Roadside parking space occupation identification method and device, computer equipment, storage medium and computer program product

Abstract

The invention relates to a roadside parking space occupation identification method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: receiving parking space planning data sent by a server; continuously collecting video streams of the target parking area through a connected roadside camera; a built-in computer vision algorithm is used to analyze the video stream in real time, and vehicle events are detected, and the vehicle events comprise at least one of a vehicle parking event and a vehicle driving-out event; when it is detected that a vehicle parking event exists in the target parking area, the current vehicle image and the parking space planning data are input into a built-in multi-mode large model, and a parking space number corresponding to the vehicle is output through fusion analysis; and forming event data by the parking space number, the current vehicle image and the timestamp, and uploading the event data to the server. The method can improve the recognition precision and reduce the cost.

Description

Technical Field

[0001] This application relates to the field of roadside parking management technology, and in particular to a method, device, computer equipment, storage medium and computer program product for identifying roadside parking space occupancy. Background Technology

[0002] With the rapid development of intelligent transportation system technology, visual perception and edge computing technologies are becoming increasingly mature. Leveraging their core characteristics of real-time data processing and intelligent analysis, they are gradually being widely applied in urban traffic management scenarios, providing technical support for the efficiency and precision of roadside parking management. Against this backdrop, roadside parking space occupancy status recognition is a core component of parking management, and traditional methods primarily rely on geomagnetic sensor detection and manual inspection.

[0003] In traditional technology, geomagnetic sensors detect changes in the magnetic field of the parking space area to determine whether an object is occupying it, and then report the occupancy status signal to the cloud management system to achieve basic parking space status monitoring; manual inspection relies on management personnel to verify the use of parking spaces on-site, record vehicle parking information, and handle abnormal situations.

[0004] However, these traditional methods have significant shortcomings: geomagnetic sensors can only detect whether an object occupies a parking space, and cannot distinguish between motor vehicles, non-motor vehicles, and other obstacles, resulting in a high false alarm rate; in roadside parking systems with deployed cameras, adding geomagnetic sensors increases the cost of hardware deployment, maintenance, and data fusion, increasing system complexity; moreover, geomagnetic sensors need to be buried in the road surface of the parking space, which not only damages the aesthetics of the road surface, but is also difficult to replace after the equipment is damaged; manual inspection is inefficient, costly, and prone to omissions, making it difficult to meet the needs of large-scale urban roadside parking management. Summary of the Invention

[0005] Therefore, it is necessary to provide a roadside parking space occupancy recognition method, device, computer equipment, computer-readable storage medium, and computer program product that can improve recognition accuracy and reduce costs, in order to address the above-mentioned technical problems.

[0006] In a first aspect, this application provides a method for identifying the occupancy of roadside parking spaces, applied to an edge computing unit, the method comprising:

[0007] Receive parking space planning data sent by the server;

[0008] The system continuously collects video streams of the target parking area via connected roadside cameras.

[0009] The video stream is analyzed in real time using built-in computer vision algorithms to detect vehicle events, including at least one of vehicle parking events and vehicle exit events.

[0010] When a vehicle is detected to have entered the target parking area, the current vehicle image and the parking space planning data are input into the built-in multimodal large model, and the parking space number corresponding to the vehicle is output after fusion analysis.

[0011] The event data, consisting of the parking space number, the current vehicle image, and the timestamp, is uploaded to the server.

[0012] In some embodiments of the method, the parking space planning data includes at least one of the following: parking space boundary frame coordinate information generated based on the monitoring range of the roadside camera, parking space planning image information, and parking space number.

[0013] In some embodiments of the method, the parking space planning data includes a parking space planning JSON file, and receiving the parking space planning data sent by the server includes:

[0014] The server receives a parking space planning JSON file sent by the server. The parking space planning JSON file is encapsulated by the server based on the structured information of the parking space boundary coordinates, the parking space planning image information, and the parking space number.

