System and method for identifying parking space position based on intelligent charging pile

Abstract

The invention discloses a system and method for recognizing a parking space position based on an intelligent charging pile, and relates to the technical field of parking space recognition, and the method comprises the steps: starting the intelligent charging pile, collecting an original environment data set in a parking lot through a multi-sensor fusion technology, and carrying out the preprocessing of the collected original environment data set, and obtaining an optimized environment data set; applying a computer vision algorithm and a geometric analysis method to the image data in the optimized environment data set, extracting static element features and confirming the position and the shape of a fixed structure in the static element features to obtain a static element layer; extracting dynamic elements from the video data in the optimized environment data set by adopting a motion detection algorithm, and mapping the position information of the dynamic elements into a space coordinate system to obtain a dynamic element layer; and respectively extracting position and shape features of a static structure and a moving object from the static and dynamic element layers through a geometric feature extraction method.

Description

Technical Field

[0001] The present invention relates to the technical field of parking space recognition, and in particular to a system and method for recognizing parking space positions based on a smart charging pile. Background Art

[0002] The parking space recognition system is an intelligent system that uses a variety of technical means to automatically detect and identify the locations of vacant parking spaces in a parking lot. It integrates advanced technologies such as sensors, computer vision, and machine learning to monitor the distribution of vehicles in the parking lot in real time and provide users with accurate parking space information.

[0003] In the field of parking space recognition, existing systems cannot adapt well to different types of parking lot environments, and their performance degrades under severe weather conditions. In addition, many parking lot management systems can only provide static parking space information, while ignoring the impact of vehicle movement and other dynamic factors. At the same time, existing systems rarely provide detailed three-dimensional models of parking lots, which limits users' understanding of the parking environment. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a system and method for identifying parking space locations based on smart charging piles to solve the problems that existing systems cannot adapt well to different types of parking lot environments, their performance degrades under severe weather conditions, and many parking lot management systems can only provide static parking space information while ignoring the impact of vehicle movement and other dynamic factors.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for identifying parking space positions based on a smart charging pile, which comprises: starting the smart charging pile, using multi-sensor fusion technology to collect an original environmental data set in the parking lot, and preprocessing the collected original environmental data set to obtain an optimized environmental data set; applying computer vision algorithms and geometric analysis methods to image data in the optimized environmental data set, extracting static element features and confirming the position and shape of fixed structures in the static element features to obtain a static element layer; extracting dynamic elements from video data in the optimized environmental data set by using a motion detection algorithm, mapping the position information of the dynamic elements to a spatial coordinate system to obtain a dynamic element layer; extracting dynamic elements from static and The position and shape features of static structures and moving objects are extracted from the dynamic element layer respectively, and the relative positions of the position and shape features of static structures and moving objects are determined using the feature matching method. The relative positions of the position and shape features of static structures and moving objects are integrated using the 3D modeling platform Blender to obtain a 3D model of the parking lot. The multi-attribute decision analysis method is used to extract the attribute indicators that affect the parking space score in the 3D model of the parking lot, and weights are assigned to each attribute indicator that affects the parking space score. The weighted summation method is used to calculate the comprehensive score of each parking space. Based on the comprehensive score of each parking space, the parking space with the highest comprehensive score is selected as the charging parking space. When the vehicle arrives at the optimal parking space, the parking space status is verified and the smart charging pile is activated for charging service.

[0008] As a preferred solution of the method for identifying parking space positions based on a smart charging pile described in the present invention, wherein: the smart charging pile is started, a multi-sensor fusion technology is used to collect an original environmental data set in the parking lot, and the collected original environmental data set is pre-processed to obtain an optimized environmental data set. The specific steps are: starting the smart charging pile and initializing the multiple sensors installed on the smart charging pile; after starting the smart charging pile, the smart charging pile checks the health status of the multiple sensors through the Self-Diagnosis Program; the smart charging pile controls the multiple sensors to collect the environment in the parking lot to obtain an original environmental data set; an anomaly detection method is used to identify the original environmental data set and eliminate obviously erroneous data points, and a Kalman filter is used to remove high-frequency noise to obtain a preliminary optimized data set; the numerical data in the preliminary optimized data set is subjected to Z-score normalization processing to obtain a standardized optimized data set; the network time protocol is used to synchronize the timestamps of the multiple sensors, and the spatial coordinate system of the multiple sensors is adjusted through the external parameter information preset by the smart sensor to make it consistent with the global coordinate system to obtain an optimized environmental data set.

[0009] As a preferred solution of the method for identifying parking space positions based on smart charging piles described in the present invention, wherein: the image data in the optimized environment data set is applied with computer vision algorithms and geometric analysis methods to extract static element features and confirm the position and shape of fixed structures in the static element features to obtain a static element layer. The specific steps are: using the Canny edge detection algorithm to identify the boundary contour of the image data in the optimized environment data set, and using the Harris corner detection algorithm to find the significant corner points in the image data in the optimized environment data set. The expression is as follows:

[0010]

[0011] R(p)=det(M(p))-k·trac e 2 (M(p));

[0012] Among them, E(p) is the boundary contour strength of point p, is the gradient operator of the Gaussian filter, I(p) is the image grayscale value of point p, M(p) is the structure tensor matrix, det(M(p)) is the determinant of the matrix M(p), trace e (M(p)) is the trace of the matrix M(p), k is an empirical constant, and R(p) represents the corner point response value; the boundary contour strength E(p) and the corner point response value R(p) are combined to obtain the feature set F 1 ; Based on the feature set, use the findContours function to outline the contours of the fixed structure in the feature set, and use SIFT to compare the feature set with the contours in the optimized environment dataset. The expression is as follows:

