Intelligent parking lot comprehensive identification system and method

Abstract

The invention is suitable for the technical field of image recognition, and provides an intelligent parking lot comprehensive recognition system and method, and the method comprises the steps: collecting real-time video data, and extracting a vehicle region in the real-time video data; in a preset vehicle model database, identifying target vehicle information corresponding to the vehicle area; matching license plate position information corresponding to the target vehicle information, and extracting a license plate area in the vehicle area based on the license plate position information; wherein the license plate position information comprises a distance value between the license plate position and each feature region; extracting license plate number information in the license plate area; calling a parking process according to the vehicle number information; wherein the parking process comprises parking lot admission release or departure fee accounting release. The method not only depends on the license plate information, but also combines the overall information of the vehicle, so that more comprehensive and accurate vehicle management is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of image recognition, and in particular relates to an intelligent parking lot comprehensive recognition system and method. Background Art

[0002] Smart parking systems have been widely used in recent years, primarily to improve parking management efficiency and user experience. Traditional parking management models often rely on manual oversight, which is not only inefficient but also prone to errors, impacting vehicle entry and exit speeds and user satisfaction. With the advancement of computer vision and artificial intelligence technologies, vehicle recognition based on video data has gradually become one of the core technologies of smart parking systems.

[0003] Existing license plate recognition technologies typically use cameras to capture vehicle images and then use image processing techniques to identify the license plate number and perform related operations. These methods typically focus solely on the license plate itself, neglecting the comprehensive utilization of the vehicle's overall information. This results in low recognition efficiency and is unable to meet the needs of large-scale intelligent parking management. Summary of the Invention

[0004] In view of this, an embodiment of the present invention provides an intelligent parking lot comprehensive identification system and method to solve the technical problem that traditional technology only focuses on the information of the license plate itself, but ignores the comprehensive utilization of the overall vehicle information, resulting in low recognition efficiency.

[0005] A first aspect of an embodiment of the present invention provides a method for comprehensive identification of an intelligent parking lot, the method comprising:

[0006] Collecting real-time video data and extracting a vehicle area in the real-time video data;

[0007] Identifying target vehicle information corresponding to the vehicle area in a preset vehicle model database;

[0008] Matching the license plate position information corresponding to the target vehicle information, and extracting the license plate area in the vehicle area based on the license plate position information; wherein the license plate position information includes the distance value between the license plate position and each feature area;

[0009] Extracting license plate number information from the license plate area;

[0010] The parking process is called according to the vehicle number information; wherein the parking process includes parking lot admission and release or exit fee calculation and release.

[0011] Furthermore, the step of identifying target vehicle information corresponding to the vehicle area in the preset vehicle model database includes:

[0012] Acquire multiple feature areas corresponding to different pre-stored vehicle information; wherein the feature areas include a headlight area, a grille area, a hood area, a front bumper area, and a vehicle logo area;

[0013] Performing contour detection on the vehicle area to obtain a plurality of first closed contour areas;

[0014] Calculating first similarities between the plurality of first closed contour regions and the plurality of feature regions;

[0015] Taking a feature region whose first similarity is greater than or equal to a first threshold as a target feature region;

[0016] Count the number of target feature areas corresponding to each type of vehicle information;

[0017] When the number of target feature areas corresponding to the current vehicle information is greater than a preset number, the current vehicle information is used as the target vehicle information.

[0018] Furthermore, the step of calculating the first similarities between the multiple closed contour regions and the multiple feature regions includes:

[0019] Extracting the center point of the first closed contour area;

[0020] Calculating a first distance between a plurality of edge pixel points of the first closed contour area and the center point;

[0021] Taking the edge pixel point corresponding to the maximum first distance as the starting point, extracting multiple target edge pixel points on the edge of the first closed contour area based on a preset sampling frequency;

[0022] Calculating the second distances between adjacent target edge pixel points according to the original arrangement order of the plurality of target edge pixel points;

[0023] Normalizing the second distances corresponding to the plurality of target edge pixel points to obtain a feature distance;

[0024] Combining the plurality of characteristic distances into a current distance characteristic vector according to the original arrangement order of the plurality of target edge pixel points;

[0025] Extracting a standard distance feature vector corresponding to the feature area;

[0026] Calculating a first vector similarity between the current distance feature vector and the standard distance feature vector;

[0027] The first vector similarity is used as the first similarity.

[0028] Furthermore, the step of matching the license plate position information corresponding to the target vehicle information and extracting the license plate area in the vehicle area based on the license plate position information includes:

[0029] Matching the license plate position information corresponding to the target vehicle information;

[0030] The center of the first closed contour area with a first similarity greater than or equal to a first threshold is used as a first anchor point, and the center of the feature area with a first similarity greater than the first threshold is used as a second anchor point;

[0031] taking a first closed contour region where the first similarity is less than a first threshold as a region to be identified;

[0032] Calculating third distances between the center of the to-be-identified area and the plurality of first anchor points respectively;

[0033] Normalize the multiple third distances and construct them into the current distance vector;

[0034] extracting fourth distances between a plurality of second anchor points and the license plate position from the license plate position information;

[0035] Normalizing the multiple fourth distances and constructing them into a standard distance vector; wherein the same element positions in the current distance vector and the standard distance vector are used to respectively accommodate the values ​​corresponding to the first anchor point and the second anchor point corresponding to the same first similarity;

[0036] Calculating a second vector similarity between the current distance vector and the standard distance vector;

[0037] If the second vector similarity is greater than a second threshold, the area to be identified corresponding to the second vector similarity is used as the license plate area.