[0015] In some embodiments of the method, the step of using a built-in computer vision algorithm to perform real-time analysis of the video stream and detect vehicle events includes:

[0016] In response to detecting that the outline of a vehicle falls completely within the bounding box of any parking space in the target parking area and the dwell time exceeds a first preset threshold, a vehicle parking event is determined.

[0017] In response to the detection that the vehicle outline has completely left the parking space boundary box and the time of departure exceeds the second preset threshold, it is determined as a vehicle leaving the parking space event.

[0018] In some embodiments of the method, when a vehicle parking event is detected in the target parking area, the current vehicle image and the parking space planning data are input into a built-in multimodal large model, and the parking space number corresponding to the vehicle is output after fusion analysis, including:

[0019] Extract vehicle contour features and vehicle position coordinates from the current vehicle image, match them with the bounding box coordinates of the corresponding parking space in the parking space planning data, and calculate the overlap between the vehicle position and the parking space boundary.

[0020] If the overlap exceeds a preset ratio, output the unique parking space number of that parking space.

[0021] In some embodiments of the method, the method further includes:

[0022] If the confidence level of the output result of the initial fusion analysis is lower than the preset confidence threshold, vehicle images from adjacent frames are obtained and re-input into the model for secondary analysis until the parking space number that meets the confidence requirement is output.

[0023] According to a second aspect of the present disclosure, a roadside parking space occupancy recognition device is provided. The device includes:

[0024] The communication module is used to receive parking space planning data sent by the server.

[0025] The acquisition module is used to continuously acquire video streams of the target parking area through connected roadside cameras;

[0026] The detection module is used to perform real-time analysis of the video stream using built-in computer vision algorithms to detect vehicle events, including at least one of vehicle parking events and vehicle exit events.

[0027] The output module is used to input the current vehicle image and the parking space planning data into the built-in multimodal large model when a vehicle parking event is detected in the target parking area, and output the parking space number corresponding to the vehicle after fusion analysis.

[0028] The communication module is also used to upload event data composed of the parking space number, the current vehicle image, and the timestamp to the server.

[0029] According to a third aspect of the present disclosure, a computer device is provided. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0030] Receive parking space planning data sent by the server;

[0031] The system continuously collects video streams of the target parking area via connected roadside cameras.

[0032] The video stream is analyzed in real time using built-in computer vision algorithms to detect vehicle events, including at least one of vehicle parking events and vehicle exit events.

[0033] When a vehicle is detected to have entered the target parking area, the current vehicle image and the parking space planning data are input into the built-in multimodal large model, and the parking space number corresponding to the vehicle is output after fusion analysis.

[0034] The event data, consisting of the parking space number, the current vehicle image, and the timestamp, is uploaded to the server.

[0035] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0036] Receive parking space planning data sent by the server;

[0037] The system continuously collects video streams of the target parking area via connected roadside cameras.

[0038] The video stream is analyzed in real time using built-in computer vision algorithms to detect vehicle events, including at least one of vehicle parking events and vehicle exit events.

[0039] When a vehicle is detected to have entered the target parking area, the current vehicle image and the parking space planning data are input into the built-in multimodal large model, and the parking space number corresponding to the vehicle is output after fusion analysis.

[0040] The event data, consisting of the parking space number, the current vehicle image, and the timestamp, is uploaded to the server.

[0041] According to a fifth aspect of the present disclosure, a computer program product is provided. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0042] Receive parking space planning data sent by the server;

[0043] The system continuously collects video streams of the target parking area via connected roadside cameras.

[0044] The video stream is analyzed in real time using built-in computer vision algorithms to detect vehicle events, including at least one of vehicle parking events and vehicle exit events.

[0045] When a vehicle is detected to have entered the target parking area, the current vehicle image and the parking space planning data are input into the built-in multimodal large model, and the parking space number corresponding to the vehicle is output after fusion analysis.

[0046] The event data, consisting of the parking space number, the current vehicle image, and the timestamp, is uploaded to the server.

