Image acquisition method and device, computer equipment and storage medium

Abstract

The invention relates to an image acquisition method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a target parking area picture; inputting the target parking area picture and a preset division instruction into a preset parking space model to obtain a sub-area of each parking space in the target area and a corresponding parking space code; receiving an image shot by a second camera and a target instruction; wherein the second camera is used for carrying out direction adjustment on the second camera, and shooting a target area; the target instruction comprises a target parking space code; inputting an image shot by the camera and the target instruction into the parking space model, and outputting to obtain a direction adjustment parameter of a second camera; and adjusting the second camera based on the orientation adjustment parameter, and shooting to obtain a target image. By adopting the method, the intelligent level of parking lot management can be effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, in particular to an image acquisition method and device, computer equipment, a storage medium and a computer program product. BACKGROUND

[0002] With the progress of intelligent parking management technology, visual-based parking space positioning and vehicle recognition technology is gradually applied to modern parking lots. This technology has the characteristics of fast response, non-contact and adaptation to complex environments, significantly improving the automation level of parking management and user car search efficiency. Traditional parking management systems usually rely on fixed-position cameras for area monitoring, and then manually check the video or use simple image recognition to determine the parking space status. When users search for a car, they need to repeatedly compare multiple pictures or rely on memory to find the car, which is inefficient. Although some systems have introduced parking sensors or static recognition algorithms, they still lack dynamic image acquisition capabilities for specific parking spaces. However, the wide-angle camera used in existing methods has limited coverage and cannot clearly capture distant license plate numbers. SUMMARY

[0003] Therefore, it is necessary to provide an image acquisition method, device, computer equipment, computer readable storage medium and computer program product to solve the above technical problems.

[0004] In a first aspect, the present application provides an image acquisition method. The method comprises:

[0005] Obtaining a target parking area picture; wherein the target parking area picture is obtained by a first camera device;

[0006] Inputting the target parking area picture and a preset division instruction into a preset parking space model to obtain sub-regions of each parking space in the target region and corresponding parking space codes;

[0007] Receiving an image captured by a second camera and a target instruction; wherein the second camera is used to capture the target region by adjusting the orientation of the second camera; and the target instruction includes a target parking space code;

[0008] Inputting the image captured by the camera and the target instruction into the parking space model to output an orientation adjustment parameter of the second camera;

[0009] Based on the orientation adjustment parameter, the second camera is adjusted to capture a target image; wherein the target image includes identification information of a vehicle corresponding to the target parking space code.

[0010] In one of the embodiments, the target parking area picture and the preset division instruction are input into a preset parking space model to obtain a sub-area of each parking space in the target area and a corresponding parking space code, including:

[0011] Based on the preset division instruction, the target parking area picture is subjected to semantic segmentation through the preset parking space model to obtain boundary information;

[0012] The boundary information is subjected to graphic fitting to obtain a sub-area of each parking space;

[0013] Each parking space sub-area is numbered to obtain a corresponding parking space code.

[0014] In one of the embodiments, the image captured by the camera and the target instruction are input into the parking space model to output a position adjustment parameter of the second camera, including:

[0015] According to the parking space code, a first position coordinate of the corresponding target parking space sub-area in the target parking area picture is obtained from the parking space model;

[0016] A second position coordinate of the target parking space sub-area in the image currently captured by the second camera is recognized;

[0017] Based on the deviation of the first position coordinate and the second position coordinate, the position adjustment parameter is determined.

[0018] In one of the embodiments, the method further includes:

[0019] The license plate area of the target image is positioned to obtain a license plate image;

[0020] The license plate image is recognized to obtain identification information of a vehicle corresponding to the target parking space code;

[0021] An association between the target parking space code and the corresponding vehicle identification information is established, and the association is stored in a target database.

[0022] In one of the embodiments, the image captured by the camera and the target instruction are input into the parking space model to output a position adjustment parameter of the second camera, including:

[0023] The image captured by the second camera, the target parking area picture, and the target parking space code are input into the parking space model to obtain an initial adjustment parameter;

[0024] According to the initial adjustment parameter, the second camera is adjusted, and a target image and identification information of a corresponding vehicle are obtained;

[0025] In a case where the identification information is not recognized, inputting the target image into the parking space model to obtain a target adjustment parameter.