[0013] C={p|p∈F 1 .E,isClosed(p)};

[0014]

[0015] Among them, F 1 .E is from F 1 The boundary contour strength E(p) extracted from isClosed(p) is used to determine whether point p forms a closed contour. i is the i-th contour, T i is the i-th template, γ is the adjustment coefficient, d(C i ,T i ) is the matching error between the ith contour and the ith template, S(C i ) represents the shape matching score of the i-th contour; the shape matching score S(C i ) to obtain the intermediate feature set F 2 ; The feature set F1 and the intermediate feature set F 2 The shape matching score S(C i ) to obtain the static element feature F 3 ; Calculate the static element feature F 3 The spatial relationship score M(C i ), the expression is:

[0016]

[0017] in, is the adjustment coefficient, M(C i ) is the spatial relationship score of the i-th contour; the comprehensive feature scoring function F(E) is combined with the static element feature F 3 , construct a static element layer, the expression is:

[0018]

[0019] L={F 1 ,M(C i ),Δ};

[0020] Among them, Δ is the feature score of the static element, n is the number of static elements, α i is the shape matching score weight of the i-th static element, F 2 .S(i) is the shape matching score of the i-th static element in the intermediate feature set, β i is the spatial relationship score weight of the i-th static element, F 3 .M(C i ) is the spatial relationship score of the ith static element in the static element feature, m is the number of computing resources used in the feature extraction process, W j is the weight of the j-th computing resource, C j is the consumption of the j-th computing resource, and L is the static element layer.

[0021] As a preferred solution of the method for identifying parking space positions based on smart charging piles described in the present invention, wherein: the motion detection algorithm is used to extract dynamic elements from the video data in the optimized environmental data set, and the position information of the dynamic elements is mapped to the spatial coordinate system to obtain a dynamic element layer. The specific steps are: selecting the optical flow method, calculating the pixel displacement between consecutive frames in the video data, and obtaining the velocity vector v(x, y) of the pixel point; using the obtained velocity vector v(x, y), using a threshold function to distinguish the static background and dynamic foreground in the velocity vector v(x, y), the expression is:

[0022]

[0023] Where D(x,y) is the binary dynamic element mask, x is the horizontal coordinate of the pixel, y is the vertical coordinate of the pixel, ∥v(x,y)∥ is the length of the velocity vector v(x,y) at the position (x,y), is the threshold of the velocity vector length; when the velocity vector length of v(x,y) exceeds the set threshold The position is considered to belong to the dynamic foreground, otherwise it is considered to be the static background, and the position information of the dynamic foreground is obtained and marked as a dynamic element; the dynamic element is mapped to the global coordinate system to obtain the dynamic element layer Q.

[0024] As a preferred solution of the method for identifying parking space positions based on smart charging piles described in the present invention, wherein: the position shape features of static structures and moving objects are extracted from static and dynamic element layers respectively by using a geometric feature extraction method, and the specific steps are: applying an edge detection algorithm and a corner point detection algorithm to extract the contours and key points of static structures and moving objects in the static element layer and the dynamic element layer respectively; using feature descriptors to encode the contours and key points of static structures and moving objects, and obtaining a static structure feature set L static and the moving object feature set Q dstatic , the expression is:

[0025] L static ={(Z i ,τ i )|Z i ωA * ∪B * ,τ i =σ(Z i )};

[0026] Q dstatic ={(q j ,e j )|q j Q dynamic ∪Q dynamic ,h j =σ(q j )};

[0027] Among them, L static is the static structural feature set, Z i Represents the position coordinates of the i-th key point, A * Represents the edge point set in the static element layer, B * represents the set of corner points in the static element layer, τ i represents the feature descriptor of the i-th key point, σ(Z i ) represents the position Z i The feature vector obtained by applying the feature descriptor encoding function, q jrepresents the position coordinates of the jth key point, Q dynamic Represents the edge point set in the dynamic element layer, Q dynamic Represents the corner point set in the dynamic element layer, h j represents the feature descriptor of the jth key point, σ(q j ) represents the position q j The feature vector obtained by applying the feature descriptor encoding function; the relative position of the position shape features of the static structure and the moving object is determined by using the feature matching method, and the relative position of the position shape features of the static structure and the moving object is integrated by using the 3D modeling platform Blender to obtain the 3D model of the parking lot. The specific steps are: using FLANN to match the descriptors in the static structure feature set and the moving object feature set to obtain a feature matching set ∩; based on the feature matching set ∩ and the static structure feature set L static and the moving object feature set Q dstatic , using the 3D modeling platform Blender, a 3D model of the parking lot is generated, and the expression is:

[0028]

[0029] Among them, ∩ is the feature matching set, θ is the parking lot 3D model, L static is the feature set of the static structure in the static element layer, Q dstatic It is a feature set of moving objects in the dynamic element layer. It is a comprehensive function; based on the 3D model of the parking lot, it uses visualization rendering technology to output a complete 3D scene including static structures and dynamic objects.

