Non-motor vehicle illegal parking detection method based on cloud-side cooperation

Abstract

The invention is suitable for the technical field of non-motor vehicle illegal parking detection, and provides a non-motor vehicle illegal parking detection method based on cloud-side cooperation. According to the invention, detection shooting is carried out through the detection camera, and illegal parking non-motor vehicles are determined based on the cloud computing technology; carrying out license plate recognition, and judging whether recognition is successful; if the identification cannot be successful, illegal parking placement identification and collaborative analysis are carried out, and collaborative motor vehicles are selected; a collaborative shooting image is acquired; and carrying out license plate recognition on illegal parking non-motor vehicles, obtaining illegal parking sign information, and carrying out illegal parking expelling alarm. According to the invention, illegal parking non-motor vehicles can be identified based on cloud edge cooperation, and when license plate identification is not successful, cooperative motor vehicles are selected, cooperative shooting images are acquired, license plate identification is carried out on the illegal parking non-motor vehicles, and illegal parking expelling alarm is carried out. The technical defect that license plates of part of non-motor vehicles in illegal parking cannot be clearly captured or completely recorded due to the poor shooting angle is overcome, and the comprehensiveness and accuracy of non-motor vehicle illegal parking detection are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of non-motor vehicle illegal parking detection, and in particular relates to a non-motor vehicle illegal parking detection method based on cloud-edge collaboration. Background Art

[0002] With the acceleration of urbanization and the increasing demand for environmentally friendly travel, non-motor vehicles have become one of the important means of transportation. However, due to the lack of effective management measures, the problem of random parking of non-motor vehicles is becoming increasingly serious, which not only affects the appearance of the city, but also may hinder traffic.

[0003] Non-motor vehicle illegal parking detection refers to the process of identifying and managing the illegal parking of non-motor vehicles such as electric bicycles on urban roads or in public places.

[0004] In the prior art, since the position, height and / or angle of non-motor vehicle illegal parking detection shooting is fixed and the on-site environment is relatively complex, when illegally parked non-motor vehicles are detected, there is a technical defect that the license plates of some illegally parked non-motor vehicles cannot be clearly captured or fully recorded due to poor shooting angles, which limits the comprehensiveness and accuracy of non-motor vehicle illegal parking detection. Summary of the invention

[0005] The purpose of an embodiment of the present invention is to provide a non-motor vehicle illegal parking detection method based on cloud-edge collaboration, aiming to solve the technical problems existing in the prior art mentioned in the background technology.

[0006] The embodiment of the present invention is implemented as follows:

[0007] A non-motor vehicle illegal parking detection method based on cloud-edge collaboration, the method specifically comprises the following steps:

[0008] Through real-time detection and shooting by detection cameras, real-time shooting data is obtained and uploaded. Based on cloud computing technology, prohibited parking identification and timing comparison are performed to determine the illegally parked non-motor vehicles;

[0009] Extracting multiple tracking images of the illegally parked non-motor vehicle from the real-time shooting data, and performing license plate recognition to determine whether the recognition is successful;

[0010] When the recognition is unsuccessful, extract the illegal parking environment image from the real-time shooting data, perform illegal parking placement recognition and collaborative analysis, and select the collaborative motor vehicle;

[0011] generating and sending a collaborative inspection application to the collaborative motor vehicle, and acquiring a collaboratively shot image after the collaborative inspection application is approved;

[0012] According to the collaboratively shot image, the license plate of the illegally parked non-motor vehicle is recognized, the illegal parking license plate information is obtained, the illegal parking main information is matched, and an illegal parking removal alarm is issued.

[0013] As a further limitation of the technical solution of the embodiment of the present invention, the real-time detection and shooting by the detection camera, obtaining and uploading the real-time shooting data, and performing the prohibited parking identification and timing comparison based on the cloud computing technology to determine the illegally parked non-motor vehicle specifically include the following steps:

[0014] Perform real-time detection and shooting through the detection camera to obtain real-time shooting data;

[0015] Uploading the real-time shooting data according to a preset cloud address;

[0016] Based on cloud computing technology, the real-time shooting data is used to identify the prohibited parking and determine the prohibited parking of non-motor vehicles;

[0017] The prohibited parking time of the prohibited non-motor vehicle is calculated, and when the prohibited parking time exceeds a preset prohibited parking threshold, the prohibited non-motor vehicle is marked as an illegally parked non-motor vehicle.