[0026] In one of the embodiments, the inputting the target parking area picture and the preset division instruction into the preset parking space model to obtain the sub-area of each parking space in the target area and the corresponding parking space code comprises:

[0027] receiving a target instruction;

[0028] In a case where the target instruction is received, inputting the target parking area picture and the target instruction into the preset parking space model to obtain the sub-area of each parking space corresponding to the target instruction and the corresponding parking space code;

[0029] In a case where the target instruction is not received, inputting the target parking area picture and the preset division instruction into the preset parking space model to obtain the sub-area of each parking space in the target area and the corresponding parking space code.

[0030] In a second aspect, the present application further provides an image acquisition device. The device comprises:

[0031] a data acquisition module configured to acquire a target parking area picture, wherein the target parking area picture is obtained by a first camera device;

[0032] a region determination module configured to input the target parking area picture and a preset division instruction into a preset parking space model to obtain a sub-area of each parking space in a target area and a corresponding parking space code;

[0033] an instruction receiving module configured to receive an image captured by a second camera and a target instruction, wherein the second camera is configured to capture the target area by adjusting an orientation of the second camera, and the target instruction comprises a target parking space code;

[0034] a parameter acquisition module configured to input the image captured by the camera and the target instruction into the parking space model to output an orientation adjustment parameter of the second camera;

[0035] an image acquisition module configured to adjust the second camera based on the orientation adjustment parameter to capture a target image, wherein the target image comprises identification information of a vehicle corresponding to the target parking space code.

[0036] In a third aspect, the present application further provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the image acquisition method according to any one of the embodiments of the present application when executing the computer program.

[0037] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the image acquisition method as described in any one of the embodiments of this disclosure.

[0038] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the image acquisition method as described in any of the embodiments of this disclosure.

[0039] The aforementioned image acquisition method, apparatus, computer equipment, storage medium, and computer program products, by combining the functions of a first camera device and a second camera device, enable the system to capture multi-level information from panoramic images to local details. This design not only improves the accuracy of parking space division and vehicle recognition but also significantly enhances the system's adaptability, allowing it to operate stably in parking lot scenarios of varying sizes and complexities. Furthermore, by optimizing the entire process of image acquisition, processing, and analysis, this method effectively reduces the need for manual intervention, thereby lowering operating costs and improving management efficiency. Attached Figure Description

[0040] Figure 1 This is an application environment diagram of an image acquisition method in one embodiment;

[0041] Figure 2 This is a flowchart illustrating an image acquisition method in one embodiment;

[0042] Figure 3 This is a structural block diagram of an image acquisition device in one embodiment;

[0043] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0044] 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.

[0045] The image acquisition method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed in the cloud or on other network servers. In this application environment, a first camera device (such as a wide-angle camera deployed at a high point in the parking lot) can be used to capture panoramic images of the target parking area and upload them to server 104. Server 104 can run a preset parking space model, automatically dividing and labeling each parking space sub-area and its code based on the panoramic images. When a user submits a query request containing the target parking space code through terminal 102 (such as a smartphone), server 104 can instruct a second camera device (such as a directional adjustable PTZ camera deployed inside the parking lot) to capture images, calculate the optimal orientation adjustment parameters using the parking space model, and finally control the second camera device to point at the target parking space, capturing a clear image containing vehicle identification information. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle systems. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0046] In one embodiment, such as Figure 2 As shown, an image acquisition method is provided, including the following steps:

[0047] Step S200: Obtain an image of the target parking area; wherein the image of the target parking area is obtained through a first camera device.

[0048] The first camera device can be a wide-angle camera fixedly installed high up in the parking lot to capture panoramic images of the entire target parking area. This type of device can have high resolution and a wide field of view, ensuring coverage of all parking spaces, and transmits the captured images to the server in real time for subsequent processing. The target parking area image may include the layout information of all parking spaces in the parking lot, as well as the possible parking status of vehicles. After receiving the image, the server can use it as the basis for subsequent processing, such as parking space division and code generation.