[0030] As a preferred solution of the method for identifying parking space positions based on intelligent charging piles described in the present invention, wherein: the multi-attribute decision analysis method is used to extract attribute indicators that affect the parking space score in the three-dimensional model of the parking lot, and weights are assigned to each attribute indicator that affects the parking space score, and a weighted summation method is used to calculate the comprehensive score of each parking space. The specific steps are: extracting attribute indicators that affect the parking space score from the three-dimensional model of the parking lot, specifically the distance from the parking space to the charging station, the area of ​​the barrier-free area around the parking space, the vehicle density near the parking space, the pedestrian flow near the parking space, and the safety score of the parking space; assigning weights to the attribute indicators to obtain a scoring function for the attribute indicators, and calculating the comprehensive score through the scoring function of the attribute indicators, and the expression is:

[0031]

[0032] Among them, R is the comprehensive score of the parking space, W(k i ) is the nonlinear normalized weight of the i-th attribute index, Z i is the scoring function of the i-th attribute index, Hj is the auxiliary adjustment function, and π is the number of auxiliary adjustment items.

[0033] As a preferred solution of the method for identifying parking space positions based on smart charging piles according to the present invention, wherein: based on the comprehensive score of each parking space, the parking space with the highest comprehensive score is selected as the charging parking space. When the vehicle arrives at the optimal parking space, the parking space status is verified and the smart charging pile is activated for charging service. The specific steps are: defining a selection function, selecting the parking space with the highest score from the comprehensive score as the optimal parking space, and the expression is:

[0034]

[0035] Among them, C is the optimal parking space, R i is the comprehensive score of the i-th parking space, represents the choice that makes R i The largest index i; after determining the optimal parking space, the optimal parking space information is sent to the car owner through the wireless communication protocol, and navigation guidance is provided; during navigation to the optimal parking space, when the vehicle approaches the optimal parking space, the multi-sensor starts the monitoring function to detect whether the parking space is occupied; when the parking space is empty, the vehicle is allowed to enter, otherwise, the car owner is notified to reselect a parking space.

[0036] In a second aspect, the present invention provides a system for identifying parking space locations based on smart charging piles, including a data acquisition and processing module, a static element extraction module, a dynamic element extraction module, a three-dimensional modeling module, and a parking space scoring and selection module; the data acquisition and processing module is used to start the smart charging pile and initialize multiple sensors, collect raw data on the environment in the parking lot, and then obtain an optimized environmental data set through steps such as cleaning, denoising, standardization, time synchronization, and spatial alignment; the static element extraction module is used to apply computer vision algorithms and geometric analysis methods to extract static element features from the optimized environmental data set; the dynamic element extraction module is used to use a motion detection algorithm to extract static element features from the optimized environmental data set. The dynamic elements are extracted from the optimized environmental data set, the position information of the dynamic elements is mapped to the spatial coordinate system, and the dynamic element layer is constructed; the three-dimensional modeling module is used to integrate the static element layer and the dynamic element layer, and the relative position between the static structure and the moving object is determined by the geometric feature extraction and feature matching method, and a complete three-dimensional model of the parking lot is generated by the three-dimensional modeling platform; the parking space scoring selection module is used to extract the attribute indicators that affect the parking space score from the three-dimensional model of the parking lot, calculate the comprehensive score of each parking space, select the optimal charging parking space, and send it to the owner through the wireless communication protocol, and when the vehicle arrives at the optimal parking space, verify the parking space status and activate the smart charging pile for charging service.

[0037] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for identifying the location of a parking space based on a smart charging pile as described in the first aspect of the present invention is implemented.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for identifying parking space locations based on smart charging piles as described in the first aspect of the present invention is implemented.

[0039] The beneficial effects of the present invention are as follows: by applying computer vision algorithms and geometric analysis methods to image data in an optimized environmental data set, static element features are extracted and the position and shape of fixed structures in the static element features are confirmed to obtain a static element layer, thereby achieving accurate extraction of static element features from the optimized environmental data set, thereby achieving the effect of enhancing the accuracy and reliability of parking space position recognition; by adopting a motion detection algorithm, dynamic elements are extracted from video data in the optimized environmental data set, and the position information of the dynamic elements is mapped to a spatial coordinate system to obtain a dynamic element layer, thereby achieving accurate tracking of dynamic objects, achieving the effect of real-time updating of the dynamic situation in the parking lot and assisting intelligent decision-making; by using a three-dimensional modeling platform Blender to integrate the relative positions of static structures and moving objects, a complete three-dimensional model of the parking lot is generated, which not only improves the user experience, but also provides a visualization tool for intelligent management and scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0041] Figure 1 This is a flow chart of the method for identifying parking space locations based on smart charging piles in Example 1.

[0042] Figure 2 This is a system diagram for identifying parking space locations based on smart charging piles in Example 1. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0046] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a method for identifying a parking space location based on a smart charging pile, comprising the following steps:

[0047] S1. Start the smart charging pile, use multi-sensor fusion technology to collect the original environmental data set in the parking lot, and pre-process the collected original environmental data set to obtain an optimized environmental data set;

[0048] Furthermore, the smart charging pile is started, and the original environmental data set in the parking lot is collected using multi-sensor fusion technology, and the collected original environmental data set is pre-processed to obtain an optimized environmental data set. The specific steps are: start the smart charging pile and initialize the multiple sensors installed on the smart charging pile; after starting the smart charging pile, the smart charging pile checks the health status of the multiple sensors through the Self-Diagnosis Program; the smart charging pile controls the multiple sensors to collect the environment in the parking lot to obtain the original environmental data set; an anomaly detection method is used to identify the original environmental data set and eliminate obviously erroneous data points, and a Kalman filter is used to remove high-frequency noise to obtain a preliminary optimized data set; the numerical data in the preliminary optimized data set is Z-score standardized to obtain a standardized optimized data set; the network time protocol is used to synchronize the timestamps of the multiple sensors, and the spatial coordinate system of the multiple sensors is adjusted through the external parameter information preset by the smart sensor to make it consistent with the global coordinate system to obtain an optimized environmental data set.