[0018] As a further limitation of the technical solution of the embodiment of the present invention, the method of performing prohibited parking identification on the real-time shooting data based on cloud computing technology and determining prohibited parking of non-motor vehicles specifically includes the following steps:

[0019] Based on cloud computing technology, the real-time shooting data is used to identify the area and determine the prohibited parking areas;

[0020] Based on cloud computing technology, the real-time shooting data is used to identify non-motor vehicles entering the camera and determine multiple target non-motor vehicles;

[0021] Tracking and identifying the plurality of target non-motor vehicles, and marking the target non-motor vehicles entering the prohibited parking area as prohibited parking non-motor vehicles.

[0022] As a further limitation of the technical solution of the embodiment of the present invention, the extracting of multiple tracking images of the illegally parked non-motor vehicle from the real-time shooting data, performing license plate recognition, and determining whether the recognition is successful specifically include the following steps:

[0023] Extracting partial shooting data of illegally parked non-motor vehicles from the real-time shooting data;

[0024] Processing the partial shooting data frame by frame to obtain a plurality of partial frame-by-frame images;

[0025] Selecting a plurality of tracking images from the plurality of partial frame-by-frame images;

[0026] Perform license plate recognition on the plurality of tracking and shooting images to determine whether the recognition is successful.

[0027] As a further limitation of the technical solution of the embodiment of the present invention, when the recognition is unsuccessful, extracting the illegal parking environment image from the real-time shooting data, performing illegal parking placement recognition and collaborative analysis, and selecting the collaborative motor vehicle specifically include the following steps:

[0028] When the recognition is unsuccessful, selecting an illegal parking environment image from the plurality of tracking and shooting images;

[0029] Based on the illegal parking environment image, the illegal parking placement of the illegal non-motor vehicle is identified to determine the direction of the illegal parking sign;

[0030] Based on the illegally parked environment image, motor vehicle recognition is performed to determine a plurality of motor vehicles in the environment;

[0031] A cooperative motor vehicle is selected from a plurality of surrounding motor vehicles according to the direction of the illegal parking sign.

[0032] As a further limitation of the technical solution of the embodiment of the present invention, generating and sending a collaborative detection application to the collaborative motor vehicle, and obtaining a collaboratively shot image after the collaborative detection application is approved specifically includes the following steps:

[0033] Generate collaborative testing applications;

[0034] sending the collaborative inspection application to the collaborative motor vehicle;

[0035] receiving feedback on the collaborative application of the collaborative motor vehicle;

[0036] If the collaborative application feedback is approved, collaboratively photographing to obtain a collaborative photographing image through the collaborative motor vehicle;

[0037] If the collaborative application is not approved, the collaborative motor vehicle is reselected and the collaborative inspection application is resent.

[0038] As a further limitation of the technical solution of the embodiment of the present invention, the license plate recognition of the illegally parked non-motor vehicle based on the collaboratively shot image, obtaining the illegal parking license plate information, matching the illegal parking main information, and issuing an illegal parking removal alarm specifically includes the following steps:

[0039] capturing a parking sign image from the collaboratively shot image;

[0040] Performing license plate recognition on the illegal parking plate image to obtain illegal parking plate information;

[0041] Matching the owner information of the illegal parking lot according to the illegal parking lot information;

[0042] Based on the illegally parked vehicle owner information, an illegal parking eviction warning is issued.

[0043] As a further limitation of the technical solution of the embodiment of the present invention, the step of issuing an illegal parking removal warning according to the illegal parking owner information specifically includes the following steps:

[0044] Identify the information of the owner of the illegally parked vehicle and obtain the communication address of the owner;

[0045] Generate illegal parking warning information;

[0046] The illegal parking warning information is sent to the vehicle owner's communication address to issue an illegal parking eviction warning.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The embodiment of the present invention detects and shoots through a detection camera, and based on cloud computing technology, determines illegally parked non-motor vehicles; performs license plate recognition to determine whether the recognition is successful; when the recognition is unsuccessful, performs illegal parking placement recognition and collaborative analysis, and selects a collaborative motor vehicle; obtains collaboratively shot images; performs license plate recognition on illegally parked non-motor vehicles, obtains illegal parking license plate information, and issues an illegal parking expulsion alarm. It is able to identify illegally parked non-motor vehicles based on cloud-edge collaboration, and when license plate recognition is unsuccessful, selects a collaborative motor vehicle, obtains collaboratively shot images, performs license plate recognition on illegally parked non-motor vehicles, and issues an illegal parking expulsion alarm, so that the technical defect that the license plates of some illegally parked non-motor vehicles cannot be clearly captured or fully recorded due to poor shooting angles is solved, and the comprehensiveness and accuracy of non-motor vehicle illegal parking detection is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flow chart of a non-motor vehicle illegal parking detection method based on cloud-edge collaboration provided by an embodiment of the present invention is shown;

