CN107274495A - A kind of unattended curb parking fee collecting system - Google Patents

Abstract

The invention discloses an unattended on-street parking charging system, which includes a data processing module and a video acquisition module, a data storage module and a data transmission module respectively connected thereto. The video acquisition module collects video frame data in the parking area, The data storage module stores user information, user parking information, parking fee standards and user blacklist. The data transmission module realizes the data interaction between the data processing module and the user client. The data processing module includes an image recognition unit, a billing unit and an alarm unit. The unit detects and tracks the vehicle target entering the parking area, the billing unit calculates the user's parking fee and sends an electronic invoice to the user client, and the alarm unit fails to confirm parking, leave the parking space or pay the parking fee within the specified threshold time The user client sends alarm information. The invention realizes the whole-process electronic management of identifying, billing and paying for the parking behavior of roadside vehicles.

Description

An unattended on-street parking charging system

technical field

The invention belongs to the technical field of parking charging, and in particular relates to an unattended roadside parking charging system.

Background technique

Although the number of motor vehicles in cities is increasing rapidly, the number of supporting parking lots is growing slowly, and the parking facilities are seriously lagging behind. In order to meet the needs of short-term parking, the public security traffic control department has set up parking spaces on the side of the road if the road traffic capacity permits. At present, roadside parking is mostly managed by people, and the labor intensity of the staff is high, and there are often charging disputes and malpractice for personal gain. In order to achieve unmanned management, a meter charging management system has emerged, and the owner of the car will insert coins or swipe the card by himself, and there will be a phenomenon of fee evasion.

The invention with the publication number CN104881900A discloses a vehicle roadside parking charge management system based on a radio frequency license plate with communication function, which includes a parking space buried antenna assembly distributed underground in each parking space, and is used for different vehicles passing through or staying. The vehicle's RFID radio frequency license plate with communication function reads information, and is connected to the roadside parking node information server and the roadside parking system data center to form a vehicle roadside parking fee management system based on the radio frequency license plate with communication function. The system needs to install an antenna for each parking space, and the license plate must also have a communication function, which is expensive and is not conducive to application and promotion.

The publication number is CN105303883A, which discloses a roadside parking management system and its method. The system includes a vehicle identification module, a main control server, a cloud platform, and a handheld mobile terminal; the vehicle identification module includes a camera for capturing vehicle information. The device and the network transmission module for transmitting vehicle information, the camera device is arranged at the entrance, exit and parking lot of the parking lot, connected to the main control server through the network transmission module, obtains vehicle information and sends it to the main control server. The system requires at least three cameras for vehicle identification, and a toll collector is still required.

Contents of the invention

In order to solve the technical problems raised by the above-mentioned background technology, the present invention aims to provide an unattended on-street parking charging system, which conducts electronic management of the whole process of identification, charging and payment of on-street vehicle parking behavior, without arranging a toll collector .

In order to realize above-mentioned technical purpose, technical scheme of the present invention is:

An unattended on-street parking charging system, comprising a data processing module and a video acquisition module, a data storage module and a data transmission module respectively connected thereto, the video acquisition module is used to collect video frame data in the parking area, The data storage module is used to store user information, user parking information, parking charging standards and user blacklist, the data transmission module is used to realize the data interaction between the data processing module and the user client, and the data processing module includes an image recognition unit , a billing unit and an alarm unit, the image recognition unit is used to detect and track the vehicle target driving into the parking area, the billing unit is used to calculate the parking fee of the user and send an electronic invoice to the user client, the alarm The unit is used to send alarm information to user clients who do not confirm parking, leave the parking space or pay parking fees within the specified threshold time. The user client is installed on the user's smart terminal, such as a mobile phone, a tablet computer, and the like. The video acquisition module uses a dot-matrix infrared high-definition camera, which can automatically measure light to control the switching of CCD and the opening of infrared lights, and is used to capture image information of vehicles and parking lots under different light and dark conditions.

Further, the working process of the image recognition unit is as follows:

(1) Initialize the background model of the roadside parking area, and identify the road and parking space;

(2) Detect target vehicles entering the parking area, extract target features, and identify their license plates;

(3) Carry out target tracking according to the characteristics of the target vehicle, and detect the position of the target vehicle in the background model;

(4) Judging whether the target vehicle has entered a certain parking space, if the position of the target vehicle coincides with the position of a certain parking space, and the target vehicle stops moving, then it is judged that the vehicle has been parked;

(5) After the vehicle is parked in place, update the background model;

(6) Detect the position of the target vehicle in the background model, and judge whether it leaves the parking space;

(7) After the target vehicle leaves the parking area, update the background model.