[0038] Furthermore, the step of extracting the license plate number information in the license plate area includes:

[0039] Performing edge detection on the license plate area to obtain a plurality of second closed contour areas;

[0040] Acquire pre-stored standard license plate character image data and the license plate character corresponding to the standard license plate character image data; wherein the standard license plate character image data includes Chinese character image data, English character image data and Roman numeral character image data;

[0041] Extracting the first lengths of the plurality of second closed contour areas, and extracting the second length of the standard license plate character image data;

[0042] Based on the first length and the second length, the plurality of second closed contour regions are respectively enlarged or reduced in equal proportion to obtain a region to be identified; wherein the current length of the region to be identified is consistent with the second length;

[0043] Binarize the area to be identified to obtain a first image;

[0044] Binarizing the standard license plate character image data to obtain a second image;

[0045] constructing an image coordinate system based on the central axis and midpoint of the first image and the second image;

[0046] Calculating a difference between a first pixel value corresponding to the first image and a second pixel value corresponding to the second image at the same image position;

[0047] Counting the number of image positions where the difference is 0;

[0048] Dividing the number of image positions by the number of pixels in the second image to obtain a first value;

[0049] If the first value is greater than a third threshold, the license plate character corresponding to the standard license plate character image data corresponding to the first value is used as the license plate information corresponding to the second closed contour area corresponding to the first value;

[0050] Based on the original arrangement order of the plurality of second closed contour areas in the vehicle area, the plurality of license plate information are combined to obtain the license plate number information.

[0051] Furthermore, the step of collecting real-time video data and extracting the vehicle area in the real-time video data includes:

[0052] Acquire a standard video frame; wherein the standard video frame refers to a video frame in which no vehicle exists in the picture;

[0053] Calculating the number of difference pixels between the real-time video frame and the standard video frame in the real-time video data; wherein the number of difference pixels refers to the number of pixels corresponding to the pixel difference at the same pixel position being greater than a preset value;

[0054] If the number of the difference pixels exceeds a fourth threshold, obtaining a plurality of subsequent video frames; wherein the subsequent video frames refer to video frames located after the real-time video frame;

[0055] extracting a foreground region and a background region based on a plurality of the subsequent video frames;

[0056] If the ratio between the foreground area and the background area exceeds a fifth threshold, the image area formed by the current difference pixel points between the standard video frame and the subsequent video frame is used as the vehicle area; wherein the current difference pixel point refers to the pixel point corresponding to the pixel difference value at the same pixel position is greater than the preset value.

[0057] Furthermore, the step of extracting the foreground area and the background area based on the plurality of subsequent video frames includes:

[0058] In the plurality of subsequent video frames, sequentially calculating current pixel differences between identical pixel positions in adjacent subsequent video frames;

[0059] Counting a first number of a plurality of current pixel difference values ​​corresponding to the same pixel position that are greater than a sixth threshold;

[0060] If the first number is 0, the same pixel position is determined as a full background pixel point;

[0061] If the first number is between the second value and the third value, determining the same pixel position as a half-background pixel point;

[0062] If the first number is greater than a third value, determining the same pixel position as a full foreground pixel;

[0063] Obtain a first region consisting of continuous all-foreground pixels, and use the largest first region as the initial region;

[0064] Acquire a second region consisting of half background pixels;

[0065] If the initial region is adjacent to the second region, merging the initial region with the second region to obtain the foreground region;

[0066] The third area formed by all background pixels and the unmerged second area are used as the background area.

[0067] A second aspect of an embodiment of the present invention provides an intelligent parking lot comprehensive identification device, comprising:

[0068] An acquisition unit, configured to acquire real-time video data and extract a vehicle region from the real-time video data;

[0069] an identification unit, configured to identify target vehicle information corresponding to the vehicle area in a preset vehicle model database;

[0070] a matching unit, configured to match the license plate position information corresponding to the target vehicle information, and extract a license plate region in the vehicle region based on the license plate position information; wherein the license plate position information includes a distance value between the license plate position and each feature region;

[0071] an extraction unit, configured to extract the license plate number information in the license plate area;

[0072] The calling unit is used to call the parking process according to the vehicle number information; wherein the parking process includes parking lot admission release or exit fee calculation release.

[0073] A third aspect of an embodiment of the present invention provides an intelligent parking lot comprehensive identification system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the intelligent parking lot comprehensive identification method described in the first aspect are implemented.

[0074] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the steps of the intelligent parking lot comprehensive identification method described in the first aspect are implemented.

[0075] Compared with the prior art, the embodiments of the present invention have the following advantages: the vehicle area is extracted by collecting real-time video data, and then the target vehicle information is identified in a preset vehicle model database. Compared with the traditional single license plate recognition method, this comprehensive recognition method can utilize more vehicle feature information, thereby improving the accuracy and robustness of recognition. By matching the license plate position information corresponding to the target vehicle information, the license plate area is extracted from the vehicle area based on the license plate position information. Such a method can effectively reduce recognition errors caused by factors such as ambient light, vehicle angle, and license plate damage, thereby improving the accuracy of license plate number recognition. By quickly matching vehicle information with a preset vehicle model database, the time required for recognition is greatly shortened, and the overall processing efficiency of the system is improved. The method of the present invention not only relies on license plate information, but also combines the overall vehicle information, thereby achieving more comprehensive and accurate vehicle management. This comprehensive information utilization method can better prevent problems such as vehicle information falsification or license plate duplication, and improve the safety and reliability of parking lots. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0077] Figure 1 A schematic flow chart of a comprehensive identification method for an intelligent parking lot provided by the present invention is shown;

[0078] Figure 2 A schematic diagram of an intelligent parking lot comprehensive identification device provided by an embodiment of the present invention is shown;

[0079] Figure 3 A schematic diagram of an intelligent parking lot comprehensive identification system provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0080] In the following description, specific details such as particular system structures and techniques are provided for purposes of illustration, not limitation, to facilitate a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

[0081] The embodiments of the present invention provide an intelligent parking lot comprehensive identification system and method to solve the technical problem that traditional technology only focuses on the information of the license plate itself, but ignores the comprehensive utilization of the overall vehicle information, resulting in low recognition efficiency.