[0049] It should be noted that the data collected by multi-sensor fusion technology undergoes a series of preprocessing steps, which not only ensures the quality and reliability of the original data, but also provides a solid foundation for subsequent feature extraction and analysis. In particular, the application of the Kalman filter effectively reduces high-frequency noise, and the Z-score standardization processing enables data from different sensors to be compared on the same scale, enhancing the consistency and accuracy of data analysis. In addition, timestamp synchronization and spatial coordinate system adjustment ensure the temporal consistency and spatial positioning accuracy of multi-source data, and improve the accuracy of three-dimensional model construction.

[0050] S2. Apply computer vision algorithms and geometric analysis methods to the image data in the optimized environment data set, extract static element features and confirm the position and shape of fixed structures in the static element features to obtain a static element layer;

[0051] Furthermore, computer vision algorithms and geometric analysis methods are applied to the image data in the optimized environment data set to extract static element features and confirm the position and shape of fixed structures in the static element features to obtain static element layers. The specific steps are: using the Canny edge detection algorithm to identify the boundary contours of the image data in the optimized environment data set, and using the Harris corner detection algorithm to find the significant corners in the image data in the optimized environment data set. The expression is as follows:

[0052]

[0053] R(p)=det(M(p))-k·trac e 2 (M(p));

[0054] Among them, E(p) is the boundary contour strength of point p, is the gradient operator of the Gaussian filter, I(p) is the image grayscale value of point p, M(p) is the structure tensor matrix, det(M(p)) is the determinant of the matrix M(p), trace e (M(p)) is the trace of the matrix M(p), k is an empirical constant, and R(p) represents the corner point response value; the boundary contour strength E(p) and the corner point response value R(p) are combined to obtain the feature set F 1 ; Based on the feature set, use the findContours function to outline the contours of the fixed structure in the feature set, and use SIFT to compare the feature set with the contours in the optimized environment dataset. The expression is as follows:

[0055] C={p|p∈F 1 .E,isClosed(p)};

[0056]

[0057] Among them, F 1 .E is from F 1 The boundary contour strength E(p) extracted from isClosed(p) is used to determine whether point p forms a closed contour. i is the i-th contour, T i is the i-th template, γ is the adjustment coefficient, d(C i ,T i ) is the matching error between the ith contour and the ith template, S(C i ) represents the shape matching score of the i-th contour; the shape matching score S(C i ) to obtain the intermediate feature set F 2 ; The feature set F 1 and the intermediate feature set F 2 The shape matching score S(C i ) to obtain the static element feature F 3 ; Calculate the static element feature F 3 The spatial relationship score M(C i ), the expression is:

[0058]

[0059] in, is the adjustment coefficient, M(C i ) is the spatial relationship score of the i-th contour; the comprehensive feature scoring function F(E) is combined with the static element feature F 3 , construct a static element layer, the expression is:

[0060]

[0061] L={F 1 ,M(C i ),Δ};

[0062] Among them, Δ is the feature score of the static element, n is the number of static elements, α i is the shape matching score weight of the i-th static element, F 2 .S(i) is the shape matching score of the i-th static element in the intermediate feature set, β i is the spatial relationship score weight of the i-th static element, F 3 .M(C i ) is the spatial relationship score of the ith static element in the static element feature, m is the number of computing resources used in the feature extraction process, W j is the weight of the j-th computing resource, C jis the consumption of the j-th computing resource, and L is the static element layer.

[0063] It should be noted that the joint application of Canny edge detection and Harris corner detection algorithms can effectively separate important geometric features from images. Features are crucial for identifying the position and shape of fixed structures. The accuracy of the contour is further verified by SIFT feature matching, which improves the robustness of static element feature extraction. The static element feature scoring function comprehensively considers the shape matching score and spatial relationship score, and introduces the weight of computing resource consumption, which optimizes the construction process of the static element layer, ensuring high precision and improving computing efficiency, laying the foundation for the subsequent extraction of dynamic elements and determination of relative positions.

[0064] S3, extracting dynamic elements from the video data in the optimized environment data set by using a motion detection algorithm, mapping the position information of the dynamic elements to a spatial coordinate system, and obtaining a dynamic element layer;

[0065] Furthermore, by adopting the motion detection algorithm, the dynamic elements are extracted from the video data in the optimized environment data set, and the position information of the dynamic elements is mapped to the spatial coordinate system to obtain the dynamic element layer. The specific steps are: select the optical flow method, calculate the pixel displacement between consecutive frames in the video data, and obtain the velocity vector v(x, y) of the pixel point; through the obtained velocity vector v(x, y), use the threshold function to distinguish the static background and dynamic foreground in the velocity vector v(x, y), the expression is:

[0066]

[0067] Where D(x,y) is the binary dynamic element mask, x is the horizontal coordinate of the pixel, y is the vertical coordinate of the pixel, ∥v(x,y)∥ is the length of the velocity vector v(x,y) at the position (x,y), is the threshold of the velocity vector length; when the velocity vector length of v(x,y) exceeds the set threshold The position is considered to belong to the dynamic foreground, otherwise it is considered to be the static background, and the position information of the dynamic foreground is obtained and marked as a dynamic element; the dynamic element is mapped to the global coordinate system to obtain the dynamic element layer Q.