Further, the specific process of step (1) is as follows:

(A) Take and save the image when there is no vehicle in the parking area along the driving direction, and perform black and white binary processing on the image;

(B) On the image, the white part extracted from top to bottom is the boundary line between the road and the parking space, and the white part extracted from left to right is the boundary line between the parking spaces;

(C) Identify the area to the left of the boundary line between the road and the parking space as a road;

(D) For the area on the right side of the boundary line between the road and the parking space, start from the boundary line between the uppermost parking spaces on the image to find the boundary line between the nearest parking spaces below, and the boundary line between the upper and lower parking spaces The area in between is marked as a parking space;

(E) Step (D) is repeated until the boundary line between the parking spaces at the bottom of the image is marked.

Further, the specific process of step (2) is as follows:

(a) Convert three frames of images taken continuously in the parking area into grayscale images;

(b) Using the inter-frame difference method to obtain the difference image of the grayscale image, and binarizing the difference image;

(c) sequentially adopt the logical operation of "exclusive OR" and "not" to the corresponding bits of the binary difference image to obtain an image containing a moving target;

(d) Perform morphological processing on the image of the moving target, save the moving target conforming to the morphological characteristics of the vehicle, and obtain the mask of the vehicle target to be detected;

(e) Convert the color space of the image from RGB to HSV, count the color histogram of the block in the vehicle target mask, and use the blue or yellow template to identify the license plate position;

(f) Use a support vector machine to judge whether the license plate position identification block has a license plate;

(g) Carry out character segmentation and contour extraction to the license plate position identification block containing the license plate;

(h) Using the multi-layer perceptron model in the artificial neural network to perform license plate number recognition on the extracted characters.

Further, the interaction process between the data processing module and the user client is as follows:

(i) After the image recognition unit judges that the vehicle has been parked, the billing unit sends parking information to the user client bound to the license plate number of the vehicle, including the license plate number, parking start time and parking fee standard;

(ii) After confirming the information sent by the billing unit on the user client, the user feeds back the parking confirmation information to the billing unit;

(Ⅲ) After the charging unit receives the parking confirmation information, the parking information of the user is saved to the data storage module; when the charging unit does not receive the parking confirmation information fed back by the user client within the specified time threshold, the alarm unit will Send an alarm message to the user client to prompt the user to confirm the parking in time;

(iv) When the parking is over, the user sends a parking end request to the billing unit on the user client, and the billing unit, after receiving the request, retrieves the user's parking information from the data storage module, according to the parking start time , end time and charging standard, calculate the parking fee and send it to the user client;

(ⅴ) The user checks and confirms the parking fee on the user client, and pays for parking by mobile payment;

(ⅵ) After the billing unit confirms that the payment is completed, it sends an electronic invoice to the user client; when the user client fails to pay within a time threshold after receiving the parking fee, the alarm unit sends an alarm message to the user client, prompting The user pays in time, and if the payment has not been made within another time threshold after the user client receives the alarm information, the user is blacklisted, saved to the data storage module, and the license plate number of the user's vehicle is sent to traffic control department;

(ⅶ) When the image recognition unit judges that the vehicle that has submitted the parking end request has not left the parking space within the specified time threshold, the alarm unit will send an alarm message to the user client, prompting the user to leave the parking space in time, and charge at the same time The unit re-determines the parking end time, and calculates the parking fee according to the newly determined parking end time.

The beneficial effect brought by adopting the above-mentioned technical scheme:

The present invention realizes the unmanned charging of public parking spaces through the whole process of electronic management of identification, billing and payment of roadside vehicle parking behaviors, and saves human resources. The invention detects the vehicle target through the video acquisition module, tracks and records the target, judges the position and state of the target vehicle in the parking area, and does not need to bury the antenna and apply for a card, which saves construction cost and time.

Description of drawings

Fig. 1 is a system composition block diagram of the present invention;

Fig. 2 is the work flowchart of image recognition unit in the present invention;

Fig. 3 is a flow chart of the image recognition unit initializing the background model of the on-street parking area;

Fig. 4 is a flowchart of the interaction between the data processing module and the user client in the present invention.

detailed description

The technical solutions of the present invention will be described in detail below in conjunction with the accompanying drawings.

A kind of unattended on-street parking charging system, as shown in Figure 1, comprises data processing module and video acquisition module, data storage module and data transmission module connected with it respectively, and described video acquisition module is used for collecting parking area The video frame data within, the data storage module is used to store user information, user parking information, parking charging standards and user blacklist, the data transmission module is used to realize the data interaction between the data processing module and the user client, the data The processing module includes an image recognition unit, a billing unit and an alarm unit, the image recognition unit is used to detect and track the vehicle target entering the parking area, and the billing unit is used to calculate the parking fee of the user and send Electronic invoice, the alarm unit is used to send alarm information to user clients who fail to confirm parking, leave the parking space or pay the parking fee within the specified threshold time.