[0082] First, the present invention provides a comprehensive identification method for intelligent parking lots. Figure 1 , Figure 1 FIG1 shows a schematic flow chart of a comprehensive identification method for an intelligent parking lot provided by the present invention. Figure 1 As shown, the intelligent parking lot comprehensive identification method may include the following steps:

[0083] Step 101: collecting real-time video data and extracting the vehicle area in the real-time video data;

[0084] Real-time video data from the parking lot is collected using video surveillance equipment. Image processing technology is then used to identify and extract the area containing vehicles from the video data. The vehicle area extraction logic is as follows:

[0085] Specifically, step 101 includes steps 1011 to 1015:

[0086] Step 1011: Acquire a standard video frame; wherein the standard video frame refers to a video frame in which no vehicle exists in the picture;

[0087] The standard video frame is a reference frame used for subsequent difference calculations. This frame is a picture without a vehicle and is used to compare with the real-time video frame to detect the presence of a vehicle.

[0088] Step 1012: Calculate the number of difference pixels between the real-time video frame and the standard video frame in the real-time video data; wherein the number of difference pixels refers to the number of pixels corresponding to a pixel difference greater than a preset value at the same pixel position;

[0089] Compare the difference between the real-time video frame and the standard video frame at each pixel position. If the difference at a certain pixel position exceeds a preset value, the pixel is considered a difference pixel. Then count all the difference pixels.

[0090] Step 1013: If the number of difference pixels exceeds a fourth threshold, obtaining multiple subsequent video frames; wherein the subsequent video frames refer to video frames located after the real-time video frame;

[0091] If the number of different pixels exceeds a preset threshold (i.e., a fourth threshold, which can be set to half the total number of pixels, for example), it is considered that a vehicle may have entered the frame. In this case, further analysis is required to obtain multiple subsequent video frames to confirm the presence and location of the vehicle.

[0092] Step 1014: extracting a foreground area and a background area based on the plurality of subsequent video frames;

[0093] By analyzing multiple subsequent video frames, the foreground (possibly a vehicle) and the background (unchanged parts) in the video frames can be distinguished. The foreground region refers to the part that appears to move or change in the subsequent video frames, while the background region remains unchanged. The specific logic for extracting the foreground and background regions is as follows:

[0094] Specifically, step 1014 includes steps 10141 to 10149:

[0095] Step 10141: in the plurality of subsequent video frames, sequentially calculating current pixel differences between the same pixel positions in adjacent subsequent video frames;

[0096] For each pixel position, the difference in pixel values ​​between adjacent subsequent video frames is calculated. This is to detect the degree of pixel value change at different time points.

[0097] Step 10142: Count a first number of the current pixels corresponding to the same pixel position whose difference values ​​are greater than a sixth threshold;

[0098] Count the difference between each pixel position in all subsequent video frames and record the number of times (first number) the difference is greater than a preset threshold (sixth threshold). The sixth threshold can be set based on the fluctuation pattern of the pixel points.

[0099] Step 10143: If the first number is 0, the same pixel position is determined as a full background pixel point;

[0100] Step 10144: If the first number is between the second value and the third value, determining the same pixel position as a half-background pixel point;

[0101] Step 10145: If the first number is greater than a third value, determining the same pixel position as a full foreground pixel;

[0102] Classify each pixel according to the value of the first quantity:

[0103] If the first number of the position is 0, it means that the pixel value of the position has not changed significantly in all subsequent frames, and it is determined to be a full background pixel.

[0104] If the first number is between two thresholds (the second value and the third value), it means that the pixel value at this position sometimes changes and sometimes remains unchanged, and it is determined to be a semi-background pixel.

[0105] If the first number is greater than the third value, it means that the pixel value at this position changes frequently and is determined to be a full foreground pixel.

[0106] Step 10146: Acquire a first region consisting of continuous all foreground pixels, and use the largest first region as the initial region;

[0107] Obtain a first region consisting of all continuous foreground pixels, and use the largest first region as the initial region. Find a continuous region consisting of all foreground pixels, and use the largest continuous region as the initial foreground region.

[0108] Step 10147: Acquire a second region consisting of half background pixels;

[0109] Find all half-background pixels and call the area they form the second area.

[0110] Step 10148: If the initial region is adjacent to the second region, merge the initial region and the second region to obtain the foreground region;

[0111] Check if the initial foreground region and the half-background region are adjacent. If so, merge the two regions into the final foreground region.

[0112] Step 10149: The third area formed by all background pixels and the unmerged second area are used as the background area.

[0113] The area consisting of all background pixels is called the third area. If there is a part in the second area that is not merged with the initial area, this part is also considered part of the background area. Ultimately, the background area includes the third area and the unmerged second area.

[0114] In the embodiment corresponding to steps 10141 to 10149, through these logical steps, the foreground area (mainly vehicles) and background area can be effectively extracted from multiple subsequent video frames to ensure the accuracy and reliability of vehicle detection.

[0115] Step 1015: If the ratio between the foreground area and the background area exceeds the fifth threshold, the image area formed by the current difference pixel points between the standard video frame and the subsequent video frame is used as the vehicle area; wherein the current difference pixel point refers to the pixel point corresponding to the same pixel position whose pixel difference value is greater than the preset value.