[0068] It should be noted that the application of the optical flow method can effectively distinguish between static background and dynamic foreground, improve the sensitivity and accuracy of dynamic object detection, and the use of the velocity vector length threshold function ensures the effective separation of dynamic foreground and static background, avoiding false detection and missed detection. The position information of dynamic elements is mapped to the global coordinate system, so that the system can grasp the precise position of moving objects such as vehicles in the parking lot in real time, providing important information for path planning and safety monitoring. The process realizes real-time updating of the dynamic situation of the parking lot and assists the operation of the intelligent decision-making system.

[0069] S4, extracting the position and shape features of the static structure and the moving object from the static and dynamic element layers respectively by using a geometric feature extraction method, determining the relative positions of the position and shape features of the static structure and the moving object by using a feature matching method, integrating the relative positions of the position and shape features of the static structure and the moving object by using a 3D modeling platform Blender, and obtaining a 3D model of the parking lot;

[0070] Furthermore, the position and shape features of the static structure and the moving object are extracted from the static and dynamic element layers respectively by using the geometric feature extraction method. The specific steps are as follows: applying the edge detection algorithm and the corner point detection algorithm to extract the contours and key points of the static structure and the moving object in the static element layer and the dynamic element layer respectively; using the feature descriptor to encode the contours and key points of the static structure and the moving object, and obtaining the static structure feature set L static and the moving object feature set Q dstatic , the expression is:

[0071] L static ={(Z i ,τ i )|Z i ωA * ∪B * ,τ i =σ(Z i )};

[0072] Q dstatic ={(q j ,e j )|q j Q dynamic ∪Q dynamic ,h j =σ(q j )};

[0073] Among them, L static is the static structural feature set, Z i Represents the position coordinates of the i-th key point, A * Represents the edge point set in the static element layer, B * represents the set of corner points in the static element layer, τi represents the feature descriptor of the i-th key point, σ(Z i ) represents the position Z i The feature vector obtained by applying the feature descriptor encoding function, q j represents the position coordinates of the jth key point, Q dynamic Represents the edge point set in the dynamic element layer, Q dynamic Represents the corner point set in the dynamic element layer, h j represents the feature descriptor of the jth key point, σ(q j ) represents the position q j The feature vector obtained by applying the feature descriptor encoding function; the relative position of the position shape features of the static structure and the moving object is determined by using the feature matching method, and the relative position of the position shape features of the static structure and the moving object is integrated by using the 3D modeling platform Blender to obtain the 3D model of the parking lot. The specific steps are as follows: FLANN is used to match the descriptors in the static structure feature set and the moving object feature set to obtain the feature matching set ∩; based on the feature matching set ∩ and the static structure feature set L static and the moving object feature set Q dstatic , using the 3D modeling platform Blender, a 3D model of the parking lot is generated, and the expression is:

[0074]

[0075] Among them, ∩ is the feature matching set, θ is the parking lot 3D model, L static is the feature set of the static structure in the static element layer, Q dstatic It is a feature set of moving objects in the dynamic element layer. It is a comprehensive function; based on the 3D model of the parking lot, it uses visualization rendering technology to output a complete 3D scene including static structures and dynamic objects.

[0076] It should be noted that the geometric feature extraction method and feature descriptor encoding provide unique "fingerprints" for static structures and moving objects, which facilitates subsequent feature matching and identification. The application of the FLANN matching algorithm improves the speed and accuracy of feature matching and provides a reliable correspondence for 3D model generation. The 3D modeling platform Blender is used to integrate the relative positions of static structures and moving objects to generate a high-precision, interactive 3D parking lot scene, which not only improves the user experience, but also provides a visualization tool for intelligent management and scheduling. The 3D model supports a variety of application scenarios, such as parking space recommendation, path planning and security monitoring.

[0077] S5. Use a multi-attribute decision analysis method to extract attribute indicators that affect the parking space score in the three-dimensional model of the parking lot, assign a weight to each attribute indicator that affects the parking space score, and use a weighted summation method to calculate the comprehensive score of each parking space;

[0078] Furthermore, a multi-attribute decision analysis method is used to extract attribute indicators that affect the parking space score in the three-dimensional model of the parking lot, and weights are assigned to each attribute indicator that affects the parking space score. The weighted summation method is used to calculate the comprehensive score of each parking space. The specific steps are: extract attribute indicators that affect the parking space score from the three-dimensional model of the parking lot, specifically the distance from the parking space to the charging station, the barrier-free area around the parking space, the vehicle density near the parking space, the pedestrian flow near the parking space, and the safety score of the parking space; assign weights to the attribute indicators to obtain the scoring function of the attribute indicators, and calculate the comprehensive score through the scoring function of the attribute indicators. The expression is:

[0079]

[0080] Among them, R is the comprehensive score of the parking space, W(k i ) is the nonlinear normalized weight of the i-th attribute index, Z i is the scoring function of the i-th attribute index, H j is the auxiliary adjustment function, and π is the number of auxiliary adjustment items.

[0081] It should be noted that the multi-attribute decision analysis method (MADM) ensures the scientificity and rationality of the scoring system by quantitatively evaluating the key attribute indicators that affect the parking space score and assigning appropriate weights. The weighted summation method comprehensively considers the influence of multiple factors, including the distance to the charging station, the area of ​​the barrier-free area, the vehicle density, the pedestrian flow and the safety score, to obtain a comprehensive score for each parking space. This method not only improves the accuracy and rationality of parking space selection, but also provides users with more personalized parking suggestions, thereby improving the overall service quality and user experience.