[0047] The roadside parking space occupancy recognition scheme provided in this application embodiment achieves full automation of the parking space occupancy recognition process through the collaborative work of the server and edge computing unit, combined with camera acquisition and multimodal large model analysis. It can complete vehicle event detection and parking space number recognition without manual intervention, which greatly improves the efficiency of roadside parking management and avoids the problems of easy omission and low efficiency of traditional manual inspection.

[0048] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0049] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0050] Figure 1 This is a flowchart illustrating a roadside parking space occupancy identification method according to an exemplary embodiment;

[0051] Figure 2 This is a schematic diagram illustrating the interaction between a server and an edge computing unit according to an exemplary embodiment;

[0052] Figure 3 This is a structural block diagram of a roadside parking space occupancy recognition device according to an exemplary embodiment;

[0053] Figure 4 This is a diagram illustrating the internal structure of a computer device according to an exemplary embodiment. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure. The terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitations, the presence of other identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded. For example, the use of terms such as "first," "second," etc., to denote names does not indicate any specific order.

[0056] In some embodiments provided in this disclosure, the execution of the roadside parking space occupancy identification method can be controlled by a unified controller or by multiple controllers. These controllers may include controllers of local terminals or controllers of remote servers. In some embodiments, the controllers of local terminals and the controllers of servers may jointly assist in completing the roadside parking space occupancy identification and control processing. The local terminal mentioned in this disclosure may include, but is not limited to, various robotic devices, vehicle-mounted devices, personal computers, laptops, smartphones, tablets, wearable devices, medical devices, VR (Virtual Reality) devices, etc. The server may also be a server, server cluster, distributed subsystem, cloud processing platform, server containing blockchain nodes, and combinations thereof. The controllers described in this disclosure may include various control units capable of implementing logic processing functions, including but not limited to CPU (Central Processing Unit), PLC (Programmable Logic Controller), ECU (Electronic Control Unit), MCU (Microcontroller Unit), FPGA (Field Programmable Gate Array), and CPLD (Complex Programmable Logic Device), as well as controllers composed of one or more logic function units, chips, etc.

[0057] In some embodiments of this disclosure, a roadside parking space occupancy identification method is provided, applied to an edge computing unit, such as... Figure 1 As shown, it includes the following steps:

[0058] S20: Receive parking space planning data sent by the server.

[0059] The server-side can include a cloud-based parking space planning and management platform, capable of parking space information planning, data storage, command issuance, and event data reception and statistics. It serves as the core management node of the entire recognition system, responsible for coordinating the edge computing units and the overall parking management operations. Edge computing units typically refer to embedded computing devices deployed on the roadside, possessing data acquisition, real-time computing, inference, and network communication capabilities. They can directly connect to roadside cameras to acquire video stream data and run computer vision algorithms and multimodal large models to complete local data processing.

[0060] Parking space planning data typically refers to a set of structured information generated by the server. It may include the coordinate information of the parking space boundary boxes drawn based on the actual monitoring range of the roadside cameras, the planning image information that intuitively shows the distribution of parking spaces, and the unique number assigned to each parking space. This information can be packaged into a parking space planning JSON file for transmission and storage.

[0061] S22. Continuously collect video streams of the target parking area through connected roadside cameras.

[0062] Roadside cameras are specialized imaging devices deployed on both sides of roads to collect visual data of roadside parking areas. The target parking area refers to a specific set of parking spaces that have been uniformly planned and delineated by the server in a roadside parking management scenario.

[0063] S24. Analyze the video stream in real time using the built-in computer vision algorithm to detect vehicle events, including at least one of vehicle parking events and vehicle exit events.

[0064] Computer vision algorithms are the core technology used in roadside parking space occupancy recognition systems to process visual data. They can simulate human visual perception and analyze and understand video streams or image data collected by roadside cameras. A vehicle entering an area refers to a specific event where a vehicle changes from a moving state to a stable parked state within the target parking area, meeting preset judgment conditions. A vehicle leaving an area refers to a specific event where a vehicle that was originally parked stably changes from an occupied state to completely leaving the parking space within the target parking area, meeting preset judgment conditions.