[0116] The ratio of the foreground area to the background area is calculated. If this ratio exceeds a preset threshold (i.e., a fifth threshold), the foreground area is considered to be mostly vehicles. In this case, the image area consisting of the difference pixels between the standard video frame and the subsequent video frame is determined to be the vehicle area.

[0117] In the embodiment corresponding to steps 1011 to 1015, the above steps can accurately extract the vehicle area in the real-time video data. These steps utilize image difference analysis and foreground-background separation technology to effectively detect and locate the vehicle, providing a foundation for subsequent license plate recognition and parking management.

[0118] Step 102: Identify target vehicle information corresponding to the vehicle area in a preset vehicle model database;

[0119] Once the vehicle area is extracted, the system uses the image information in the area to match it in a pre-built vehicle model database.

[0120] Specifically, step 102 includes steps 1021 to 1026:

[0121] Step 1021: Acquire multiple feature areas corresponding to different pre-stored vehicle information; wherein the feature areas include a headlight area, a grille area, a hood area, a front bumper area, and a vehicle logo area;

[0122] First, we extract multiple feature regions corresponding to each vehicle from a pre-defined vehicle model database. These feature regions are key vehicle components, such as the headlights, grille, hood, front bumper, and logo. These regions have distinct visual characteristics, making them easy to identify and compare.

[0123] Step 1022: performing contour detection on the vehicle area to obtain a plurality of first closed contour areas;

[0124] Contour detection is an image processing technique used to detect edges and shapes in an image, thereby obtaining multiple closed regions that may contain characteristic parts of a vehicle.

[0125] Step 1023: Calculating first similarities between the plurality of first closed contour regions and the plurality of feature regions;

[0126] The detected multiple closed contour regions are compared with the pre-stored feature regions to calculate the similarity between them. The first similarity is a quantitative indicator used to represent the matching degree between the closed contour region and the feature region.

[0127] Specifically, step 1023 includes steps 10231 to 10239:

[0128] Step 10231: extracting the center point of the first closed contour area;

[0129] Extract the center point from the detected closed contour area. This center point is the geometric center of the closed area, usually obtained by calculating the average position of all edge pixels in the area.

[0130] Step 10232: Calculate first distances between a plurality of edge pixel points of the first closed contour area and the center point;

[0131] Calculate the distance between each edge pixel and the center point in the closed contour area. These distances reflect the shape characteristics of the closed contour.

[0132] Step 10233: Using the edge pixel point corresponding to the maximum first distance as a starting point, extracting multiple target edge pixel points on the edge of the first closed contour area based on a preset sampling frequency;

[0133] Find the edge pixel farthest from the center point and use it as the starting point. Then, starting from the starting point, evenly extract multiple target edge pixels along the edge of the closed contour according to the preset sampling frequency. These points will be used for further distance calculations.

[0134] Step 10234: Calculate the second distances between adjacent target edge pixel points according to the original arrangement order of the plurality of target edge pixel points;

[0135] Based on the order of the extracted target edge pixels, the distance between each pair of adjacent points is calculated. These distances are called second distances and reflect the local shape characteristics of the closed contour area.

[0136] Step 10235: normalize the second distances corresponding to the multiple target edge pixels to obtain a feature distance;

[0137] The calculated second distance is normalized to eliminate the effects of image scaling or distance change. Normalization generally involves scaling the distance value to a standard range (eg, between 0 and 1).

[0138] Step 10236: combining the plurality of characteristic distances into a current distance feature vector according to the original arrangement order of the plurality of target edge pixels;

[0139] The normalized feature distances are combined in their original order to form a current distance feature vector, which represents the shape features of the closed contour area.

[0140] Step 10237: extracting the standard distance feature vector corresponding to the feature area;

[0141] Extract the standard distance feature vector for the corresponding feature area from the preset vehicle model database. This vector is pre-calculated and represents the shape characteristics of the standard vehicle feature. The calculation logic for the standard distance feature vector is similar to that of the current distance feature vector and will not be repeated here.

[0142] Step 10238: Calculate the first vector similarity between the current distance feature vector and the standard distance feature vector;

[0143] Calculate the similarity between the current distance feature vector and the standard distance feature vector. Similarity calculation methods include, but are not limited to, cosine similarity or Euclidean distance. This similarity measures the degree of shape matching between the current closed contour region and the standard feature region.

[0144] Step 10239: Use the first vector similarity as the first similarity.

[0145] Finally, the calculated vector similarity is used as the first similarity. This first similarity is used to indicate the matching degree between the closed contour area and the feature area, helping to identify vehicle information.

[0146] In the embodiment corresponding to steps 10231 to 10239, the similarity between the closed contour region and the feature region is calculated by calculating the distance between edge pixels and measuring vector similarity. This method improves the accuracy of shape matching and helps to more accurately identify vehicle information.

[0147] Step 1024: taking the feature region whose first similarity is greater than or equal to the first threshold as the target feature region;

[0148] By setting a first similarity threshold, feature regions with similarity greater than or equal to the threshold are screened out. These feature regions that pass the screening are called target feature regions, which means that they have sufficient similarity with the pre-stored feature regions and can be considered as corresponding vehicle features.

[0149] Step 1025: Count the number of target feature areas corresponding to each type of vehicle information;

[0150] For each vehicle model, the number of corresponding target feature areas is counted. This step further confirms the accuracy of the vehicle information by comparing the number of target feature areas.

[0151] Step 1026: When the number of target feature areas corresponding to the current vehicle information is greater than a preset number, the current vehicle information is used as the target vehicle information.

[0152] Finally, a preset threshold is set. If the number of target feature regions corresponding to a particular vehicle model exceeds this threshold (for example, three), the vehicle model is considered the target vehicle for the current detection. This step ensures that vehicle information is only confirmed when a sufficient number of feature regions match, thereby improving recognition accuracy and reliability.