[0049] In one exemplary embodiment, the first camera device can also automatically adjust exposure parameters according to ambient lighting conditions to ensure clear images are acquired under different lighting conditions. Furthermore, the device can support timed image capture, periodically acquiring images of the target parking area at preset time intervals to achieve dynamic monitoring of parking space status. After receiving these images, the server can preprocess them, such as removing noise and enhancing contrast, to provide high-quality data support for subsequent parking space model analysis. Simultaneously, the server can also record the timestamp information of image acquisition for subsequent data traceability and analysis.

[0050] Step S202: Input the target parking area image and the preset division instructions into the preset parking space model to obtain the sub-areas of each parking space in the target area and the corresponding parking space codes.

[0051] The parking space model can include a deep learning-based algorithm capable of intelligently analyzing input images of target parking areas. This model can automatically identify the boundaries of each parking space within the target parking area using pre-defined partitioning instructions and image feature information, generating corresponding sub-region partitioning results. Each sub-region can represent an independent parking space, and each parking space is assigned a unique parking space code for subsequent location and management operations. In practical applications, the parking space model can employ convolutional neural networks (CNNs) or semantic segmentation techniques to perform pixel-level processing on the target parking area image, thereby accurately extracting the boundary information of the parking spaces. By performing graphical fitting on these boundary information, the model can further optimize the partitioning accuracy of parking space sub-regions, ensuring that the scope of each parking space is clear and non-overlapping. Furthermore, the generation process of parking space codes can follow certain rules, such as numbering according to the arrangement order of parking spaces, or naming them according to their specific location in the parking lot.

[0052] In one exemplary embodiment, the parking space model can also dynamically adjust its partitioning strategy by combining historical data and real-time image information. For example, when the parking lot layout changes, the model can regenerate parking space sub-regions and codes based on the latest image data without manual intervention. This intelligent processing method not only improves the efficiency of parking space partitioning but also provides reliable foundational support for subsequent vehicle recognition and positioning.

[0053] Step S204: Receive an image captured by the second camera and a target instruction; wherein the second camera is used to capture an image of the target area by adjusting the orientation of the second camera; the target instruction includes a target parking space code.

[0054] The second camera can be an adjustable camera device with pan-tilt functionality, capable of flexibly adjusting the shooting angle and focal length according to instructions. The target instruction, typically submitted by the user via a terminal device, includes a specific target parking space code, instructing the system to locate the designated parking space. Upon receiving the target instruction, the server parses the parking space code and calculates the optimal orientation adjustment parameters for the second camera based on a pre-set parking space model. These parameters may include horizontal rotation angle, vertical tilt angle, and focal length adjustment value, ensuring the camera accurately aligns with the target parking space and captures a clear image. Furthermore, the second camera can automatically optimize shooting parameters based on ambient lighting conditions, such as adjusting exposure time or white balance settings, thereby improving image quality. After completing the orientation adjustment, the second camera will re-capture the target area and output image data containing the target parking space and its related information for subsequent processing.

[0055] In one exemplary embodiment, the second camera may include a spherical camera with 360-degree rotation and a wide range of tilt adjustment capabilities, enabling it to cover a wider monitoring area. The spherical camera may be equipped with a high-magnification optical zoom lens, capable of clearly capturing detailed information about the target parking space from a distance. Furthermore, this type of camera may support autofocus, quickly adjusting the focus based on the location of the target parking space to ensure the clarity and accuracy of the captured image. In practical applications, the flexibility and high performance of spherical cameras make them an ideal choice for capturing dynamic images inside parking lots.

[0056] In one exemplary embodiment, the second camera can also integrate an edge computing module. This module can perform some image processing tasks locally on the camera, such as preliminary target detection and parking space location, thereby reducing data transmission latency and lowering the computational burden on the server. In this way, the system can generate orientation adjustment parameters more quickly and achieve accurate shooting of the target parking space. Furthermore, in some complex scenarios, such as when there are obstructions or uneven lighting conditions in a parking lot, the second camera can also incorporate multi-frame image fusion technology to improve the shooting effect. Specifically, the camera can continuously capture multiple images under different exposure parameters and synthesize these images into a high-quality output result through an algorithm. This method effectively solves image quality problems caused by environmental factors, ensuring that the final acquired image meets the needs of subsequent recognition and analysis.

[0057] Step S206: Input the image captured by the camera and the target command into the parking space model, and output the orientation adjustment parameters of the second camera.

