-10-20 | 秦皇岛玥朋科技有限公司 | A kind of multi-storied garage safety-protection system and method based on intelligent video monitoring |
| CN108831158A (en) * | 2018-08-20 | 2018-11-16 | 贵州宜行智通科技有限公司 | It disobeys and stops monitoring method, device and electric terminal |
| CN109348179A (en) * | 2018-11-06 | 2019-02-15 | 常州信息职业技术学院 | A kind of road monitoring detection system and method based on artificial intelligence |
| CN110443178A (en) * | 2019-07-29 | 2019-11-12 | 浙江工贸职业技术学院 | A kind of monitoring system and its method of vehicle violation parking |
| CN111260738A (en) * | 2020-01-08 | 2020-06-09 | 天津大学 | Multi-scale target tracking method based on relevant filtering and self-adaptive feature fusion |
| CN111667514A (en) * | 2020-05-22 | 2020-09-15 | 浙江工贸职业技术学院 | Quick and accurate vehicle tracking system |
| CN112464982A (en) * | 2020-10-27 | 2021-03-09 | 河北科技大学 | Target detection model, method and application based on improved SSD algorithm |
-
2021
- 2021-07-12 CN CN202110786710.1A patent/CN113593250A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389988A (en) * | 2015-12-07 | 2016-03-09 | 北京航空航天大学 | Multi-unmanned aerial vehicle cooperation highway intelligent inspection system |
| CN106741895A (en) * | 2016-12-28 | 2017-05-31 | 合肥工业大学 | The operating method of inspection unmanned plane violating the regulations and inspection unmanned plane violating the regulations |
| CN107273880A (en) * | 2017-07-31 | 2017-10-20 | 秦皇岛玥朋科技有限公司 | A kind of multi-storied garage safety-protection system and method based on intelligent video monitoring |
| CN108831158A (en) * | 2018-08-20 | 2018-11-16 | 贵州宜行智通科技有限公司 | It disobeys and stops monitoring method, device and electric terminal |
| CN109348179A (en) * | 2018-11-06 | 2019-02-15 | 常州信息职业技术学院 | A kind of road monitoring detection system and method based on artificial intelligence |
| CN110443178A (en) * | 2019-07-29 | 2019-11-12 | 浙江工贸职业技术学院 | A kind of monitoring system and its method of vehicle violation parking |
| CN111260738A (en) * | 2020-01-08 | 2020-06-09 | 天津大学 | Multi-scale target tracking method based on relevant filtering and self-adaptive feature fusion |
| CN111667514A (en) * | 2020-05-22 | 2020-09-15 | 浙江工贸职业技术学院 | Quick and accurate vehicle tracking system |
| CN112464982A (en) * | 2020-10-27 | 2021-03-09 | 河北科技大学 | Target detection model, method and application based on improved SSD algorithm |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114267182A (en) * | 2021-12-29 | 2022-04-01 | 湖南凌翔磁浮科技有限责任公司 | Violation reminding robot system and violation reminding method |
| CN114512005A (en) * | 2022-02-15 | 2022-05-17 | 复亚智能科技(太仓)有限公司 | Road self-inspection method and device, unmanned aerial vehicle and storage medium |
| CN114664093A (en) * | 2022-05-20 | 2022-06-24 | 湖南第一师范学院 | Traffic control system based on computer vision |
| CN114648748A (en) * | 2022-05-23 | 2022-06-21 | 科大天工智能装备技术（天津）有限公司 | Motor vehicle illegal parking intelligent identification method and system based on deep learning |
| CN115188091A (en) * | 2022-07-13 | 2022-10-14 | 国网江苏省电力有限公司泰州供电分公司 | Unmanned aerial vehicle grid inspection system and method integrating power transmission and transformation equipment |
| CN115188091B (en) * | 2022-07-13 | 2023-10-13 | 国网江苏省电力有限公司泰州供电分公司 | Unmanned aerial vehicle gridding inspection system and method integrating power transmission and transformation equipment |
| CN115082903A (en) * | 2022-08-24 | 2022-09-20 | 深圳市万物云科技有限公司 | Non-motor vehicle illegal parking identification method and device, computer equipment and storage medium |
| CN115082903B (en) * | 2022-08-24 | 2022-11-11 | 深圳市万物云科技有限公司 | Non-motor vehicle illegal parking identification method and device, computer equipment and storage medium |
| CN116264037A (en) * | 2023-05-17 | 2023-06-16 | 山东科技大学 | Illegal parking detection system and method based on intelligent network vehicle connection |
| CN117037501A (en) * | 2023-10-10 | 2023-11-10 | 成都创一博通科技有限公司 | Urban parking management method and management system based on artificial intelligence |
| CN117037501B (en) * | 2023-10-10 | 2023-12-12 | 成都创一博通科技有限公司 | Urban parking management method and management system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113593250A (en) | Illegal parking detection system based on visual identification | |
| Rasheed et al. | Automated number plate recognition using hough lines and template matching | |
| CN103824452B (en) | A kind of peccancy parking detector based on panoramic vision of lightweight | |
| CN104506804B (en) | Motor vehicle abnormal behaviour monitoring device and its method on a kind of through street | |
| CN106373430B (en) | Intersection traffic early warning method based on computer vision | |
| Negru et al. | Image based fog detection and visibility estimation for driving assistance systems | |
| CN110717387B (en) | Real-time vehicle detection method based on unmanned aerial vehicle platform | |
| CA3124173C (en) | System and method for detecting and transmitting incidents of interest of a roadway to a remote server | |
| KR102122859B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| SG191237A1 (en) | Calibration device and method for use in a surveillance system for event detection | |
| US11436839B2 (en) | Systems and methods of detecting moving obstacles | |
| KR102122850B1 (en) | Solution for analysis road and recognition vehicle license plate employing deep-learning | |
| US9292743B1 (en) | Background modeling for fixed, mobile, and step- and-stare video camera surveillance | |
| CN115797873B (en) | Crowd density detection method, system, equipment, storage medium and robot | |
| CN111967396A (en) | Processing method, device and equipment for obstacle detection and storage medium | |
| CN112084892B (en) | Road abnormal event detection management device and method thereof | |
| CN114841910A (en) | Vehicle-mounted lens shielding identification method and device | |
| CN112257683A (en) | Cross-mirror tracking method for vehicle running track monitoring | |
| FAN et al. | Robust lane detection and tracking based on machine vision | |
| CN112926415A (en) | Pedestrian avoiding system and pedestrian monitoring method | |
| Matsuda et al. | A system for real-time on-street parking detection and visualization on an edge device | |
| CN115761668A (en) | Camera stain recognition method and device, vehicle and storage medium | |
| CN112686136B (en) | Object detection method, device and system | |
| CN112818837B (en) | Aerial photography vehicle weight recognition method based on attitude correction and difficult sample perception | |
| CN113408514A (en) | Method and device for detecting roadside parking lot berth based on deep learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211102 |
|
| RJ01 | Rejection of invention patent application after publication |