[0082] S6. Based on the comprehensive score of each parking space, the parking space with the highest comprehensive score is selected as the charging parking space. When the vehicle arrives at the optimal parking space, the parking space status is verified and the smart charging pile is activated for charging service.

[0083] Furthermore, based on the comprehensive score of each parking space, the parking space with the highest comprehensive score is selected as the charging parking space. When the vehicle arrives at the optimal parking space, the parking space status is verified and the smart charging pile is activated for charging service. The specific steps are: define a selection function, select the parking space with the highest comprehensive score as the optimal parking space, and the expression is:

[0084]

[0085] Among them, C is the optimal parking space, Ri is the comprehensive score of the i-th parking space, represents the choice that makes R i The largest index i; after determining the optimal parking space, the optimal parking space information is sent to the car owner through the wireless communication protocol, and navigation guidance is provided; during navigation to the optimal parking space, when the vehicle approaches the optimal parking space, the multi-sensor starts the monitoring function to detect whether the parking space is occupied; when the parking space is empty, the vehicle is allowed to enter, otherwise, the car owner is notified to reselect a parking space.

[0086] It should be noted that the selection function based on comprehensive scoring ensures that the selection of the optimal parking space is based on the results of a comprehensive evaluation, maximizing user convenience and satisfaction. The application of wireless communication protocols enables the optimal parking space information to be quickly transmitted to the car owner and provides navigation guidance, simplifying the parking process. The multi-sensor activation monitoring function detects the parking space status when the vehicle approaches the optimal parking space, ensuring the availability of the parking space and preventing the parking space from being occupied. When the parking space is empty, the vehicle is allowed to enter and the smart charging pile is activated for charging service, realizing the integrated management of smart parking and charging, and improving the operating efficiency and service level of the parking lot.

[0087] The present embodiment also provides a system for identifying parking space positions based on smart charging piles, including: a data acquisition and processing module, a static element extraction module, a dynamic element extraction module, a three-dimensional modeling module, and a parking space scoring and selection module; the data acquisition and processing module is used to start the smart charging pile and initialize multiple sensors, collect raw data of the environment in the parking lot, and then obtain an optimized environmental data set through steps such as cleaning, denoising, standardization, time synchronization, and spatial alignment; the static element extraction module is used to apply computer vision algorithms and geometric analysis methods to extract static element features from the optimized environmental data set; the dynamic element extraction module is used to use motion detection algorithms to extract static element features from the optimized environmental data set The dynamic elements are extracted from the environmental data set, the position information of the dynamic elements is mapped to the spatial coordinate system, and the dynamic element layer is constructed; the 3D modeling module is used to integrate the static element layer and the dynamic element layer, and the relative position between the static structure and the moving object is determined by the geometric feature extraction and feature matching methods, and the complete 3D model of the parking lot is generated by the 3D modeling platform; the parking space scoring selection module is used to extract the attribute indicators that affect the parking space score from the 3D model of the parking lot, calculate the comprehensive score of each parking space, select the optimal charging parking space and send it to the car owner through the wireless communication protocol, and when the vehicle arrives at the optimal parking space, verify the parking space status and activate the smart charging pile for charging service.

[0088] This embodiment also provides a computer device, which is suitable for the method of identifying the location of a parking space based on a smart charging pile, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the method of identifying the location of a parking space based on a smart charging pile as proposed in the above embodiment.

[0089] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0090] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for identifying the location of a parking space based on a smart charging pile as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (Static Random Access Memory, referred to as SRAM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), an erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), a programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), a read-only memory (Read-Only Memory, referred to as ROM), a magnetic storage device, a flash memory, a disk or an optical disk.

[0091] In summary, the present invention applies computer vision algorithms and geometric analysis methods to image data in the optimized environmental data set, extracts static element features and confirms the position and shape of fixed structures in the static element features to obtain a static element layer, thereby achieving accurate extraction of static element features from the optimized environmental data set, thereby achieving the effect of enhancing the accuracy and reliability of parking space position recognition. By adopting a motion detection algorithm, dynamic elements are extracted from video data in the optimized environmental data set, and the position information of the dynamic elements is mapped to a spatial coordinate system to obtain a dynamic element layer, thereby achieving accurate tracking of dynamic objects, achieving the effect of real-time updating of the dynamic situation in the parking lot and assisting intelligent decision-making. The three-dimensional modeling platform Blender is used to integrate the relative positions of static structures and moving objects to generate a complete three-dimensional model of the parking lot, which not only improves the user experience, but also provides a visualization tool for intelligent management and scheduling.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for identifying parking space locations based on smart charging piles, characterized in that: include: Start the smart charging pile, use multi-sensor fusion technology to collect the original environmental data set in the parking lot, and pre-process the collected original environmental data set to obtain an optimized environmental data set; Apply computer vision algorithms and geometric analysis methods to the image data in the optimized environmental data set, extract static element features and confirm the position and shape of fixed structures in the static element features, and obtain a static element layer; By adopting the motion detection algorithm, the dynamic elements are extracted from the video data in the optimized environment data set, and the position information of the dynamic elements is mapped to the spatial coordinate system to obtain the dynamic element layer; The position and shape features of static structures and moving objects are extracted from the static and dynamic element layers respectively by using the geometric feature extraction method, and the relative positions of the position and shape features of static structures and moving objects are determined by using the feature matching method. The relative positions of the position and shape features of static structures and moving objects are integrated by using the 3D modeling platform Blender to obtain the 3D model of the parking lot. The multi-attribute decision analysis method is used to extract the attribute indicators that affect the parking space score in the three-dimensional model of the parking lot, and a weight is assigned to each attribute indicator that affects the parking space score. The weighted summation method is used to calculate the comprehensive score of each parking space. Based on the comprehensive score of each parking space, the parking space with the highest comprehensive score is selected as the charging parking space. When the vehicle arrives at the optimal parking space, the parking space status is verified and the smart charging pile is activated for charging service.