[0058] The parking space model, upon receiving images and target instructions from the second camera, can perform deep analysis based on the input data. Specifically, the model can extract features from the target area in the image, identifying key information related to the target parking space's code. Combining the specific requirements of the target instructions, the model calculates the deviation between the second camera's current position and the target parking space. This process may involve precise coordinate positioning of the target parking space and geometric analysis of the camera's current field of view. Through comprehensive processing of this data, the parking space model can generate a set of precise orientation adjustment parameters. These parameters include not only horizontal and vertical angle adjustments but may also encompass optimization suggestions for shooting parameters such as focal length and aperture size, ensuring that the final captured image clearly presents the target parking space and its related information.

[0059] In one exemplary embodiment, in certain complex scenarios, such as parking lots with multi-layered structures or densely packed parking spaces, the parking space model can also incorporate spatial mapping technology. By constructing a three-dimensional spatial model of the target parking area, the system can more accurately determine the specific location of the target parking space and optimize the adjustment strategy of the second camera accordingly. This method can effectively avoid shooting deviations caused by viewpoint occlusion or distance estimation errors, further improving the system's positioning accuracy and adaptability. Simultaneously, to improve processing efficiency, the parking space model can also utilize incremental learning technology to continuously optimize its algorithm performance based on the results of each shot, thereby achieving faster and more accurate orientation adjustments in subsequent operations.

[0060] Step S208: Based on the orientation adjustment parameters, adjust the second camera to capture a target image; wherein the target image includes the identification information of the vehicle corresponding to the target parking space code.

[0061] After adjusting the orientation of the second camera, it can accurately target the parking space and capture a clear image. This image not only contains detailed information about the parking space but also accurately captures the identification information of the parked vehicle, such as license plate number, body color, or special markings. By extracting and analyzing this information, the system can further complete vehicle identification and status confirmation. In this process, the parking space model can combine various image processing technologies, such as edge detection, optical character recognition (OCR), and deep learning algorithms, to ensure high-precision target data extraction from complex backgrounds. Furthermore, to improve recognition accuracy, the system can also utilize comparative analysis of multiple frames to eliminate misjudgments caused by light reflection, shadows, or other interference factors.

[0062] In one exemplary embodiment, the process of acquiring the target image may further include dynamic adaptation to the shooting environment. For example, when there is direct sunlight or low illumination in the parking lot, the second camera can automatically adjust exposure parameters or enable infrared fill light to ensure that the image quality is always optimal. Simultaneously, the system can flexibly select different shooting modes based on the specific location of the target parking space and the characteristics of the surrounding environment. For example, when shooting at a distance, the camera can switch to a high-magnification zoom mode; while when shooting at close range, it prioritizes using a wide-angle mode to cover more details. This intelligent shooting strategy not only improves the efficiency of image acquisition but also provides a reliable foundation for subsequent data processing.

[0063] In one exemplary embodiment, the target image can be transmitted to a server for further analysis and storage. By combining it with a preset parking space model and user query requests, the system can quickly generate feedback information and send it to the user's terminal device. This process not only automates parking space location and vehicle recognition but also significantly improves the user's car-finding experience. For example, users can view real-time images and vehicle information of the target parking space through a smartphone application, thereby quickly finding their vehicle's location. Simultaneously, the system can also record historical data from each query, providing valuable operational references for parking lot management, such as statistics on parking space usage frequency and peak-hour distribution. This vision-based intelligent parking management system offers a novel solution for the efficient operation and user experience optimization of modern parking lots.

[0064] In the image acquisition method described above, by combining the functions of the first and second camera devices, the system can achieve multi-level information capture, from panoramic images to local details. This design not only improves the accuracy of parking space division and vehicle recognition but also significantly enhances the system's adaptability, enabling it to operate stably in parking lot scenarios of varying sizes and complexities. Furthermore, by optimizing the entire process of image acquisition, processing, and analysis, this method can effectively reduce the need for manual intervention, thereby lowering operating costs and improving management efficiency.

[0065] In one embodiment, the step of inputting the target parking area image and a preset division instruction into a preset parking space model to obtain sub-regions of each parking space in the target area and their corresponding parking space codes includes:

[0066] Based on preset segmentation instructions, the target parking area image is semantically segmented using a preset parking space model to identify boundary information.

[0067] The boundary information is graphically fitted to obtain the sub-regions of each parking space.