[0153] In the embodiment corresponding to steps 1021 to 1026 , multiple characteristic regions of the vehicle are used to identify vehicle information, thereby enhancing the accuracy and robustness of the identification process.

[0154] Step 103: matching the license plate position information corresponding to the target vehicle information, and extracting the license plate region in the vehicle region based on the license plate position information; wherein the license plate position information includes the distance value between the license plate position and each feature region;

[0155] After identifying the vehicle's specific information, the system uses this information to locate the license plate. This preset license plate location information is derived by statistically analyzing common features of license plate positions on a large number of similar vehicles (such as the distance of the license plate relative to features such as headlights and windows). This approach improves the accuracy of license plate location. Based on this location information, the system can accurately extract the license plate area within the vehicle image.

[0156] Specifically, step 103 includes steps 1031 to 1039:

[0157] Step 1031: Matching the license plate position information corresponding to the target vehicle information;

[0158] Based on the target vehicle information, the corresponding license plate location information is found. This usually involves obtaining the standard location data of the license plate from a database or a pre-defined template.

[0159] Step 1032: The center of the first closed contour region whose first similarity is greater than or equal to the first threshold is used as the first anchor point, and the center of the feature region whose first similarity is greater than the first threshold is used as the second anchor point;

[0160] Select closed contour regions that closely match the license plate features (with a similarity greater than or equal to a first threshold). Their center points are called first anchor points. Similarly, select the center points of feature regions that closely match the license plate features as second anchor points. The first and second anchor points are used for subsequent distance calculations.

[0161] Step 1033: taking the first closed contour region with a first similarity less than a first threshold as a region to be identified;

[0162] Those closed contour areas with a low degree of matching with the license plate features (similarity less than the first threshold) are marked as areas to be identified. These areas need further verification to determine whether they contain the license plate.

[0163] Step 1034: Calculate the third distances between the center of the to-be-identified area and the plurality of first anchor points respectively;

[0164] Calculate the distance between the center point of each region to be identified and all first anchor points (center points of highly matched regions). These distances reflect the positional relationship of the region to be identified relative to the highly matched regions.

[0165] Step 1035: normalize the multiple third distances and construct a current distance vector;

[0166] The calculated third distance is normalized to eliminate the effects of different measurement units or scales. Normalization can standardize the distance values ​​to a fixed range (e.g., 0 to 1). These normalized distance values ​​are then combined into a current distance vector.

[0167] Step 1036: extracting fourth distances between a plurality of second anchor points and the license plate position from the license plate position information;

[0168] The distances between multiple second anchor points (center points of the feature area) and the standard position of the license plate are extracted from the license plate position information. These distances are used to establish a standard distance relationship.

[0169] Step 1037: normalize the multiple fourth distances and construct a standard distance vector; wherein the same element positions in the current distance vector and the standard distance vector are used to respectively accommodate the values ​​corresponding to the first anchor point and the second anchor point corresponding to the same first similarity;

[0170] The extracted fourth distance is normalized to standardize it, and the normalized distance values ​​are combined into a standard distance vector.

[0171] When constructing the current distance vector and the standard distance vector, ensure that the elements at the same position in the two vectors correspond to the same anchor point pair (i.e., the first anchor point and the second anchor point corresponding to the same first similarity). This ensures that the vector comparison is based on the same reference point.

[0172] Step 1038: Calculate a second vector similarity between the current distance vector and the standard distance vector;

[0173] The similarity between the current distance vector and the standard distance vector can be calculated by using methods such as cosine similarity and Euclidean distance. The second vector similarity reflects the degree of matching between the position relationship of the area to be identified and the position of the standard license plate.

[0174] Step 1039: If the second vector similarity is greater than a second threshold, the area to be identified corresponding to the second vector similarity is used as the license plate area.

[0175] If the calculated second vector similarity exceeds a preset second threshold (eg, 0.9), it indicates that the degree of matching between the area to be identified and the standard license plate position is sufficiently high. At this point, the area to be identified can be identified as a license plate area.

[0176] In the embodiment corresponding to steps 1031 to 1039, the method of accurately locating the license plate area using distance vectors and vector similarity can effectively improve the accuracy of license plate recognition through multi-level filtering and matching.

[0177] Step 104: extracting the license plate number information in the license plate area;

[0178] Specifically, step 104 includes steps A1 to A12:

[0179] Step A1: performing edge detection on the license plate area to obtain a plurality of second closed contour areas;

[0180] Edge detection is performed on the license plate area to obtain multiple second closed contour areas. Edge detection is an image processing technique used to identify edges in an image. Through this step, the outlines of the characters in the license plate area can be identified.

[0181] Step A2: obtaining pre-stored standard license plate character image data and the license plate characters corresponding to the standard license plate character image data; wherein the standard license plate character image data includes Chinese character image data, English character image data and Roman numeral character image data;

[0182] Obtaining pre-stored standard license plate character image data and the license plate characters corresponding to the standard license plate character image data. This step is to obtain standard license plate character image data from a database, including Chinese characters, English characters and Roman numeral character image data.

[0183] Step A3: extracting the first lengths of the plurality of second closed contour areas, and extracting the second length of the standard license plate character image data;

[0184] Step A4: Based on the first length and the second length, the plurality of second closed contour regions are respectively enlarged or reduced in equal proportion to obtain a region to be identified; wherein the current length of the region to be identified is consistent with the second length;

[0185] Extracting a first length of multiple second closed contour areas and extracting a second length of the standard license plate character image data. Measuring the length of the character contours extracted from the license plate image and the length of the standard character image. Based on the first and second lengths, proportionally enlarging or reducing the multiple second closed contour areas to obtain an area to be recognized; wherein the current length of the area to be recognized is consistent with the second length. Proportionally adjusting the extracted character contours to the same length as the standard character image to facilitate subsequent image matching.