[0065] S26. When a vehicle parking event is detected in the target parking area, the current vehicle image and the parking space planning data are input into the built-in multimodal large model, and the parking space number corresponding to the vehicle is output after fusion analysis.

[0066] The current vehicle image refers to a single or multiple static images captured by the edge computing unit from the real-time video stream when a vehicle parking event is detected. These images contain the complete shape of the vehicle and its parking position. This is the visual data used by the multimodal large-scale model for parking space number recognition. The multimodal large-scale model is an artificial intelligence model capable of processing and fusing multiple types of data. It can be used to integrate visual data (current vehicle image) with structured data (parking space planning data) to achieve accurate matching between vehicles and parking spaces. Through pre-training, this model possesses image understanding and data fusion capabilities.

[0067] S28. The parking space number, the current vehicle image, and the timestamp are combined to form event data and uploaded to the server.

[0068] The parking space number is a unique identifier assigned by the server to each individual parking space within the target parking area, used to distinguish different parking spaces and achieve refined management. The timestamp refers to the specific identifier recorded by the edge computing unit indicating the time of a vehicle event (vehicle entering or leaving).

[0069] In some embodiments of this disclosure, through the collaborative work of the server and edge computing unit, combined with camera acquisition and multimodal large model analysis, the entire process of parking space occupancy recognition is automated. Vehicle event detection and parking space number recognition can be completed without manual intervention, which greatly improves the efficiency of roadside parking management and avoids the problems of easy omissions and low efficiency of traditional manual inspections.

[0070] The following is combined Figure 2 Further explanation is needed.

[0071] In some embodiments of this disclosure, the parking space planning data includes at least one of the following: parking space boundary frame coordinate information generated based on the monitoring range of the roadside camera, parking space planning image information, and parking space number.

[0072] In some embodiments of this disclosure, the parking space planning data includes a parking space planning JSON file. S20 includes: receiving a parking space planning JSON file sent by a server. The parking space planning JSON file is obtained by the server encapsulating structured information based on parking space boundary coordinate information, parking space planning image information, and parking space number.

[0073] In some implementations, the server first draws an irregular quadrilateral bounding box consisting of four vertices for each parking space based on the monitoring coverage of the roadside cameras, records the coordinate information of each vertex, generates a parking space planning image, and assigns a unique number to each parking space. The server then encapsulates this structured information, including bounding box coordinates, planning image, and parking space number, into a parking space planning JSON file, and sends it to the corresponding roadside edge computing unit via network communication. The edge computing unit receives this file through its built-in communication module and stores it for later use.

[0074] In some embodiments of this disclosure, it can be ensured that the edge computing unit obtains sufficient structured data for accurate identification, providing comprehensive support for the fusion analysis of multimodal large models, effectively improving the accuracy of parking space matching, and reducing identification errors caused by incomplete data; JSON files can be used to encapsulate parking space planning data, facilitating data transmission and parsing between the server and the edge computing unit, improving the efficiency and compatibility of data interaction, while the structured encapsulation method also reduces the complexity of data storage and retrieval, reducing the burden of system data processing.

[0075] In some embodiments of this disclosure, S24 includes:

[0076] In response to detecting that the outline of a vehicle falls completely within the bounding box of any parking space in the target parking area and the dwell time exceeds a first preset threshold, a vehicle parking event is determined.

[0077] In response to the detection that the vehicle outline has completely left the parking space boundary box and the time of departure exceeds the second preset threshold, it is determined as a vehicle leaving the parking space event.