[0050] Figure 2 A flow chart showing prohibited parking identification and timing comparison in the method provided by an embodiment of the present invention is shown;

[0051] Figure 3 A flowchart of determining a non-motor vehicle parking ban in a method provided by an embodiment of the present invention is shown;

[0052] Figure 4 A flowchart showing license plate recognition and judgment in the method provided by an embodiment of the present invention is shown;

[0053] Figure 5 A flow chart showing illegal parking identification and collaborative analysis in the method provided by an embodiment of the present invention is shown;

[0054] Figure 6 A flowchart of obtaining collaboratively shot images in a method provided in an embodiment of the present invention is shown;

[0055] Figure 7A flow chart of obtaining illegal parking license plate information in a method provided by an embodiment of the present invention is shown;

[0056] Figure 8 A flow chart of performing an illegal parking removal warning in the method provided in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] It is understandable that in the prior art, due to the fixed position, height and / or angle of non-motor vehicle illegal parking detection shooting and the relatively complex on-site environment, when illegally parked non-motor vehicles are detected, there is a technical defect that the license plates of some illegally parked non-motor vehicles cannot be clearly captured or fully recorded due to poor shooting angles, which limits the comprehensiveness and accuracy of non-motor vehicle illegal parking detection.

[0059] To solve the above problems, an embodiment of the present invention discloses a non-motor vehicle illegal parking detection method based on cloud-edge collaboration, which performs real-time detection and shooting through a detection camera, obtains and uploads real-time shooting data, and performs prohibited parking identification and timing comparison based on cloud computing technology to determine illegally parked non-motor vehicles; extracts multiple tracking and shooting images of illegally parked non-motor vehicles from the real-time shooting data, and performs license plate recognition to determine whether the recognition is successful; when the recognition is not successful, extracts the illegal parking environment image from the real-time shooting data, performs illegal parking placement identification and collaborative analysis, and selects a collaborative motor vehicle; generates and sends a collaborative detection application to the collaborative motor vehicle, and obtains the collaborative shooting image after the collaborative detection application is passed; based on the collaborative shooting image, performs license plate recognition on the illegally parked non-motor vehicle, obtains the illegal parking license plate information, matches the illegal parking main information, and issues an illegal parking removal alarm. Based on cloud-edge collaboration, it can identify illegally parked non-motor vehicles, and when license plate recognition is unsuccessful, it can select a collaborative motor vehicle to obtain collaboratively shot images, perform license plate recognition on illegally parked non-motor vehicles, and issue an illegal parking expulsion alarm. This solves the technical defect that the license plates of some illegally parked non-motor vehicles cannot be clearly captured or fully recorded due to poor shooting angles, thereby improving the comprehensiveness and accuracy of non-motor vehicle illegal parking detection.

[0060] Specifically, Figure 1 A flow chart of a non-motor vehicle illegal parking detection method based on cloud-edge collaboration provided by an embodiment of the present invention is shown.

[0061] In a preferred embodiment of the present invention, a non-motor vehicle illegal parking detection method based on cloud-edge collaboration is provided, and the method specifically comprises the following steps:

[0062] Step S101: Real-time detection and shooting are performed through the detection camera, and real-time shooting data is obtained and uploaded. Based on cloud computing technology, prohibited parking recognition and timing comparison are performed to determine illegally parked non-motor vehicles.