2. The method for identifying parking space location based on a smart charging pile according to claim 1, characterized in that: The smart charging pile is started, a multi-sensor fusion technology is used to collect the original environmental data set in the parking lot, and the collected original environmental data set is preprocessed to obtain an optimized environmental data set. The specific steps are: Start the smart charging pile and initialize the multiple sensors installed on the smart charging pile; After starting the smart charging pile, the smart charging pile checks the health status of multiple sensors through the Self-Diagnosis Program; The smart charging pile controls multiple sensors to collect the environment in the parking lot and obtain the original environmental data set; The original environmental data set is identified using anomaly detection methods to remove obviously erroneous data points, and a Kalman filter is applied to remove high-frequency noise to obtain a preliminary optimized data set; Perform Z-score normalization on the numerical data in the preliminary optimized data set to obtain a standardized optimized data set; The network time protocol is used to synchronize the timestamps of multiple sensors, and the spatial coordinate system of multiple sensors is adjusted through the external parameter information preset by the intelligent sensor to make it consistent with the global coordinate system, thereby obtaining an optimized environmental data set.

3. The method for identifying parking space location based on smart charging pile according to claim 2, characterized in that: The specific steps of applying computer vision algorithms and geometric analysis methods to the image data in the optimized environment data set, extracting static element features and confirming the position and shape of fixed structures in the static element features, and obtaining a static element layer are as follows: The Canny edge detection algorithm is used to identify the boundary contours of the image data in the optimized environment dataset, and the Harris corner detection algorithm is used to find the significant corners in the image data in the optimized environment dataset. The expression is as follows: R(p)=det(M(p))-k·trac e 2 (M(p)); Among them, E(p) is the boundary contour strength of point p, is the gradient operator of the Gaussian filter, I(p) is the image grayscale value of point p, M(p) is the structure tensor matrix, det(M(p)) is the determinant of the matrix M(p), trace e (M(p)) is the trace of the matrix M(p), k is an empirical constant, and R(p) represents the corner point response value; The boundary contour strength E(p) and the corner point response value R(p) are combined to obtain the feature set F1; Based on the feature set, the findContours function is used to outline the contours of the fixed structure in the feature set, and SIFT is used to compare the feature set with the contours in the optimized environment dataset. The expression is as follows: C={p|p∈F1.E,isClosed(p)}; Among them, F1.E is the boundary contour strength E(p) extracted from F1, isClosed(p) is used to determine whether point p forms a closed contour, and C i is the i-th contour, T i is the i-th template, γ is the adjustment coefficient, d(C i ,T i ) is the matching error between the ith contour and the ith template, S(C i ) represents the shape matching score of the i-th contour; The shape matching score S(C i ) are collected to obtain the intermediate feature set F2; The shape matching score S(C i ) are integrated to obtain the static element feature F3; Calculate the spatial relationship score M(C i ), the expression is: in, is the adjustment coefficient, M(C i ) is the spatial relationship score of the i-th contour; The static element layer is constructed by using the comprehensive feature scoring function F(E) combined with the static element feature F3. The expression is: L={F1,M(C i ),Δ}; Among them, Δ is the feature score of the static element, n is the number of static elements, α i is the shape matching score weight of the i-th static element, F2.S(i) is the shape matching score of the i-th static element in the intermediate feature set, β i is the spatial relationship score weight of the ith static element, F3.M(C i ) is the spatial relationship score of the ith static element in the static element feature, m is the number of computing resources used in the feature extraction process, W j is the weight of the j-th computing resource, C j is the consumption of the j-th computing resource, and L is the static element layer.

4. The method for identifying parking space location based on a smart charging pile according to claim 3, characterized in that: The method uses a motion detection algorithm to extract dynamic elements from the video data in the optimized environment data set, maps the position information of the dynamic elements to a spatial coordinate system, and obtains a dynamic element layer. The specific steps are as follows: Select the optical flow method to calculate the pixel displacement between consecutive frames in the video data and obtain the velocity vector v(x,y) of the pixel point; Through the obtained velocity vector v(x,y), a threshold function is used to distinguish the static background and dynamic foreground in the velocity vector v(x,y). The expression is: Where D(x,y) is the binary dynamic element mask, x is the horizontal coordinate of the pixel, y is the vertical coordinate of the pixel, ∥v(x,y)∥ is the length of the velocity vector v(x,y) at the position (x,y), is the threshold of velocity vector length; When the velocity vector length of v(x,y) exceeds the set threshold If the position is a dynamic foreground, it is considered to be a static background, and the position information of the dynamic foreground is obtained and marked as a dynamic element; The dynamic elements are mapped into the global coordinate system to obtain the dynamic element layer Q.