[0186] It is understandable that since the lengths of the characters in the license plate are consistent but the widths vary, this embodiment performs scaling based on the length.

[0187] Step A5: binarizing the area to be identified to obtain a first image;

[0188] The region to be identified is binarized to obtain a first image. The binarization process converts the image into a black and white format to facilitate subsequent pixel comparison.

[0189] Step A6: binarizing the standard license plate character image data to obtain a second image;

[0190] Step A7: constructing an image coordinate system based on the central axis and midpoint of the first image and the second image;

[0191] By aligning the central axis and midpoint of the image, a unified coordinate system is established to facilitate accurate comparison of each pixel in the image.

[0192] Step A8: calculating the difference between a first pixel value corresponding to the first image and a second pixel value corresponding to the second image at the same image position;

[0193] Calculate the difference between a first pixel value corresponding to the first image and a second pixel value corresponding to the second image at the same image position. Compare the two binary images pixel by pixel and calculate the difference at each pixel position.

[0194] Step A9: Counting the number of image positions where the difference is 0;

[0195] Count the number of pixels with the same value (difference is 0) at the same pixel position in two images.

[0196] Step A10: Dividing the number of image positions by the number of pixels in the second image to obtain a first value;

[0197] The ratio of the number of matched pixels to the total number of pixels in the standard character image is calculated to obtain a similarity value (first value).

[0198] Step A11: If the first value is greater than a third threshold, taking the license plate character corresponding to the standard license plate character image data corresponding to the first value as the license plate information corresponding to the second closed contour area corresponding to the first value;

[0199] If the similarity value is higher than a third threshold value (eg, 0.8), it is considered that the character corresponding to the contour area matches the standard license plate character.

[0200] Step A12: Based on the original arrangement order of the plurality of second closed contour regions in the vehicle region, the plurality of license plate information are combined to obtain the license plate number information.

[0201] Finally, the recognized characters are combined according to their original arrangement order in the license plate image to obtain the complete license plate number information.

[0202] In the embodiment corresponding to steps A1 to A12, the specific steps of extracting and identifying the license plate number from the license plate image ensure accurate recognition of the license plate characters through a series of image processing and comparison operations.

[0203] Step 105: Call the parking process according to the vehicle number information; wherein the parking process includes parking lot admission or exit fee calculation and release.

[0204] Finally, the system uses the identified license plate number information to perform appropriate parking lot management operations. For vehicles entering the parking lot, the system verifies their access rights and automatically allows them to pass. For vehicles leaving the parking lot, the system calculates the parking fee based on the length of time parked and automatically allows them to pass after payment is confirmed.

[0205] In the embodiment corresponding to steps 101 to 105, the vehicle area is extracted by collecting real-time video data, and then the target vehicle information is identified in a preset vehicle model database. Compared to traditional single license plate recognition methods, this comprehensive recognition method can utilize more vehicle feature information, thereby improving recognition accuracy and robustness. By matching the license plate position information corresponding to the target vehicle information, the license plate area is extracted from the vehicle area based on the license plate position information. This method can effectively reduce recognition errors caused by factors such as ambient light, vehicle angle, and license plate damage, thereby improving the accuracy of license plate number recognition. By quickly matching vehicle information with a preset vehicle model database, the time required for recognition is greatly shortened, and the overall processing efficiency of the system is improved. The method of the present invention not only relies on license plate information, but also combines overall vehicle information (such as vehicle model, body features, etc.), thereby achieving more comprehensive and accurate vehicle management. This comprehensive information utilization method can better prevent problems such as vehicle information falsification or license plate duplication, and improve the safety and reliability of parking lots.

[0206] like Figure 2 The present invention provides a comprehensive identification device for intelligent parking lots, see Figure 2 , Figure 2 A schematic diagram of an intelligent parking lot comprehensive identification device provided by the present invention is shown in FIG. Figure 2 The intelligent parking lot comprehensive identification device shown includes:

[0207] The acquisition unit 21 is used to acquire real-time video data and extract the vehicle area in the real-time video data;

[0208] An identification unit 22 is configured to identify target vehicle information corresponding to the vehicle area in a preset vehicle model database;

[0209] a matching unit 23 configured to match the license plate position information corresponding to the target vehicle information, and extract a license plate region in the vehicle region based on the license plate position information; wherein the license plate position information includes a distance value between the license plate position and each feature region;

[0210] An extraction unit 24 is used to extract the license plate number information in the license plate area;

[0211] The calling unit 25 is used to call the parking process according to the vehicle number information; wherein the parking process includes parking lot admission release or exit fee calculation release.

[0212] The present invention provides an intelligent parking lot comprehensive identification device, which extracts the vehicle area by collecting real-time video data, and then identifies the target vehicle information in a preset vehicle model database. Compared with the traditional single license plate recognition method, this comprehensive identification method can utilize more vehicle feature information, thereby improving the accuracy and robustness of recognition. By matching the license plate position information corresponding to the target vehicle information, the license plate area is extracted in the vehicle area based on the license plate position information. Such a method can effectively reduce recognition errors caused by factors such as ambient light, vehicle angle, and license plate damage, and improve the accuracy of license plate number recognition. By quickly matching vehicle information with a preset vehicle model database, the time required for recognition is greatly shortened, and the overall processing efficiency of the system is improved. The method of the present invention not only relies on license plate information, but also combines the overall vehicle information (such as vehicle model, body features, etc.), thereby achieving more comprehensive and accurate vehicle management. This comprehensive use of information can better prevent problems such as vehicle information falsification or license plate duplication, and improve the safety and reliability of parking lots.