[0068] Number the sub-areas of each parking space to obtain the corresponding parking space number.

[0069] In the semantic segmentation process, the parking space model can accurately distinguish the boundaries of different parking spaces within the target parking area. Through the classification of image pixels, the system can identify the specific range of each parking space and generate clear boundary information. This boundary information, after image fitting, further optimizes the division accuracy of parking space sub-regions, ensuring that the range of each parking space is non-overlapping and conforms to the actual layout. In the numbering stage, the system uniquely identifies each parking space sub-region according to preset rules, such as generating codes based on the parking space arrangement order or geographical location. This process not only improves the automation level of parking space division but also lays a data foundation for subsequent vehicle positioning and management. Furthermore, by combining real-time acquired image information, the system can dynamically adjust the division strategy to adapt to changes in the parking lot layout, thereby achieving more efficient parking space management.

[0070] In one exemplary embodiment, the parking space model can also utilize incremental learning techniques to continuously optimize its algorithm performance based on each partitioning result. For example, when the parking lot layout changes, the system can automatically detect new boundary information and incorporate it into the model training data, thereby improving the accuracy of subsequent partitioning. Simultaneously, the system can record historical data from each adjustment and refer to this information when regenerating parking space codes to avoid duplicate or confusing identifiers. Furthermore, the application of semantic segmentation technology is not limited to static image processing but can be extended to dynamic scene analysis. For example, when temporary obstacles or construction areas exist within the parking lot, the system can identify these changes through real-time updated image data and temporarily mark the relevant areas as unavailable to ensure the practicality of the parking space partitioning results. This intelligent processing method significantly improves the system's flexibility and adaptability, enabling it to cope with various complex parking scenarios.

[0071] In this embodiment, by combining semantic segmentation and incremental learning techniques, the parking space model can achieve dynamic monitoring and precise division of target parking areas. This method not only improves the accuracy of parking space division but also enhances the system's adaptability to complex scenarios.

[0072] In one embodiment, inputting the image captured by the camera and the target command into the parking space model, and outputting the orientation adjustment parameters of the second camera, includes:

[0073] Based on the parking space code, the first position coordinates of the corresponding target parking space sub-region in the target parking area image are obtained from the parking space model.

[0074] Identify the second position coordinates of a sub-region of the target parking space in the image currently captured by the second camera.

[0075] The orientation adjustment parameters are determined based on the deviation between the first position coordinates and the second position coordinates.

[0076] The parking space model can accurately calculate the deviation of the target parking space in the currently captured image by comparing the coordinates of the first and second positions. This process may involve geometric positioning analysis of the target parking space sub-region. Combining the camera's field of view and focal length parameters, the system can deduce the specific angle and focal length change that the second camera needs to adjust. For example, when the target parking space is outside the current shooting range, the system will prioritize calculating the adjustment values ​​of the horizontal rotation angle and the vertical tilt angle to ensure that the target parking space can be brought back into the shooting field of view. At the same time, if the target parking space is blurry or too small in the image, the system will further optimize the focal length parameters to ensure that the detailed information of the target parking space is clearly presented.

[0077] In one exemplary embodiment, the parking space model can also incorporate a multi-level calibration mechanism. The system performs a coarse adjustment based on initially calculated deviation values ​​to quickly include the target parking space within the shooting range. A secondary calibration of the target parking space's position can be performed using high-precision image analysis technology to ensure that the final captured image meets recognition requirements. Furthermore, this mechanism can dynamically adjust the calibration strategy according to changes in environmental conditions. For example, in low-light environments, the system may prioritize adjusting exposure parameters before performing orientation calibration to avoid misjudging the target parking space due to insufficient light.

[0078] In one exemplary embodiment, the parking space model can also incorporate a real-time feedback mechanism to further improve the accuracy of orientation adjustments. Specifically, the system re-captures an image after each adjustment and inputs it into the parking space model for deviation verification. If the verification result shows that the target parking space is still not perfectly aligned, the system generates a new set of adjustment parameters based on the new deviation value and repeats the above process until the target parking space is accurately captured. This closed-loop control design not only improves the system's robustness but also significantly shortens the adjustment time in complex scenarios, providing more reliable support for subsequent vehicle recognition and positioning.