[0063] In an embodiment of the present invention, the detection camera is controlled to perform real-time detection and shooting, real-time shooting data is obtained, and the real-time shooting data is uploaded to the cloud according to a preset cloud address, and then based on cloud computing technology, according to the preset no-parking line marking, the real-time shooting data is regionally identified to determine the no-parking area, and based on cloud computing technology, according to the preset non-motor vehicle characteristics, the real-time shooting data is identified as non-motor vehicles entering the camera, multiple target non-motor vehicles are determined, and multiple target non-motor vehicles are tracked and identified, and the target non-motor vehicles entering the no-parking area are marked as no-parking non-motor vehicles, and then the no-parking time of the no-parking non-motor vehicle is calculated, and the no-parking time is compared with the preset no-parking threshold. When the no-parking time exceeds the preset no-parking threshold, the corresponding no-parking non-motor vehicle is marked as an illegally parked non-motor vehicle, thereby realizing the detection and identification of illegally parked non-motor vehicles based on cloud-edge collaboration.

[0064] It can be understood that in one embodiment of the present invention, the yellow grid no-parking line on the ground is identified by a detection camera to confirm the no-parking area, and the non-motor vehicle entering the screen is identified and tracked. Once the non-motor vehicle enters the no-parking area and the stay time reaches a preset threshold, it is determined to be an illegally parked non-motor vehicle, and the picture and related information of the illegally parked non-motor vehicle are uploaded to the cloud. While performing illegal parking identification, video image data is collected and cached in the camera storage medium after pre-processing such as cleaning and formatting. The data is uploaded to the cloud server regularly according to the preset threshold for deep learning training; the cloud automatically labels and extracts features of the received pre-processed data to generate a large-scale labeled data set, and uses the large-scale labeled data set to train, adjust parameters and optimize the model. The trained model is exported and sent to the front-end smart camera for real-time target recognition and reasoning; the cloud focuses on centralized training and optimization of the model, and with its abundant computing resources and large-scale data processing capabilities, the model can be continuously improved and perfected, while the edge is responsible for the reasoning work of the model and the collection and pre-processing of image data. This means that data processing and decision-making can be performed quickly at the front end close to the data source, greatly shortening the response time. This collaborative working mode greatly improves recognition efficiency and response speed, effectively avoids delays and losses caused by long-distance data transmission, and achieves fast and accurate processing. At the same time, by continuously collecting data and feedback, and regularly uploading new data to the cloud for model retraining and updating, it can sensitively adapt to new data changes and environmental dynamics.

[0065] Specifically, Figure 2The flowchart of the prohibited parking identification and timing comparison in the method provided by the embodiment of the present invention is shown.

[0066] Among them, in the preferred embodiment provided by the present invention, the real-time detection and shooting by the detection camera, obtaining and uploading the real-time shooting data, and performing the prohibited parking identification and timing comparison based on the cloud computing technology to determine the illegally parked non-motor vehicle specifically include the following steps:

[0067] Step S1011: Perform real-time detection and shooting through a detection camera to obtain real-time shooting data.

[0068] Step S1012: Upload the real-time shooting data according to the preset cloud address.

[0069] Step S1013: Based on cloud computing technology, the real-time shooting data is subjected to parking prohibition identification to determine the non-motor vehicles that are prohibited from parking.

[0070] Specifically, Figure 3 A flow chart of determining the prohibition of parking of non-motor vehicles in the method provided by an embodiment of the present invention is shown.

[0071] Among them, in the preferred embodiment provided by the present invention, the method of performing parking prohibition identification on the real-time shooting data based on cloud computing technology and determining the parking prohibition of non-motor vehicles specifically includes the following steps:

[0072] Step S10131: Based on cloud computing technology, perform area recognition on the real-time shooting data to determine the prohibited parking area.

[0073] Step S10132: Based on cloud computing technology, the real-time shooting data is used to identify non-motor vehicles entering the camera, and multiple target non-motor vehicles are determined.

[0074] Step S10133: Track and identify the plurality of target non-motor vehicles, and mark the target non-motor vehicles that enter the prohibited parking area as prohibited non-motor vehicles.

[0075] Furthermore, the real-time detection and shooting by the detection camera, obtaining and uploading the real-time shooting data, and performing the prohibited parking identification and timing comparison based on the cloud computing technology to determine the illegally parked non-motor vehicle also include the following steps:

[0076] Step S1014: Calculate the prohibited parking duration of the prohibited non-motor vehicle, and when the prohibited parking duration exceeds a preset prohibited parking threshold, mark the prohibited non-motor vehicle as an illegally parked non-motor vehicle.

[0077] Furthermore, the non-motor vehicle illegal parking detection method based on cloud-edge collaboration also includes the following steps:

[0078] Step S102: extract multiple tracking images of the illegally parked non-motor vehicle from the real-time shooting data, perform license plate recognition, and determine whether the recognition is successful.