[0213] Figure 3 FIG. 1 is a schematic diagram of an intelligent parking lot comprehensive identification system provided by an embodiment of the present invention. Figure 3 As shown, the intelligent parking lot comprehensive identification system 3 of this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30, such as an intelligent parking lot comprehensive identification program. When the processor 30 executes the computer program 32, the steps of each of the above-mentioned intelligent parking lot comprehensive identification method embodiments are implemented, such as Figure 1 Alternatively, when the processor 30 executes the computer program 32, the functions of each unit in the above-mentioned device embodiments are realized, for example, Figure 2 Function of the unit shown.

[0214] For example, the computer program 32 can be divided into one or more units, which are stored in the memory 31 and executed by the processor 30 to implement the present invention. The one or more units can be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution process of the computer program 32 in the intelligent parking lot integrated identification system 3. For example, the specific functions of each unit of the computer program 32 can be divided as follows:

[0215] An acquisition unit, configured to acquire real-time video data and extract a vehicle region from the real-time video data;

[0216] an identification unit, configured to identify target vehicle information corresponding to the vehicle area in a preset vehicle model database;

[0217] a matching unit, configured to match the license plate position information corresponding to the target vehicle information, and extract a license plate region in the vehicle region based on the license plate position information; wherein the license plate position information includes a distance value between the license plate position and each feature region;

[0218] an extraction unit, configured to extract the license plate number information in the license plate area;

[0219] The calling unit is used to call the parking process according to the vehicle number information; wherein the parking process includes parking lot admission release or exit fee calculation release.

[0220] The intelligent parking lot comprehensive identification system includes but is not limited to a processor 30 and a memory 31. It will be understood by those skilled in the art that Figure 3 It is only an example of an intelligent parking lot comprehensive identification system 3 and does not constitute a limitation of an intelligent parking lot comprehensive identification system 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the intelligent parking lot comprehensive identification system may also include input and output devices, network access devices, buses, etc.

[0221] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0222] The memory 31 can be an internal storage unit of the intelligent parking lot comprehensive identification system 3, such as a hard disk or memory of the intelligent parking lot comprehensive identification system 3. The memory 31 can also be an external storage device of the intelligent parking lot comprehensive identification system 3, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the intelligent parking lot comprehensive identification system 3. Furthermore, the memory 31 can also include both the internal storage unit of the intelligent parking lot comprehensive identification system 3 and an external storage device. The memory 31 is used to store the computer program and other programs and data required by the roaming control device. The memory 31 can also be used to temporarily store data that has been output or is to be output.

[0223] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0224] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0225] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0226] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned various method embodiments can be implemented.

[0227] An embodiment of the present invention provides a computer program product. When the computer program product is run on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0228] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the camera device / intelligent parking lot integrated identification system, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electric carrier signal, a telecommunication signal and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0229] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0230] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0231] In the embodiments provided by the present invention, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0232] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units.

[0233] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0234] It will also be understood that the term "and / or" used in the present description and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0235] As used in the present specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to monitoring," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is monitored" may be interpreted as meaning "upon determination" or "in response to determining" or "upon monitoring [described condition or event]" or "in response to monitoring [described condition or event]," depending on the context.

[0236] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0237] References to "one embodiment" or "some embodiments" in the present specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present invention. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in yet other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0238] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A comprehensive identification method for intelligent parking lots, characterized in that: The intelligent parking lot comprehensive identification method includes: Collecting real-time video data and extracting a vehicle area in the real-time video data; Identifying target vehicle information corresponding to the vehicle area in a preset vehicle model database; Matching the license plate position information corresponding to the target vehicle information, and extracting the license plate area in the vehicle area based on the license plate position information; wherein the license plate position information includes the distance value between the license plate position and each feature area; Extracting license plate number information from the license plate area; The parking process is called according to the vehicle number information; wherein the parking process includes parking lot admission and release or exit fee calculation and release.

2. The intelligent parking lot comprehensive identification method according to claim 1, characterized in that: The step of identifying target vehicle information corresponding to the vehicle area in the preset vehicle model database includes: Acquire multiple feature areas corresponding to different pre-stored vehicle information; wherein the feature areas include a headlight area, a grille area, a hood area, a front bumper area, and a vehicle logo area; Performing contour detection on the vehicle area to obtain a plurality of first closed contour areas; Calculating first similarities between the plurality of first closed contour regions and the plurality of feature regions; Taking a feature region whose first similarity is greater than or equal to a first threshold as a target feature region; Count the number of target feature areas corresponding to each type of vehicle information; When the number of target feature areas corresponding to the current vehicle information is greater than a preset number, the current vehicle information is used as the target vehicle information.

3. The intelligent parking lot comprehensive identification method according to claim 2, characterized in that: The step of calculating the first similarity between the plurality of closed contour regions and the plurality of feature regions comprises: Extracting the center point of the first closed contour area; Calculating a first distance between a plurality of edge pixel points of the first closed contour area and the center point; Taking the edge pixel point corresponding to the maximum first distance as the starting point, extracting multiple target edge pixel points on the edge of the first closed contour area based on a preset sampling frequency; Calculating the second distances between adjacent target edge pixel points according to the original arrangement order of the plurality of target edge pixel points; Normalizing the second distances corresponding to the plurality of target edge pixel points to obtain a feature distance; Combining the plurality of characteristic distances into a current distance characteristic vector according to the original arrangement order of the plurality of target edge pixel points; Extracting a standard distance feature vector corresponding to the feature area; Calculating a first vector similarity between the current distance feature vector and the standard distance feature vector; The first vector similarity is used as the first similarity.

