License plate recognition method, system and device for multi-license-plate vehicle and storage medium

Abstract

The invention provides a license plate recognition method, system and device for a multi-license-plate vehicle and a storage medium, and the method comprises the steps: photographing a first image through a first camera, extracting a first feature from the first image, photographing a second image through a second camera, and extracting a second feature from the second image; and comparing the first feature with the second feature, correspondingly outputting the first feature or combining the first feature with the second feature according to a comparison result so as to obtain an accurate result, when only one license plate exists, directly outputting a result of a single license plate, and when two license plates exist, combining and outputting all the license plates, so that misjudgment is avoided, and the accuracy of license plate recognition is improved. And the recognition accuracy is improved. Furthermore, the two cameras are used for cooperative work, license plate recognition and vehicle information recognition are carried out, a final license plate recognition result is confirmed through a matching mechanism, and the double verification mechanism can significantly improve the accuracy of license plate recognition and reduce the misjudgment rate.

Description

Technical Field

[0001] The present application relates to the technical field of license plate recognition, and in particular to a license plate recognition method, system, device and storage medium for vehicles with multiple license plates. Background Art

[0002] The license plate recognition system is an application of computer video image recognition technology in vehicle license plate recognition. License plate recognition is widely used in highway vehicle management. In the electronic toll collection system, it is also the main means of identifying the identity of the vehicle in combination with DSRC technology. The license plate recognition technology is combined with the electronic non-stop toll collection system to identify the vehicle. Passing vehicles do not need to stop when passing through the intersection, which means that automatic vehicle identity recognition and automatic toll collection can be achieved.

[0003] Domestic license plates have different rules in different regions, such as mainland license plates and Hong Kong, Macau and Taiwan license plates. Some vehicles have multiple license plates A, B, and C. When identifying, due to various factors such as angle, light, or different function cameras (for example, camera 1 can only recognize mainland license plates, and camera 2 can recognize mainland license plates and Hong Kong, Macau and Taiwan license plates), the camera only recognizes a certain license plate.

[0004] If a vehicle has multiple license plates, the system will generate multiple different records when processing the vehicle. The vehicle identity cannot be correctly identified, resulting in confusion when the vehicle passes through the identification system. Summary of the invention

[0005] The technical problem to be solved by this application is that if a vehicle has multiple license plates, the system will generate multiple different records when processing the vehicle. The vehicle identity cannot be correctly identified, resulting in confusion in vehicle identity identification when the vehicle passes through the identification system.

[0006] In order to solve the above problems, in order to solve the above technical problems or at least partially solve the above technical problems, the present application provides a license plate recognition method, system, device and storage medium for multi-license plate vehicles.

[0007] In a first aspect, the present invention discloses a license plate recognition method for a vehicle with multiple license plates, which comprises the following steps:

[0008] The vehicle to be detected enters the detection area, the first camera and the second camera are started to shoot, a first image and a second image are obtained, and feature extraction processing is performed on the first image and the second image in turn to obtain a first feature and a second feature; the feature extraction processing includes vehicle feature extraction and license plate feature extraction, the first feature includes a first license plate feature and a first vehicle feature, the second feature includes a second license plate feature and a second vehicle feature, and the vehicle feature extraction includes obtaining the body color, vehicle brand, and vehicle model of the vehicle to be detected;

[0009] The first feature is compared with the second feature, wherein the comparison includes a similarity calculation between the first feature and the second feature, and the first feature or a result of combining the first feature and the second feature is output according to the comparison result.

[0010] Preferably, the step of sequentially performing feature extraction processing on the first image and the second image to obtain the first feature and the second feature specifically includes the following steps:

[0011] Extracting license plate features and vehicle features from the first image to obtain first features, where the first features include first license plate features and first vehicle features;

[0012] Indexing the first feature and the timestamp of acquisition of the first image and temporarily storing them;

[0013] Perform license plate feature extraction and vehicle feature extraction on the second image to obtain second features, where the second features include second license plate features and second vehicle features.