[0079] In an embodiment of the present invention, partial shooting data of illegally parked non-motor vehicles are extracted from real-time shooting data, and then the partial shooting data are processed frame by frame to obtain multiple partial frame-by-frame images. Then, from the multiple partial frame-by-frame images, multiple tracking shooting images with clear pictures of illegally parked non-motor vehicles are screened, and license plate recognition is performed on the multiple tracking shooting images to determine whether the recognition is successful.

[0080] Specifically, Figure 4 A flow chart of license plate recognition and judgment in the method provided by an embodiment of the present invention is shown.

[0081] Among them, in the preferred embodiment provided by the present invention, extracting multiple tracking images of the illegally parked non-motor vehicle from the real-time shooting data, and performing license plate recognition, and judging whether the recognition is successful specifically include the following steps:

[0082] Step S1021: extracting partial shooting data of illegally parked non-motor vehicles from the real-time shooting data.

[0083] Step S1022: perform frame-by-frame processing on the partial shooting data to obtain a plurality of partial frame-by-frame images.

[0084] Step S1023: Filter a plurality of tracking images from the plurality of partial frame-by-frame images.

[0085] Step S1024: performing license plate recognition on the plurality of tracking and shooting images to determine whether the recognition is successful.

[0086] Furthermore, the non-motor vehicle illegal parking detection method based on cloud-edge collaboration also includes the following steps:

[0087] Step S103: When the recognition is unsuccessful, extract the illegal parking environment image from the real-time shooting data, perform illegal parking placement recognition and collaborative analysis, and select a collaborative motor vehicle.

[0088] In an embodiment of the present invention, when license plate recognition is unsuccessful, the highest-definition illegal parking environment image is selected from multiple tracking images, and the illegal parking placement is identified on the illegal parking environment image to determine the direction of the illegal parking license plate of the illegally parked non-motorized vehicle. The illegal parking environment image is also identified by performing motor vehicle identification to determine multiple environmental motor vehicles in the image screen, and according to the direction of the illegal parking license plate, an environmental motor vehicle at the position corresponding to the direction of the illegal parking license plate is selected from the multiple environmental motor vehicles and marked as a cooperative motor vehicle.

[0089] Specifically, Figure 5 A flow chart of illegal parking identification and collaborative analysis in the method provided by an embodiment of the present invention is shown.

[0090] Among them, in the preferred embodiment provided by the present invention, when the recognition is unsuccessful, extracting the illegal parking environment image from the real-time shooting data, performing illegal parking placement recognition and collaborative analysis, and selecting the collaborative motor vehicle specifically includes the following steps:

[0091] Step S1031: When the recognition is unsuccessful, an illegal parking environment image is selected from the plurality of tracking images.

[0092] Step S1032: Based on the illegal parking environment image, the illegal parking placement of the illegally parked non-motor vehicle is identified to determine the direction of the illegal parking sign.

[0093] Step S1033: Based on the illegally parked environment image, motor vehicle identification is performed to determine multiple motor vehicles in the environment.

[0094] Step S1034: Select a cooperative motor vehicle from multiple environmental motor vehicles based on the direction of the illegal parking sign.

[0095] Furthermore, the non-motor vehicle illegal parking detection method based on cloud-edge collaboration also includes the following steps:

[0096] Step S104: Generate and send a collaborative detection application to the collaborative motor vehicle, and obtain a collaborative shooting image after the collaborative detection application is approved.

[0097] In an embodiment of the present invention, a collaborative detection application is generated and sent to a collaborative motor vehicle. The owner of the collaborative motor vehicle can view the collaborative detection application through a mobile terminal, provide collaborative application feedback to approve or reject the application, and transmit the collaborative application feedback through the original channel through the collaborative motor vehicle. The owner of the collaborative motor vehicle can then receive the collaborative application feedback from the collaborative motor vehicle, identify the collaborative application feedback, and when the collaborative application feedback is approval, the collaborative application feedback is passed. At this time, collaborative shooting can be performed through the collaborative motor vehicle to obtain collaborative shooting images. When the collaborative application feedback is rejection, the collaborative application feedback is not passed. At this time, it is necessary to reselect the collaborative motor vehicle and resend the collaborative detection application.