4. The intelligent parking lot comprehensive identification method according to claim 1, characterized in that: The step of matching the license plate position information corresponding to the target vehicle information and extracting the license plate area in the vehicle area based on the license plate position information includes: Matching the license plate position information corresponding to the target vehicle information; The center of the first closed contour area with a first similarity greater than or equal to a first threshold is used as a first anchor point, and the center of the feature area with a first similarity greater than the first threshold is used as a second anchor point; taking a first closed contour region where the first similarity is less than a first threshold as a region to be identified; Calculating third distances between the center of the to-be-identified area and the plurality of first anchor points respectively; Normalize the multiple third distances and construct them into the current distance vector; extracting fourth distances between a plurality of second anchor points and the license plate position from the license plate position information; Normalizing the multiple fourth distances and constructing them into a standard distance vector; wherein the same element positions in the current distance vector and the standard distance vector are used to respectively accommodate the values ​​corresponding to the first anchor point and the second anchor point corresponding to the same first similarity; Calculating a second vector similarity between the current distance vector and the standard distance vector; If the second vector similarity is greater than a second threshold, the area to be identified corresponding to the second vector similarity is used as the license plate area.

5. The intelligent parking lot comprehensive identification method according to claim 1, characterized in that: The step of extracting the license plate number information in the license plate area includes: Performing edge detection on the license plate area to obtain a plurality of second closed contour areas; Acquire pre-stored standard license plate character image data and the license plate character corresponding to the standard license plate character image data; wherein the standard license plate character image data includes Chinese character image data, English character image data and Roman numeral character image data; Extracting the first lengths of the plurality of second closed contour areas, and extracting the second length of the standard license plate character image data; Based on the first length and the second length, the plurality of second closed contour regions are respectively enlarged or reduced in equal proportion to obtain a region to be identified; wherein the current length of the region to be identified is consistent with the second length; Binarize the area to be identified to obtain a first image; Binarizing the standard license plate character image data to obtain a second image; constructing an image coordinate system based on the central axis and midpoint of the first image and the second image; Calculating a difference between a first pixel value corresponding to the first image and a second pixel value corresponding to the second image at the same image position; Counting the number of image positions where the difference is 0; Dividing the number of image positions by the number of pixels in the second image to obtain a first value; If the first value is greater than a third threshold, the license plate character corresponding to the standard license plate character image data corresponding to the first value is used as the license plate information corresponding to the second closed contour area corresponding to the first value; Based on the original arrangement order of the plurality of second closed contour areas in the vehicle area, the plurality of license plate information are combined to obtain the license plate number information.

6. The intelligent parking lot comprehensive identification method according to claim 1, characterized in that: The steps of collecting real-time video data and extracting the vehicle area in the real-time video data include: Acquire a standard video frame; wherein the standard video frame refers to a video frame in which no vehicle exists in the picture; Calculating the number of difference pixels between the real-time video frame and the standard video frame in the real-time video data; wherein the number of difference pixels refers to the number of pixels corresponding to the pixel difference at the same pixel position being greater than a preset value; If the number of the difference pixels exceeds a fourth threshold, obtaining a plurality of subsequent video frames; wherein the subsequent video frames refer to video frames located after the real-time video frame; extracting a foreground region and a background region based on a plurality of the subsequent video frames; If the ratio between the foreground area and the background area exceeds a fifth threshold, the image area formed by the current difference pixel points between the standard video frame and the subsequent video frame is used as the vehicle area; wherein the current difference pixel point refers to the pixel point corresponding to the pixel difference value at the same pixel position is greater than the preset value.

7. The intelligent parking lot comprehensive identification method according to claim 6, characterized in that: The step of extracting the foreground area and the background area according to the plurality of subsequent video frames comprises: In the plurality of subsequent video frames, sequentially calculating current pixel differences between identical pixel positions in adjacent subsequent video frames; Counting a first number of a plurality of current pixel difference values ​​corresponding to the same pixel position that are greater than a sixth threshold; If the first number is 0, the same pixel position is determined as a full background pixel point; If the first number is between the second value and the third value, determining the same pixel position as a half-background pixel point; If the first number is greater than a third value, determining the same pixel position as a full foreground pixel; Obtain a first region consisting of continuous all-foreground pixels, and use the largest first region as the initial region; Acquire a second region consisting of half background pixels; If the initial region is adjacent to the second region, merging the initial region with the second region to obtain the foreground region; The third area formed by all background pixels and the unmerged second area are used as the background area.

8. An intelligent parking lot comprehensive identification device, characterized in that: The intelligent parking lot comprehensive identification device includes: An acquisition unit, configured to acquire real-time video data and extract a vehicle region from the real-time video data; an identification unit, configured to identify target vehicle information corresponding to the vehicle area in a preset vehicle model database; a matching unit, configured to match the license plate position information corresponding to the target vehicle information, and extract a license plate region in the vehicle region based on the license plate position information; wherein the license plate position information includes a distance value between the license plate position and each feature region; an extraction unit, configured to extract the license plate number information in the license plate area; The calling unit is used to call the parking process according to the vehicle number information; wherein the parking process includes parking lot admission release or exit fee calculation release.

9. An intelligent parking lot comprehensive identification system, characterized in that: The intelligent parking lot comprehensive identification system includes: a memory, a processor, and an intelligent parking lot comprehensive identification program stored in the memory and executable on the processor. The intelligent parking lot comprehensive identification program is configured to implement the steps in the intelligent parking lot comprehensive identification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the intelligent parking lot comprehensive identification method according to any one of claims 1 to 7 are implemented.

Images