[0014] Preferably, the first feature and the second feature are compared, and the comparison process includes a similarity calculation process between the first feature and the second feature. The first feature or the result of combining the first feature and the second feature is output according to the comparison process result. Specifically, the following steps are performed:

[0015] Calculating the similarity between the first license plate feature and the second license plate feature to obtain a first similarity result;

[0016] Calculating the similarity between the first vehicle feature and the second vehicle feature to obtain a similarity result between the first vehicle feature and the second vehicle feature;

[0017] The similarity between the first feature and the second feature is determined based on the first similarity result and the second similarity result, and the first feature or a result of combining the first feature and the second feature is output based on the determination result.

[0018] Preferably, judging the similarity between the first feature and the second feature according to the first similarity result and the second similarity result, and outputting the first feature or the result of combining the first feature and the second feature according to the judgment result, specifically comprises the following steps:

[0019] When the first similarity result is judged to be the same, the second feature is directly output; when the first similarity result is judged to be different, the second similarity result is judged;

[0020] Calculating similarity one by one according to each parameter of the vehicle feature, combining the similarity results of each parameter to obtain a second similarity result, and judging whether the first vehicle feature and the second vehicle feature are the same according to the second similarity result;

[0021] The second similarity result is judged to be the same, and the first license plate feature and the second license plate feature are combined and output;

[0022] The second similarity result is judged to be different, and is output in combination with the first feature and the second feature.

[0023] Preferably, when the first similarity result is judged to be identical, the second feature is directly outputted, and when the first similarity result is judged to be different, the second similarity result is judged, which specifically includes the following steps:

[0024] The first similarity result is greater than or equal to 9, the first license plate feature is the same as the second license plate feature, and it can be directly determined that the first feature is the same as the second feature;

[0025] The first similarity result is less than 9, and the first license plate feature is different from the second license plate feature.

[0026] Preferably, similarity is calculated one by one according to each parameter of the vehicle feature, and a second similarity result is obtained by combining the similarity results of each parameter, which specifically includes the following steps:

[0027] For the three parameter features of vehicle body color, vehicle brand, and vehicle model in the vehicle features, the first vehicle feature and the second vehicle feature are parameter-by-parameter feature for calculating similarity values ​​to obtain a body color similarity result, a vehicle brand similarity result, and a vehicle model similarity result;

[0028] Take the average of the body color similarity results, vehicle brand similarity results, and vehicle model similarity results. When the average value is greater than or equal to 8, they are the same vehicle. If the average value is less than 8, they are different vehicles.

[0029] Preferably, the following steps are then included:

[0030] The current license plate information is recorded, the first feature, the second feature, the first image, the second image and the acquired timestamp are stored, and uploaded to the upper system.

[0031] In a second aspect, the present invention discloses a license plate recognition system for a vehicle with multiple license plates, which includes the steps of the above-mentioned license plate recognition method for a vehicle with multiple license plates.

[0032] In a third aspect, the present invention discloses an electronic device, which includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0033] Memory, used to store computer programs;

[0034] The processor is used to implement the steps of the license plate recognition method for multiple license plate vehicles when executing the program stored in the memory.

[0035] In a fourth aspect, the present invention discloses a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-mentioned license plate recognition method for vehicles with multiple license plates.

[0036] The above technical solution provided by this application has the following advantages compared with the prior art:

[0037] The present application provides a license plate recognition method, system, device and storage medium for vehicles with multiple license plates. The method uses a first camera to capture a first image and extract a first feature from the first image, then captures a second image from a second camera and extracts a second feature from the second image, performs comparison processing on the first feature and the second feature, and outputs the first feature or combines the first feature with the second feature according to the comparison processing result, so as to obtain an accurate result. When there is only one license plate, the result of the single license plate is directly output, and when there are two license plates, all the license plates are combined and output together to avoid misjudgment and improve the recognition accuracy.