[0079] In this embodiment, by calculating the deviation between the first position coordinates and the second position coordinates, the system can accurately derive the orientation adjustment parameters of the second camera.

[0080] In one embodiment, the method further includes:

[0081] The license plate area of ​​the target image is located to obtain the license plate image.

[0082] The license plate image is identified to obtain the vehicle identification information corresponding to the target parking space code.

[0083] Establish the association between the target parking space code and the corresponding vehicle identification information, and store the association in the target database.

[0084] In the license plate region localization process, image segmentation technology can be used to extract the specific area containing the license plate from the target image. By analyzing the pixel features of this area, the license plate can be effectively distinguished from background information, thus ensuring the accuracy of localization. After localization, the system preprocesses the extracted license plate image, such as adjusting contrast, removing noise, or correcting tilt angles, to improve the reliability of subsequent recognition. This series of operations lays the foundation for the accurate extraction of vehicle identification information.

[0085] In one exemplary embodiment, during the license plate image recognition stage, the system can combine optical character recognition (OCR) technology and a deep learning model to parse the characters on the license plate. Through training with a multi-layer neural network, the model can recognize license plate information with different fonts, colors, and formats, and adapt to various complex shooting conditions. Furthermore, the system can incorporate contextual information to assist in judgment, such as verifying the rationality of the recognition results according to license plate numbering rules, thereby reducing the possibility of misidentification. During the process of establishing associations, the system can bind the target parking space code with the recognized vehicle identification information and generate a unique mapping record. These records are stored in the target database for subsequent querying and management. The system can also set up data verification mechanisms, such as repeatedly verifying the matching between license plate information and parking space codes to eliminate potential erroneous records. Simultaneously, the database supports efficient retrieval functions, allowing users to quickly locate relevant records using parking space codes or vehicle identification information, thus meeting the diverse needs of parking lot management.

[0086] In one exemplary embodiment, the system can also introduce a timestamp mechanism to add time-related information to each associated record. This design not only facilitates tracking vehicle parking history but also provides data support for parking lot operation analysis. For example, by analyzing vehicle parking duration and frequency, managers can optimize parking space resource allocation or develop dynamic charging strategies. Furthermore, the system supports integration with other smart devices, such as interfacing with automatic gate systems at parking lot entrances, enabling seamless vehicle entry and exit management and further improving overall operational efficiency.

[0087] In this embodiment, by accurately locating the license plate region, the system can extract a clear license plate image from the target image. This process typically relies on advanced image segmentation techniques, combined with edge detection and morphological processing, to ensure accurate separation of the license plate region even against complex backgrounds.

[0088] In one embodiment, inputting the image captured by the camera and the target command into the parking space model, and outputting the orientation adjustment parameters of the second camera, includes:

[0089] The image captured by the second camera, the image of the target parking area, and the target parking space code are input into the parking space model to obtain initial adjustment parameters.

[0090] The second camera is adjusted according to the initial adjustment parameters, and the target image and the corresponding vehicle identification information are acquired.

[0091] If the identification information is not recognized, the target image is input into the parking space model to obtain the target adjustment parameters.

[0092] In cases where no identification information is detected, the system can automatically trigger the calculation process for target adjustment parameters. Specifically, the parking space model performs depth analysis on the target image, combining the target parking area image and the target parking space code to re-evaluate the shooting range and focal length settings of the second camera. This process may involve secondary verification of the boundary information of the target parking space sub-region to ensure the accuracy of the adjustment direction. Simultaneously, the system prioritizes factors that may lead to recognition failure, such as shooting angle deviations, lighting interference, or the special parking status of the target vehicle, and optimizes the adjustment strategy accordingly.

[0093] In one exemplary embodiment, the generation of target adjustment parameters can incorporate a multi-dimensional environmental perception mechanism. For example, when an obstruction is detected around the target parking space, the system dynamically adjusts the camera's height or tilt angle to avoid the obstruction and capture a clearer target image. Furthermore, if low-light conditions cause image quality degradation, the system can improve the shooting effect by enhancing supplementary lighting or switching to infrared imaging mode. This intelligent adjustment method not only improves the success rate of target image acquisition but also provides more reliable data support for subsequent license plate recognition. Further, the system can continuously improve the accuracy of the target adjustment parameters through iterative optimization. After each adjustment, the system re-captures an image and inputs it into the parking space model for verification. If the verification result shows that the target parking space is still not fully captured or the license plate information is not successfully extracted, the system generates a new round of adjustment parameters based on the new deviation value and repeats the above process. This closed-loop control design significantly improves the system's robustness, enabling it to maintain efficient positioning capabilities even in complex scenarios.