[0078] In some implementations, the edge computing unit establishes a stable connection with a roadside low-position camera via wired or wireless means, controlling the camera to continuously capture the target parking area, forming real-time video stream data. The camera installation position is pre-calibrated to ensure that the shooting range completely covers all planned parking spaces and can clearly capture the dynamic process of vehicles entering and leaving parking spaces. The video stream data is transmitted in real time to the storage module of the edge computing unit for subsequent analysis. The edge computing unit calls the built-in computer vision algorithm to perform frame-by-frame real-time analysis of the received video stream. The algorithm compares the vehicle outline features in the image with the coordinate information of the parking space boundary box: when a vehicle outline is detected to be completely within the boundary box of a parking space, and this state continues to exceed a first preset threshold, it is determined as a vehicle parking event; when a vehicle outline originally parked in the parking space is detected to be completely outside the boundary box, and this state continues to exceed a second preset threshold, it is determined as a vehicle leaving event. The first preset threshold and the second preset threshold can be the same or different; the first preset threshold and the second preset threshold can be preset with a standard value, or they can be configured by the user or operator.

[0079] In some embodiments of this disclosure, vehicle events are determined by setting contour matching conditions and time thresholds, which effectively avoids false event triggering caused by temporary vehicle parking, misidentification of contours, etc., improves the accuracy of vehicle event detection, reduces invalid identification and data reporting, and ensures the stability of system operation.

[0080] In some embodiments of this disclosure, S26 includes:

[0081] Extract vehicle contour features and vehicle position coordinates from the current vehicle image, match them with the bounding box coordinates of the corresponding parking space in the parking space planning data, and calculate the overlap between the vehicle position and the parking space boundary.

[0082] If the overlap exceeds a preset ratio, output the unique parking space number of that parking space.

[0083] In some implementations, upon detecting a vehicle parking event, the edge computing unit extracts an image containing the vehicle from the current video frame, extracts the vehicle's outline features and vehicle position coordinates from the image, and inputs them along with pre-stored parking space planning data into a built-in multimodal large model. The model accurately matches the vehicle's position coordinates with the bounding box coordinates of each parking space, calculates the overlap between the vehicle's position and the parking space boundary, and outputs a unique number corresponding to the parking space when the overlap degree meets preset requirements.

[0084] In some embodiments of this disclosure, by extracting vehicle features and accurately matching them with parking space coordinates, and combining the overlap to determine the output parking space number, the accuracy of parking space identification is further improved, avoiding confusion between different parking spaces and solving the false alarm problem caused by the inability of traditional geomagnetic sensors to distinguish the type of occupied object.

[0085] In some embodiments of this disclosure, the method further includes:

[0086] If the confidence level of the output result of the initial fusion analysis is lower than the preset confidence threshold, vehicle images from adjacent frames are obtained and re-input into the model for secondary analysis until the parking space number that meets the confidence requirement is output.

[0087] In some implementations, if the confidence level of the initial analysis result does not meet the preset standard, the system will automatically acquire vehicle images from adjacent frames and re-input them into the model for secondary analysis until a parking space number that meets the confidence level requirement is output. The edge computing unit integrates the identified parking space number, the captured current vehicle image, and the timestamp of the event to form complete event data, which is then uploaded to the server via the communication module. After receiving the data, the server stores it in the database for subsequent parking billing, parking space status display, and other management operations.

[0088] In some embodiments of this disclosure, secondary verification is performed on analysis results with insufficient confidence. The reliability of the output results is ensured by re-analyzing multiple frames of images, which effectively reduces recognition errors caused by factors such as blurry or occluded images in a single instance, and further improves the stability and recognition accuracy of the entire recognition system.

[0089] The roadside parking space occupancy identification methods disclosed herein achieve full automation of the parking space occupancy identification process through the collaborative work of the server and edge computing unit, combined with camera acquisition and multimodal large model analysis. Vehicle event detection and parking space number identification can be completed without manual intervention, which greatly improves the efficiency of roadside parking management and avoids the problems of easy omission and low efficiency of traditional manual inspection.

[0090] It is understood that the various embodiments of the methods described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. Related details can be found in the descriptions of other method embodiments.

[0091] It should be understood that although the steps in the flowcharts shown in the accompanying drawings are displayed sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the accompanying drawings may include multiple steps or stages, which are not necessarily completed at the same time, but may be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but may be performed alternately or in turn with other steps or at least a portion of the steps or stages of other steps.