[0038] Furthermore, two cameras are used to work together to perform license plate recognition and vehicle information recognition, and the final license plate recognition result is confirmed through a matching mechanism. This double verification mechanism can significantly improve the accuracy of license plate recognition and reduce the misjudgment rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0041] Figure 1 A flowchart of a license plate recognition method for a vehicle with multiple license plates provided in this application Figure 1 ;

[0042] Figure 2 A flowchart of a license plate recognition method for a vehicle with multiple license plates provided in this application Figure 2 . DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0044] First, see Figure 1-2 The present invention discloses a license plate recognition method for a vehicle with multiple license plates, which comprises the following steps:

[0045] Step S1: When a vehicle to be detected enters a detection area, the first camera and the second camera are started to shoot, a first image and a second image are obtained, and feature extraction processing is performed on the first image and the second image in turn to obtain a first feature and a second feature; the feature extraction processing includes vehicle feature extraction and license plate feature extraction, the first feature includes a first license plate feature and a first vehicle feature, the second feature includes a second license plate feature and a second vehicle feature, and the vehicle feature extraction includes obtaining the body color, vehicle brand, and vehicle model of the vehicle to be detected;

[0046] Step S2: performing a comparison process on the first feature and the second feature, wherein the comparison process includes a similarity calculation process between the first feature and the second feature, and outputting the first feature or a result of combining the first feature and the second feature according to the comparison process result.

[0047] Step S3: Record the current license plate information, store the first feature, the second feature, the first image, the second image and the acquired timestamp, and upload them to the upper system.

[0048] Specifically, in step S1, the vehicle to be detected enters the detection area, the first camera and the second camera are awakened, and start to recognize the license plate. The first camera first shoots the vehicle to be detected, and recognizes the features in the first image obtained by shooting to obtain the first feature. The extraction of vital signs includes the extraction of the vehicle's own features and the extraction of the license plate features. The first image and the first feature are then temporarily cached, and then the second camera shoots the second image, and recognizes the second feature from the second image. Among them, the image recognition algorithm is used to capture features from the image, and any one of convolutional neural network (CNN), scale-invariant feature transform (SIFT), and local binary pattern (LBP) can be used to extract features. Vehicle feature extraction includes obtaining the body color, vehicle brand, and vehicle model of the vehicle to be detected. Among them, the database of the parking lot management system pre-stores the current vehicle brands and vehicle models on the market, which can be obtained by comparing the first image obtained with the data in the database; license plate feature extraction includes obtaining the license plate character arrangement, number of characters, and license plate color. Since the license plate colors used by fuel vehicles and new energy vehicles in the mainland are different, the colors of mainland license plates and Hong Kong, Macao and Taiwan license plates are also different. The power source and location of the vehicle can be identified by the license plate color, and the character arrangement and number of characters of the corresponding type of vehicle can be obtained based on the power source and location of the vehicle obtained by color.

[0049] Specifically, in step S2, the first vehicle feature and the first license plate feature are compared with the second vehicle feature and the second license plate feature, respectively. The comparison process uses each feature to calculate the similarity, obtains the similarity result of each feature, and judges the similarity of each feature according to the similarity. Then, according to the rule of license plate feature priority, when the similarity of the license plate feature is high, the license plate feature is given priority and the second feature is output. If the similarity of the license plate feature is low, that is, the license plates in the first feature and the second feature are different, the average similarity of each feature of the vehicle feature is considered, and whether the vehicles are the same is judged from the result, and the combination result of the first feature and the second feature is output to avoid missing the output license plate feature and ensure accuracy.

[0050] Specifically, in step S3, after the recognition result is output, the recognition result needs to be recorded and uploaded to the system. At the same time, the timestamps of the first image and the second image are recorded and indexed with the vehicle to be detected to facilitate subsequent inspection.

[0051] The present application provides a license plate recognition method, system, device and storage medium for vehicles with multiple license plates, wherein the method uses a first camera to capture a first image and extract a first feature from the first image, then captures a second image from a second camera and extracts a second feature from the second image, performs a comparison process on the first feature and the second feature, and outputs the first feature or combines the first feature with the second feature according to the comparison process result, thereby obtaining an accurate result. When there is only one license plate, the result of a single license plate is directly output, and when there are two license plates, all license plates are combined and output together to avoid misjudgment and improve the accuracy of recognition. The license plate information of the same vehicle is combined together, and the vehicle is indexed to avoid repeated recognition data being stored in the upper system.