[0094] In this embodiment, by repeatedly adjusting and verifying the target image, the system can gradually approach the optimal shooting state, thereby ensuring the accurate association between the target parking space code and the corresponding vehicle identification information. This method not only enhances the system's adaptability but also provides a more stable technical guarantee for automated operations in parking lot management.

[0095] In one embodiment, the step of inputting the target parking area image and a preset division instruction into a preset parking space model to obtain sub-regions of each parking space in the target area and their corresponding parking space codes includes:

[0096] Receive target instructions.

[0097] Upon receiving a target instruction, the target parking area image and the target instruction are input into a preset parking space model to obtain the sub-area of ​​each parking space corresponding to the target instruction and the corresponding parking space code.

[0098] If no target instruction is received, the target parking area image and the preset division instruction are input into the preset parking space model to obtain the sub-areas of each parking space in the target area and the corresponding parking space code.

[0099] The system can flexibly adapt to different operational needs by parsing received instructions. When a target instruction is received, the parking space model can prioritize processing the target parking area image related to that instruction, ensuring the relevance and accuracy of the output results. If no target instruction is received, the system can automatically execute parking space allocation tasks based on preset allocation instructions, ensuring the continuity and stability of parking lot management. This dual-mode design not only enhances the system's intelligence level but also provides users with a more convenient operating experience. Furthermore, by dynamically switching between target instructions and preset instructions, the system can achieve efficient task allocation in different scenarios, thereby optimizing overall operational efficiency.

[0100] In one exemplary embodiment, the system can also introduce an instruction priority mechanism to dynamically adjust the processing order based on the importance and urgency of the target instruction. For example, during peak parking hours, the system can prioritize responses to instructions related to parking space allocation or vehicle guidance to ensure a smooth parking process. During off-peak hours, the system can optimize the parking lot layout based on preset instruction divisions to improve resource utilization. Furthermore, the system supports the analysis of historical instruction data, further optimizing the instruction parsing algorithm by mining high-frequency instruction types and operation patterns, thereby improving response speed and accuracy.

[0101] In one exemplary embodiment, the parking space model can incorporate multi-source data fusion technology to comprehensively analyze images of the target parking area combined with information collected from other sensors. For example, by fusing data from ultrasonic sensors or lidar, the system can more accurately identify the actual occupancy status of parking spaces, avoiding misjudgments caused by lighting conditions or image occlusion. Simultaneously, this multi-source data fusion approach enhances the system's adaptability to complex environments, enabling it to maintain stable performance even in adverse weather conditions or special scenarios.

[0102] In this embodiment, by optimizing the instruction parsing mechanism and integrating multi-source data, the system can achieve efficient task execution and accurate parking space management in different scenarios, providing more intelligent technical support for parking lot operation.

[0103] It should be understood that although the steps in the flowcharts of the embodiments described above are shown 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 steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0104] Based on the same inventive concept, this application also provides an image acquisition apparatus for implementing the image acquisition method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more image acquisition apparatus embodiments provided below can be found in the limitations of the image acquisition method described above, and will not be repeated here.

[0105] In one embodiment, such as Figure 3 As shown, an image acquisition device 300 is provided, including: a data acquisition module 301, a region determination module 303, an instruction receiving module 305, a parameter acquisition module 307, and an image acquisition module 309, wherein:

[0106] A data acquisition module is used to acquire images of a target parking area; wherein, the images of the target parking area are obtained through a first camera device;

[0107] The area determination module is used to input the target parking area image and the preset division instructions into the preset parking space model to obtain the sub-areas of each parking space in the target area and the corresponding parking space codes.

[0108] The instruction receiving module is used to receive images captured by the second camera and target instructions; wherein, the second camera is used to capture images of the target area by adjusting the orientation of the second camera; the target instructions include the target parking space code;

[0109] The parameter acquisition module is used to input the image captured by the camera and the target command into the parking space model, and output the orientation adjustment parameters of the second camera.