5. The method for identifying parking space location based on smart charging pile according to claim 4, characterized in that: The geometric feature extraction method is used to extract the position and shape features of the static structure and the moving object from the static and dynamic element layers respectively, and the specific steps are as follows: Apply edge detection algorithm and corner detection algorithm to extract the outline and key points of static structures and moving objects in static element layer and dynamic element layer respectively; Use feature descriptors to encode the contours and key points of static structures and moving objects, and obtain the static structure feature set L static and the moving object feature set Q dstatic , the expression is: L static ={(Z i ,t i )∣Z i ohA * ∪B * ,t i =σ(Z i )}; Q dstatic ={(q j ,e j )∣q j ωQ dynamic ∪Q dynamic ,h j =σ(q j )}; Among them, L static is the static structural feature set, Z i Represents the position coordinates of the i-th key point, A * Represents the edge point set in the static element layer, B * represents the set of corner points in the static element layer, τ i represents the feature descriptor of the i-th key point, σ(Z i ) represents the position Z i The feature vector obtained by applying the feature descriptor encoding function, q j represents the position coordinates of the jth key point, Q dynamic Represents the edge point set in the dynamic element layer, Q dynamic Represents the corner point set in the dynamic element layer, h j represents the feature descriptor of the jth key point, σ(q j ) represents the position q j Applying the feature descriptor encoding function to obtain the feature vector; The feature matching method is used to determine the relative positions of the position shape features of the static structure and the moving object, and the relative positions of the position shape features of the static structure and the moving object are integrated using the 3D modeling platform Blender to obtain a 3D model of the parking lot. The specific steps are: FLANN is used to match the descriptors in the static structure feature set and the moving object feature set to obtain the feature matching set ∩; Based on the feature matching set ∩ and the static structure feature set L static and the moving object feature set Q dstatic , using the 3D modeling platform Blender, a 3D model of the parking lot is generated, and the expression is: Among them, ∩ is the feature matching set, θ is the parking lot 3D model, L static is the feature set of the static structure in the static element layer, Q dstatic It is a feature set of moving objects in the dynamic element layer. is a comprehensive function; Based on the 3D model of the parking lot, visualization rendering technology is used to output a complete 3D scene including static structures and dynamic objects.

6. The method for identifying parking space location based on a smart charging pile according to claim 5, characterized in that: The multi-attribute decision analysis method is used to extract attribute indicators that affect the parking space score in the three-dimensional model of the parking lot, and a weight is assigned to each attribute indicator that affects the parking space score. The weighted summation method is used to calculate the comprehensive score of each parking space. The specific steps are: Extract attribute indicators that affect the parking space score from the three-dimensional model of the parking lot, specifically the distance from the parking space to the charging station, the area of ​​the barrier-free area around the parking space, the vehicle density near the parking space, the pedestrian flow near the parking space, and the safety score of the parking space; Assign weights to the attribute indicators and obtain the scoring function of the attribute indicators. The comprehensive score is calculated by the scoring function of the attribute indicators. The expression is: Among them, R is the comprehensive score of the parking space, W(k i ) is the nonlinear normalized weight of the i-th attribute index, Z i is the scoring function of the i-th attribute index, H j is the auxiliary adjustment function, and π is the number of auxiliary adjustment items.

7. The method for identifying parking space location based on a smart charging pile according to claim 6, characterized in that: The comprehensive score of each parking space is selected as the charging parking space with the highest comprehensive score. When the vehicle arrives at the optimal parking space, the parking space status is verified and the smart charging pile is activated for charging service. The specific steps are as follows: Define the selection function and select the parking space with the highest score from the comprehensive score as the optimal parking space. The expression is: Among them, C is the optimal parking space, R i is the comprehensive score of the i-th parking space, represents the choice that makes R i The largest index i; After determining the best parking space, the best parking space information is sent to the car owner through the wireless communication protocol, and navigation guidance is provided; When navigating to the optimal parking space, when the vehicle approaches the optimal parking space, multiple sensors start monitoring to detect whether the parking space is occupied; When the parking space is empty, the vehicle is allowed to enter. Otherwise, the owner is notified to reselect a parking space.

8. A system for identifying parking spaces based on a smart charging pile, based on the method for identifying parking spaces based on a smart charging pile according to any one of claims 1 to 7, characterized in that: It includes data acquisition and processing module, static element extraction module, dynamic element extraction module, 3D modeling module and parking space scoring and selection module; The data acquisition and processing module is used to start the smart charging pile and initialize multiple sensors, collect raw data of the environment in the parking lot, and then obtain an optimized environmental data set through steps such as cleaning, denoising, standardization, time synchronization and spatial alignment; The static element extraction module is used to extract static element features from the optimized environmental data set by applying computer vision algorithms and geometric analysis methods; The dynamic element extraction module is used to extract dynamic elements from the optimized environment data set using a motion detection algorithm, map the position information of the dynamic elements to a spatial coordinate system, and construct a dynamic element layer; The three-dimensional modeling module is used to integrate the static element layer and the dynamic element layer, use geometric feature extraction and feature matching methods to determine the relative position between the static structure and the moving object, and use the three-dimensional modeling platform to generate a complete three-dimensional model of the parking lot; The parking space scoring selection module is used to extract attribute indicators that affect the parking space score from the three-dimensional model of the parking lot, calculate the comprehensive score of each parking space, select the optimal charging parking space and send it to the car owner through the wireless communication protocol, and when the vehicle arrives at the optimal parking space, verify the parking space status and activate the smart charging pile for charging service.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for identifying the location of a parking space based on a smart charging pile according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for identifying the location of a parking space based on a smart charging pile according to any one of claims 1 to 7 are implemented.

Images