[0052] In particular, the first camera and the second camera can perform license plate recognition on all license plates in China and obtain the license plate numbers on the license plates.

[0053] Furthermore, two cameras are used to work together to perform license plate recognition and vehicle information recognition, and the final license plate recognition result is confirmed through a matching mechanism. This double verification mechanism can significantly improve the accuracy of license plate recognition and reduce the misjudgment rate.

[0054] Step S1 specifically includes the following steps:

[0055] Step S11: extracting license plate features and vehicle features from the first image to obtain a first feature, where the first feature includes a first license plate feature and a first vehicle feature;

[0056] Step S12: indexing the first feature and the timestamp of obtaining the first image and temporarily storing them;

[0057] Step S13: extracting license plate features and vehicle features from the second image to obtain second features, where the second features include second license plate features and second vehicle features.

[0058] Specifically, the feature extraction of the first image and the second image both includes vehicle feature extraction and license plate feature extraction. Therefore, when feature extraction is performed on the first image, the first license plate feature and the first vehicle feature will be obtained, and when feature extraction is performed on the second image, the second license plate feature and the second vehicle feature will be obtained. The first camera works first, so the first image is photographed and recognized first, and the first image and the recognition result of the first image are temporarily stored for subsequent comparison processing, and then the features of the second image are extracted.

[0059] Among them, license plate feature extraction includes obtaining license plate characters, character arrangement, number of characters, and license plate color; vehicle feature extraction includes body color, vehicle brand, and vehicle model.

[0060] Step S2 specifically includes the following steps:

[0061] Step S21: Calculate the similarity between the first license plate feature and the second license plate feature to obtain a first similarity result;

[0062] Step S22: Calculating the similarity between the first vehicle feature and the second vehicle feature to obtain a similarity result between the first vehicle feature and the second vehicle feature;

[0063] Step S23: judging the similarity between the first feature and the second feature according to the first similarity result and the second similarity result, and outputting the first feature or the result of combining the first feature and the second feature according to the judgment result.

[0064] Specifically, the license plate is a unique identification of the vehicle. By calculating the similarity of the license plate features, it is possible to very accurately determine whether they are the same vehicle or highly related vehicles. In addition, in addition to the license plate, the vehicle has many other features, such as model, color, etc. Considering the similarity of the overall features of the vehicle, it is possible to more comprehensively determine whether the two vehicles are truly similar, and avoid the possible misjudgment (such as the license plate being stolen) that may occur when relying solely on the license plate. Combining the similarity of the license plate features with the similarity of the overall features of the vehicle for judgment can make more accurate and reliable decisions. This comprehensive judgment method can adapt to various complex actual situations and improve the accuracy and robustness of the recognition system. A threshold is pre-set respectively, and a logical judgment is performed on the first similarity result and the second similarity result. If both meet the threshold requirements, the first feature is judged to be the same as the second feature; if not, it is processed according to the corresponding situation.

[0065] Step S23 specifically includes the following steps:

[0066] Step S231: when the first similarity result is judged to be the same, directly output the second feature; when the first similarity result is judged to be different, then judge the second similarity result;

[0067] Step S232: calculating similarities one by one according to the parameters of the vehicle feature, combining the similarity results of the parameters to obtain a second similarity result, and judging whether the first vehicle feature is the same as the second vehicle feature according to the second similarity result;

[0068] Step S233: the second similarity result is judged to be the same, and the first license plate feature and the second license plate feature are combined and output;

[0069] Step S234: The second similarity result is judged to be different, and the first feature and the second feature are combined and outputted.