[0092] Based on the description of the roadside parking space occupancy identification method embodiments described above, this disclosure also provides a roadside parking space occupancy identification device for implementing the roadside parking space occupancy identification method mentioned above. The device may include a system (including a distributed system), software (application), module, component, controller, server, terminal, etc., using the method described in the embodiments of this specification, combined with necessary implementation hardware. Based on the same innovative concept, the devices in one or more embodiments provided by the embodiments of this disclosure are as described in the following embodiments. Since the implementation schemes and methods for solving the problem by the devices are similar, the implementation of specific devices in the embodiments of this specification can refer to the implementation of the aforementioned method, and repeated details will not be repeated. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0093] Figure 3 This is a schematic block diagram illustrating a roadside parking space occupancy recognition device according to an exemplary embodiment. The device can be the aforementioned terminal, a server, or a module, component, device, control unit, etc., integrated into the terminal. For details, please refer to... Figure 3The device 100 may include: a communication module 120, a data acquisition module 140, a detection module 160, and an output module 180. The communication module 120 receives parking space planning data sent by the server; the data acquisition module 140 continuously acquires video streams of the target parking area via connected roadside cameras; the detection module 160 uses a built-in computer vision algorithm to perform real-time analysis of the video stream and detect vehicle events, including at least one of vehicle entry events and vehicle exit events; the output module 180, when a vehicle entry event is detected in the target parking area, inputs the current vehicle image and the parking space planning data into a built-in multimodal large model, performs fusion analysis, and outputs the parking space number corresponding to the vehicle; the communication module 120 is also used to upload event data composed of the parking space number, the current vehicle image, and a timestamp to the server.

[0094] In some embodiments of the device, the parking space planning data includes at least one of the following: parking space boundary frame coordinate information generated based on the monitoring range of the roadside camera, parking space planning image information, and parking space number.

[0095] In some embodiments of the device, the communication module 120 is further configured to receive a parking space planning JSON file sent by the server. The parking space planning JSON file is encapsulated by the server based on structured information such as parking space boundary coordinates, parking space planning image information, and parking space number.

[0096] In some embodiments of the device, the detection module 160 is further configured to determine a vehicle parking event in response to detecting that the vehicle outline completely falls within the boundary frame of any parking space in the target parking area and the dwell time exceeds a first preset threshold; and to determine a vehicle exit event in response to detecting that the vehicle outline completely leaves the boundary frame of the parking space and the departure time exceeds a second preset threshold.

[0097] In some embodiments of the device, the output module 180 is further configured to extract vehicle contour features and vehicle body position coordinates from the current vehicle image, match them with the bounding box coordinates of the corresponding parking space in the parking space planning data, and calculate the overlap between the vehicle position and the parking space boundary; it is also configured to output a unique parking space number when the overlap exceeds a preset ratio.

[0098] In some embodiments of the device, the output module 180 is further configured to acquire vehicle images from adjacent frames and re-input them into the model for secondary analysis if the confidence level of the first fusion analysis output result is lower than a preset confidence threshold, until a parking space number that meets the confidence level requirement is output.

[0099] Each module in the aforementioned roadside parking space occupancy recognition device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0100] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a roadside parking space occupancy identification method.

[0101] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0102] Based on the foregoing description of the relevant methods and apparatus embodiments, this disclosure also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the roadside parking space occupancy identification method described in any embodiment of this specification.

[0103] Based on the foregoing description of the relevant methods and apparatus embodiments, this disclosure also provides a computer-readable storage medium that, when the instructions in the computer-readable storage medium are executed by the processor of a computer device, enables the computer device to implement the roadside parking space occupancy identification method as described in any embodiment of this disclosure.

[0104] Based on the foregoing description of the relevant methods and apparatus embodiments, this disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the roadside parking space occupancy identification method described in any embodiment of this specification.