[0098] Specifically, Figure 6 A flow chart of acquiring collaboratively shot images in the method provided in an embodiment of the present invention is shown.

[0099] Among them, in the preferred embodiment provided by the present invention, the generating and sending of the collaborative detection application to the collaborative motor vehicle, and after the collaborative detection application is approved, obtaining the collaborative shooting image specifically comprises the following steps:

[0100] Step S1041: Generate a collaborative testing application.

[0101] Step S1042: Send the collaborative detection application to the collaborative motor vehicle.

[0102] Step S1043, receiving the collaborative application feedback of the collaborative motor vehicle.

[0103] Step S1044: If the collaborative application feedback is passed, collaborative photography is performed by the collaborative motor vehicle to obtain a collaborative photography image.

[0104] Step S1045: If the collaborative application is not approved, reselect the collaborative motor vehicle and resend the collaborative inspection application.

[0105] Furthermore, the non-motor vehicle illegal parking detection method based on cloud-edge collaboration also includes the following steps:

[0106] Step S105: Based on the collaboratively shot image, the license plate of the illegally parked non-motor vehicle is recognized, the illegal parking license plate information is obtained, the illegal parking main information is matched, and an illegal parking removal alarm is issued.

[0107] In an embodiment of the present invention, an illegal parking plate image is captured from the collaboratively shot image, and then the license plate recognition is performed on the illegal parking plate image to obtain the illegal parking plate information, and then the illegal parking owner information is matched according to the illegal parking plate information. By recognizing the illegal parking owner information, the owner's communication address is obtained, and illegal parking alarm information is generated and sent to the owner's communication address, thereby implementing an illegal parking eviction alarm to the owner of the illegally parked non-motor vehicle.

[0108] Specifically, Figure 7 A flow chart of obtaining illegal parking license plate information in the method provided in an embodiment of the present invention is shown.

[0109] Among them, in the preferred embodiment provided by the present invention, the license plate recognition of the illegally parked non-motor vehicle is performed according to the collaboratively shot image, the illegal parking license plate information is obtained, the illegal parking main information is matched, and the illegal parking removal alarm is issued, which specifically includes the following steps:

[0110] Step S1051: capturing an image of a parking sign that is in violation of the regulations from the collaboratively shot image.

[0111] Step S1052: Perform license plate recognition on the illegal parking plate image to obtain illegal parking plate information.

[0112] Step S1053: Match the illegal parking owner information according to the illegal parking plate information.

[0113] Step S1054: issuing an illegal parking eviction warning based on the illegal parking owner information.

[0114] Specifically, Figure 8 A flow chart of performing an illegal parking removal warning in the method provided in an embodiment of the present invention is shown.

[0115] Among them, in the preferred implementation manner provided by the present invention, the step of issuing an illegal parking removal warning according to the illegal parking owner information specifically includes the following steps:

[0116] Step S10541, identifying the information of the illegally parked car owner and obtaining the communication address of the car owner.

[0117] Step S10542: Generate illegal parking warning information.

[0118] Step S10543: Send the illegal parking warning information to the vehicle owner's communication address to issue an illegal parking eviction warning.

[0119] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0120] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0121] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A non-motor vehicle illegal parking detection method based on cloud-edge collaboration, characterized in that: The method specifically comprises the following steps: Through real-time detection and shooting by detection cameras, real-time shooting data is obtained and uploaded. Based on cloud computing technology, prohibited parking identification and timing comparison are performed to determine the illegally parked non-motor vehicles; Extracting multiple tracking images of the illegally parked non-motor vehicle from the real-time shooting data, and performing license plate recognition to determine whether the recognition is successful; When the recognition is unsuccessful, extract the illegal parking environment image from the real-time shooting data, perform illegal parking placement recognition and collaborative analysis, and select the collaborative motor vehicle; generating and sending a collaborative inspection application to the collaborative motor vehicle, and acquiring a collaboratively shot image after the collaborative inspection application is approved; According to the collaboratively shot image, the license plate of the illegally parked non-motor vehicle is recognized, the illegal parking license plate information is obtained, the illegal parking main information is matched, and an illegal parking removal alarm is issued.