As shown in Figure 2, the working process of the image recognition unit is as follows:

(1) Initialize the background model of the roadside parking area, and identify the road and parking space;

(2) Detect target vehicles entering the parking area, extract target features, and identify their license plates;

(3) Carry out target tracking according to the characteristics of the target vehicle, and detect the position of the target vehicle in the background model;

(4) Judging whether the target vehicle has entered a certain parking space, if the position of the target vehicle coincides with the position of a certain parking space, and the target vehicle stops moving, then it is judged that the vehicle has been parked;

(5) After the vehicle is parked in place, update the background model;

(6) Detect the position of the target vehicle in the background model, and judge whether it leaves the parking space;

(7) After the target vehicle leaves the parking area, update the background model.

Further, as shown in Figure 3, the specific process of step (1) is as follows:

(A) Take and save the image when there is no vehicle in the parking area along the driving direction, and perform black and white binary processing on the image;

(B) On the image, the white part extracted from top to bottom is the boundary line between the road and the parking space, and the white part extracted from left to right is the boundary line between the parking spaces;

(C) Identify the area to the left of the boundary line between the road and the parking space as a road;

(D) For the area on the right side of the boundary line between the road and the parking space, start from the boundary line between the uppermost parking spaces on the image to find the boundary line between the nearest parking spaces below, and the boundary line between the upper and lower parking spaces The area in between is marked as a parking space;

(E) Step (D) is repeated until the boundary line between the parking spaces at the bottom of the image is marked.

Further, the specific process of step (2) is as follows:

(a) Convert three frames of images taken continuously in the parking area into grayscale images;

(b) Using the inter-frame difference method to obtain the difference image of the grayscale image, and binarizing the difference image;

(c) sequentially adopt the logical operation of "exclusive OR" and "not" to the corresponding bits of the binary difference image to obtain an image containing a moving target;

(d) Perform morphological processing on the image of the moving target, save the moving target conforming to the morphological characteristics of the vehicle, and obtain the mask of the vehicle target to be detected;

(e) Convert the color space of the image from RGB to HSV, count the color histogram of the block in the vehicle target mask, and use the blue or yellow template to identify the license plate position;

(f) Use a support vector machine to judge whether the license plate position identification block has a license plate;

(g) Carry out character segmentation and contour extraction to the license plate position identification block containing the license plate;

(h) Using the multi-layer perceptron model in the artificial neural network to perform license plate number recognition on the extracted characters.

As shown in Figure 4, the interaction process between the data processing module and the user client is as follows:

(i) After the image recognition unit judges that the vehicle has been parked, the billing unit sends parking information to the user client bound to the license plate number of the vehicle, including the license plate number, parking start time and parking fee standard;

(ii) After confirming the information sent by the billing unit on the user client, the user feeds back the parking confirmation information to the billing unit;

(Ⅲ) After the charging unit receives the parking confirmation information, the parking information of the user is saved to the data storage module; when the charging unit does not receive the parking confirmation information fed back by the user client within the specified time threshold, the alarm unit will Send an alarm message to the user client to prompt the user to confirm the parking in time;

(iv) When the parking is over, the user sends a parking end request to the billing unit on the user client, and the billing unit, after receiving the request, retrieves the user's parking information from the data storage module, according to the parking start time , end time and charging standard, calculate the parking fee and send it to the user client;

(ⅴ) The user checks and confirms the parking fee on the user client, and pays for parking by mobile payment;

(ⅵ) After the billing unit confirms that the payment is completed, it sends an electronic invoice to the user client.

The invention provides protection for the parking charging system by establishing an alarm unit. When the user client fails to pay within a time threshold after receiving the parking fee, the alarm unit sends an alarm message to the user client to prompt the user to pay in time, if another time threshold after the user client receives the alarm message If the payment has not been made within the period, the user is blacklisted, saved to the data storage module, and the license plate number of the user's vehicle is sent to the traffic control department. When the image recognition unit judges that the vehicle that has submitted the parking end request has not left the parking space within the specified time threshold, the alarm unit sends an alarm message to the user client to prompt the user to leave the parking space in time, and the billing unit re-determines Parking end time, the parking fee is calculated according to the newly determined parking end time.

The above embodiments are only to illustrate the technical ideas of the present invention, and cannot limit the protection scope of the present invention with this. All technical ideas proposed according to the present invention, any changes made on the basis of technical solutions, all fall within the protection scope of the present invention. Inside.