[0105] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. In particular, hardware + program embodiments are relatively simple in description because they are fundamentally similar to method embodiments; relevant parts can be referred to the descriptions in the method embodiments.

[0106] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0107] It should be noted that the apparatus, computer equipment, storage medium, and computer program products described above may also include other implementation methods according to the description of the method embodiments. Specific implementation methods can be found in the description of the relevant method embodiments. Furthermore, new embodiments formed by combinations of features from various methods, apparatuses, devices, and server embodiments still fall within the scope of this disclosure and will not be elaborated upon here.

[0108] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, when implementing one or more of these specifications, the functions of each module can be implemented in the same or different software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling and communication connections between the devices or units shown or described can be implemented through direct and / or indirect coupling / connection, through standard or custom interfaces or protocols, and can be implemented electrically, mechanically, or in other forms.

[0109] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0110] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for identifying the occupancy of roadside parking spaces, characterized in that, Applied to edge computing units, the method includes: Receive parking space planning data sent by the server; The system continuously collects video streams of the target parking area via connected roadside cameras. The video stream is analyzed in real time using built-in computer vision algorithms to detect vehicle events, including at least one of vehicle parking events and vehicle exit events. When a vehicle is detected to have entered the target parking area, the current vehicle image and the parking space planning data are input into the built-in multimodal large model, and the parking space number corresponding to the vehicle is output after fusion analysis. The event data, consisting of the parking space number, the current vehicle image, and the timestamp, is uploaded to the server.

2. The method according to claim 1, characterized in that, The parking space planning data includes at least one of the following: parking space boundary frame coordinates generated based on the monitoring range of roadside cameras, parking space planning image information, and parking space number.

3. The method according to claim 2, characterized in that, The parking space planning data includes a parking space planning JSON file, and the receipt of the parking space planning data sent by the server includes: The server receives a parking space planning JSON file sent by the server. The parking space planning JSON file is obtained by the server encapsulating structured information such as parking space boundary coordinates, parking space planning image information, and parking space number.

4. The method according to claim 1, characterized in that, The method of using built-in computer vision algorithms to perform real-time analysis of the video stream and detect vehicle events includes: In response to detecting that the outline of a vehicle falls completely within the bounding box of any parking space in the target parking area and the dwell time exceeds a first preset threshold, a vehicle parking event is determined. In response to the detection that the vehicle outline has completely left the parking space boundary box and the time of departure exceeds the second preset threshold, it is determined as a vehicle leaving the parking space event.

5. The method according to claim 1, characterized in that, When a vehicle is detected parking in the target parking area, the current vehicle image and the parking space planning data are input into a built-in multimodal large model. After fusion analysis, the corresponding parking space number for the vehicle is output, including: Extract vehicle contour features and vehicle position coordinates from the current vehicle image, match them with the bounding box coordinates of the corresponding parking space in the parking space planning data, and calculate the overlap between the vehicle position and the parking space boundary. If the overlap exceeds a preset ratio, output the unique parking space number of that parking space.

6. The method according to claim 5, characterized in that, The method further includes: If the confidence level of the output result of the initial fusion analysis is lower than the preset confidence threshold, vehicle images from adjacent frames are obtained and re-input into the model for secondary analysis until the parking space number that meets the confidence requirement is output.

7. A roadside parking space occupancy recognition device, characterized in that, The device includes: The communication module is used to receive parking space planning data sent by the server. The acquisition module is used to continuously acquire video streams of the target parking area through connected roadside cameras; The detection module is used to perform real-time analysis of the video stream using built-in computer vision algorithms to detect vehicle events, including at least one of vehicle parking events and vehicle exit events. The output module is used to input the current vehicle image and the parking space planning data into the built-in multimodal large model when a vehicle parking event is detected in the target parking area, and output the parking space number corresponding to the vehicle after fusion analysis. The communication module is also used to upload event data composed of the parking space number, the current vehicle image, and the timestamp to the server.

8. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, It stores a computer program thereon, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1 to 6.

Images