2. The non-motor vehicle illegal parking detection method based on cloud-edge collaboration according to claim 1 is characterized in that: The method of performing real-time detection and shooting by a detection camera, acquiring and uploading real-time shooting data, performing prohibited parking identification and timing comparison based on cloud computing technology, and determining illegally parked non-motor vehicles specifically includes the following steps: Perform real-time detection and shooting through the detection camera to obtain real-time shooting data; Uploading the real-time shooting data according to a preset cloud address; Based on cloud computing technology, the real-time shooting data is used to identify the prohibited parking and determine the prohibited parking of non-motor vehicles; The prohibited parking time of the prohibited non-motor vehicle is calculated, and when the prohibited parking time exceeds a preset prohibited parking threshold, the prohibited non-motor vehicle is marked as an illegally parked non-motor vehicle.

3. The non-motor vehicle illegal parking detection method based on cloud-edge collaboration according to claim 2 is characterized in that: The method of performing parking prohibition identification on real-time shooting data based on cloud computing technology and determining the parking prohibition of non-motor vehicles specifically includes the following steps: Based on cloud computing technology, the real-time shooting data is used to identify the area and determine the prohibited parking areas; Based on cloud computing technology, the real-time shooting data is used to identify non-motor vehicles entering the camera and determine multiple target non-motor vehicles; Tracking and identifying the plurality of target non-motor vehicles, and marking the target non-motor vehicles entering the prohibited parking area as prohibited parking non-motor vehicles.

4. The non-motor vehicle illegal parking detection method based on cloud-edge collaboration according to claim 1 is characterized in that: The step of extracting a plurality of tracking images of the illegally parked non-motor vehicle from the real-time shooting data, performing license plate recognition, and determining whether the recognition is successful specifically comprises the following steps: Extracting partial shooting data of illegally parked non-motor vehicles from the real-time shooting data; Processing the partial shooting data frame by frame to obtain a plurality of partial frame-by-frame images; Selecting a plurality of tracking images from the plurality of partial frame-by-frame images; Perform license plate recognition on the plurality of tracking and shooting images to determine whether the recognition is successful.

5. The non-motor vehicle illegal parking detection method based on cloud-edge collaboration according to claim 4 is characterized in that: When the recognition is unsuccessful, extracting the illegal parking environment image from the real-time shooting data, performing illegal parking placement recognition and collaborative analysis, and selecting the collaborative motor vehicle specifically includes the following steps: When the recognition is unsuccessful, selecting an illegal parking environment image from the plurality of tracking and shooting images; Based on the illegal parking environment image, the illegal parking placement of the illegal non-motor vehicle is identified to determine the direction of the illegal parking sign; Based on the illegally parked environment image, motor vehicle recognition is performed to determine a plurality of motor vehicles in the environment; A cooperative motor vehicle is selected from a plurality of surrounding motor vehicles according to the direction of the illegal parking sign.

6. The non-motor vehicle illegal parking detection method based on cloud-edge collaboration according to claim 1 is characterized in that: The generating and sending of a collaborative detection application to the collaborative motor vehicle, and obtaining a collaboratively shot image after the collaborative detection application is approved specifically comprises the following steps: Generate collaborative testing applications; sending the collaborative inspection application to the collaborative motor vehicle; receiving feedback on the collaborative application of the collaborative motor vehicle; If the collaborative application feedback is approved, collaboratively photographing to obtain a collaborative photographing image through the collaborative motor vehicle; If the collaborative application is not approved, the collaborative motor vehicle is reselected and the collaborative inspection application is resent.

7. The non-motor vehicle illegal parking detection method based on cloud-edge collaboration according to claim 1 is characterized in that: The method of performing license plate recognition on the illegally parked non-motor vehicle according to the collaboratively shot image, obtaining the illegally parked license plate information, matching the illegally parked vehicle main information, and issuing an illegally parked vehicle removal warning specifically includes the following steps: capturing a parking sign image from the collaboratively shot image; Performing license plate recognition on the illegal parking plate image to obtain illegal parking plate information; Matching the owner information of the illegal parking lot according to the illegal parking lot information; Based on the illegally parked vehicle owner information, an illegal parking eviction warning is issued.

8. The non-motor vehicle illegal parking detection method based on cloud-edge collaboration according to claim 7 is characterized in that: The step of issuing an illegal parking removal warning according to the illegal parking owner information specifically includes the following steps: Identify the information of the owner of the illegally parked vehicle and obtain the communication address of the owner; Generate illegal parking warning information; The illegal parking warning information is sent to the vehicle owner's communication address to issue an illegal parking eviction warning.

Images