Claims (5)

1. a kind of unattended curb parking fee collecting system, it is characterised in that：Including data processing module and difference therewith Video acquisition module, data memory module and the data transmission module of connection, the video acquisition module are used to gather parking area Video requency frame data in domain, the data memory module be used for store user profile, user's parking information, parking rate with And subscriber blacklist, the data transmission module is used for the data interaction for realizing data processing module and subscription client, described Data processing module include image recognition unit, Charging Detail Record unit and alarm unit, the image recognition unit be used for detect and with Track drives into the vehicle target of parking area, and the Charging Detail Record unit is used to calculate the Parking Fee of user and sent out to subscription client Electronic invoice is sent, the alarm unit is used to stop to confirming parking not in defined threshold time, sailing out of parking stall or pay The subscription client of fare sends warning message.

2. unattended curb parking fee collecting system according to claim 1, it is characterised in that：The image recognition unit The course of work it is as follows：

(1) background model in initialization curb parking region, identifies road and parking stall；

(2) target vehicle for entering parking area is detected, extracts target signature, recognize its car plate；

(3) target following, position of the detection target vehicle in background model are carried out according to target vehicle feature；

(4) judge whether target vehicle enters certain parking stall, if target vehicle position and certain parking stall position are overlapped, and target Vehicle stop motion, then judge vehicle parking；

(5) vehicle parking enters behind position, updates background model；

(6) position of the detection target vehicle in background model, judges whether it sails out of parking stall；

(7) target vehicle is sailed out of after parking area, updates background model.

3. unattended curb parking fee collecting system according to claim 2, it is characterised in that：The specific mistake of step (1) Journey is as follows：

(A) shot along direction of traffic and preserve image when parking area does not have vehicle, black white binarization processing is carried out to image；

(B) part for extracting white from top to down on image is the boundary line of road and parking interdigit, and extremely left and right extraction is white The part of color is the boundary line of parking interdigit；

(C) road and the left field of the boundary line of parking interdigit are designated road；

(D) to the boundary line right side area of road and parking interdigit, since the boundary line of the parking interdigit of the top on image Find closest to lower section parking interdigit boundary line, up and down stop interdigit boundary line between area identification for parking Position；

(E) repeat step (D), until being identified to the boundary line of the parking interdigit of bottom on image.

4. unattended curb parking fee collecting system according to claim 2, it is characterised in that：The specific mistake of step (2) Journey is as follows：

(a) the three frame images that parking area is continuously shot are converted into gray level image；

(b) difference image of gray level image is obtained using frame difference method, and by difference image binaryzation；

(c) the difference image correspondence position of binaryzation is obtained containing motion mesh successively using the logical operation of distance and " non-" Target image；

(d) Morphological scale-space is carried out to movement destination image, the moving target for meeting vehicle morphological feature is preserved, obtained To vehicle target mask to be detected；

(e) color space of image is switched to the color histogram of segment in HSV, calculating vehicle target mask from RGB, indigo plant is used Color or yellow template come out car plate station location marker；

(f) go to differentiate whether car plate station location marker segment has car plate using SVMs；

(g) Character segmentation and contours extract are carried out to the car plate station location marker segment containing car plate；

(h) license plate number identification is carried out to the character extracted using the multiple perceptron model in artificial neural network.

5. unattended curb parking fee collecting system according to claim 2, it is characterised in that：Data processing module is with using The interaction of family client is as follows：

(I) image recognition unit judges vehicle after parking, the user visitor bound in Charging Detail Record unit to the vehicle license plate number Family end sends parking information, including license plate number, parking initial time and parking rate；

(II) user is confirmed on subscription client after the information that Charging Detail Record unit is sent, to Charging Detail Record unit feedback parking confirmation letter Breath；

(III) Charging Detail Record unit is received after parking confirmation, and the parking information of the user is preserved to data memory module；Work as meter Expense unit does not receive the parking confirmation of subscription client feedback in defined time threshold, and alarm unit will be to user visitor Family end sends warning message, points out user to confirm parking in time；

(iv) when parking terminates, user sends parking ending request on subscription client to Charging Detail Record unit, and Charging Detail Record unit is connecing Receive after the request, the parking information of the user transferred from data memory module, according to parking initial time, the end time with And expenses standard, calculate Parking Fee and send to subscription client；

(V) user is consulted on subscription client and confirms Parking Fee, and parking pay is carried out using mobile payment mode；

(VI) Charging Detail Record unit is confirmed after the completion of paying, and electronic invoice is sent to subscription client；When subscription client stops receiving Do not paid in a time threshold after fare, then alarm unit sends warning message to subscription client, point out to use Family timely payment, will if not paid yet in another time threshold after subscription client receives warning message The user is piped off, and preserves to data memory module, and the license plate number of user's vehicle is sent into best friend's pipe portion door；

(VII) judges that the vehicle for having been filed on stopping ending request is not sailed out of yet in defined time threshold when image recognition unit Parking stall, then alarm unit is to subscription client transmission warning message, and prompting user sails out of parking stall in time, while Charging Detail Record unit The parking end time is redefined, Parking Fee is calculated according to the parking end time newly determined.