[0110] The image acquisition module is used to adjust the second camera based on the orientation adjustment parameters to capture a target image; wherein the target image includes the identification information of the vehicle corresponding to the target parking space code.

[0111] Each module in the aforementioned image acquisition 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 operations corresponding to each module.

[0112] 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, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. 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 stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. 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 the computer program is executed by the processor, it implements an image acquisition method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0113] Those skilled in the art will understand thatFigure 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.

[0114] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0115] 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. When executed, the computer program 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.

[0116] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0117] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An image acquisition method, characterized in that, The method includes: Acquire an image of the target parking area; wherein the image of the target parking area is obtained through a first camera device; The target parking area image and the preset division instructions are input into the preset parking space model to obtain the sub-areas of each parking space in the target area and the corresponding parking space codes. The system receives images captured by a second camera and target instructions; wherein the second camera is used to capture images of the target area by adjusting its orientation; the target instructions include the target parking space code. The image captured by the camera and the target command are input into the parking space model, and the orientation adjustment parameters of the second camera are output. Based on the orientation adjustment parameters, the second camera is adjusted to capture a target image; wherein, the target image includes the identification information of the vehicle corresponding to the target parking space code.

2. The method according to claim 1, characterized in that, The step of inputting the target parking area image and preset division instructions into a preset parking space model to obtain sub-regions of each parking space in the target area and their corresponding parking space codes includes: Based on preset division instructions, the target parking area image is semantically segmented using a preset parking space model to identify boundary information; The boundary information is graphically fitted to obtain sub-regions for each parking space; Number the sub-areas of each parking space to obtain the corresponding parking space number.

3. The method according to claim 1, characterized in that, The step of inputting the image captured by the camera and the target command into the parking space model, and outputting the orientation adjustment parameters of the second camera, includes: Based on the parking space code, obtain the first position coordinates of the corresponding target parking space sub-region in the target parking area image from the parking space model; Identify the second position coordinates of a sub-region of the target parking space in the image currently captured by the second camera; The orientation adjustment parameters are determined based on the deviation between the first position coordinates and the second position coordinates.

4. The method according to claim 1, characterized in that, The method further includes: The license plate region of the target image is located to obtain the license plate image; The license plate image is identified to obtain the vehicle identification information corresponding to the target parking space code; Establish the association between the target parking space code and the corresponding vehicle identification information, and store the association in the target database.

5. The method according to claim 4, characterized in that, The step of inputting the image captured by the camera and the target command into the parking space model, and outputting the orientation adjustment parameters of the second camera, includes: The image captured by the second camera, the image of the target parking area, and the target parking space code are input into the parking space model to obtain initial adjustment parameters; Based on the initial adjustment parameters, the second camera is adjusted, and the target image and the corresponding vehicle identification information are acquired; If the identification information is not recognized, the target image is input into the parking space model to obtain the target adjustment parameters.

6. The method according to claim 1, characterized in that, The step of inputting the target parking area image and preset division instructions into a preset parking space model to obtain sub-regions of each parking space in the target area and their corresponding parking space codes includes: Receive target instructions; Upon receiving a target instruction, the target parking area image and the target instruction are input into a preset parking space model to obtain the sub-area of ​​each parking space corresponding to the target instruction and the corresponding parking space code. If no target instruction is received, the target parking area image and the preset division instruction are input into the preset parking space model to obtain the sub-areas of each parking space in the target area and the corresponding parking space code.

7. An image acquisition device, characterized in that, The device includes: A data acquisition module is used to acquire images of a target parking area; wherein, the images of the target parking area are obtained through a first camera device; The area determination module is used to input the target parking area image and the preset division instructions into the preset parking space model to obtain the sub-areas of each parking space in the target area and the corresponding parking space codes. The instruction receiving module is used to receive images captured by the second camera and target instructions; wherein, the second camera is used to capture images of the target area by adjusting the orientation of the second camera; the target instructions include the target parking space code; The parameter acquisition module is used to input the image captured by the camera and the target command into the parking space model, and output the orientation adjustment parameters of the second camera. The image acquisition module is used to adjust the second camera based on the orientation adjustment parameters to capture a target image; wherein the target image includes the identification information of the vehicle corresponding to the target parking space code.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

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

Images