[0070] Specifically, the first similarity and the second similarity are judged by first considering the first similarity and then the second similarity. The result of the license plate feature is given priority, and then the comparison result of the vehicle feature is analyzed. When the first similarity result is higher than the predetermined threshold, the second feature can be directly output to reduce the subsequent unnecessary calculation amount and improve the processing efficiency of the system; when the first similarity result is lower than the predetermined threshold, before the first similarity and the second similarity are judged, a threshold needs to be set accordingly to set a reference value to judge whether the first similarity and the second similarity are high or low. The threshold is obtained from a large amount of historical data. In addition, the threshold can be adjusted according to manual input. When the second similarity result is judged to be the same, it can be understood that the vehicle is the same and there are multiple license plates for the vehicle; when the second similarity result is judged to be different, the first feature and the second feature are output at the same time. At this time, the current vehicle needs to be recorded, judged as abnormal, uploaded to the upper system, and the management personnel are notified for manual processing.

[0071] Optionally, when the first similarity result determines that the first license plate feature is the same as the second license plate feature, the first feature or the second feature may be output.

[0072] As an embodiment, step S231 specifically includes the following steps:

[0073] The first similarity result is greater than or equal to 9, the first license plate feature is the same as the second license plate feature, and it can be directly determined that the first feature is the same as the second feature;

[0074] The first similarity result is less than 9, and the first license plate feature is different from the second license plate feature.

[0075] Specifically, the threshold value of the first similarity setting is determined to be 9. By comparing the obtained first similarity result with the threshold value 9, if it is higher than 9 or equal to 9, it can be determined that the vehicles in the first image and the second image are the same vehicle, and the second feature is directly output. When the first similarity result is less than 9, it can be considered that the first license plate feature is different from the second license plate feature, that is, the license plate extracted from the first image is different from the license plate extracted from the second image. It is necessary to determine whether the license plate belongs to the same vehicle through the vehicle features to determine whether a vehicle has multiple license plates.

[0076] As an embodiment, step S232 specifically includes the following steps:

[0077] For the three parameter features of vehicle body color, vehicle brand, and vehicle model in the vehicle features, the first vehicle feature and the second vehicle feature are parameter-by-parameter feature for calculating similarity values ​​to obtain a body color similarity result, a vehicle brand similarity result, and a vehicle model similarity result;

[0078] Take the average of the body color similarity results, vehicle brand similarity results, and vehicle model similarity results. When the average value is greater than or equal to 8, they are the same vehicle. If the average value is less than 8, they are different vehicles.

[0079] Specifically, to determine whether the vehicles in the two images belong to the same vehicle, it is necessary to judge based on the body color, vehicle brand and vehicle model. The brand and license plate are better obtained when they are in the same position. The vehicle model can be obtained by comparison from the preset vehicle database to determine which model the current vehicle belongs to. After the similarity of each parameter in the vehicle characteristics is calculated, the similarity results of each parameter are obtained. It is necessary to take the average value as the second similarity result for subsequent judgment. The preset threshold is 8. The second similarity is compared with the preset threshold to determine whether the vehicles are the same.

[0080] As an embodiment, if the license plate number recognized by the first camera is slightly different from that recognized by the second camera, potential errors can be identified through a character matching algorithm (such as misjudgment of the letter "L" and the number "1", or a car has license plates from the mainland, Hong Kong, Macao and Taiwan, camera 1 misses one of the license plates, and camera 2 recognizes the license plate missed by camera 1). Therefore, by combining the features of the first camera and the second camera, the content missed by the two cameras can be overcome to avoid errors.

[0081] Similarity calculation process: create a two-dimensional array with the data in the first feature and the second feature, then initialize the first column and the first row, calculate the edit distance by dynamic programming, use a double-layer loop to traverse the character combination of the two-dimensional array, determine whether the characters are equal and determine the temporary value, calculate and update the edit distance based on the obtained temporary value, obtain the minimum edit distance, and then calculate the similarity value through the minimum edit distance.

[0082] Specifically, first create a two-dimensional array named dif, the number of rows of which is the length of the string str1 plus 1 (i.e., len1+1), and the number of columns of which is the length of the string str2 plus 1 (i.e., len2+1). Each element dif[i][j] of this two-dimensional array will be used to store the edit distance information from the first i characters of str1 to the first j characters of str2. Initialize the first column, and initialize the first column element dif[a][0] of the two-dimensional array dif through the loop for(int a=0;a<=len1;a++). The initialization value here is a, which means that the number of operations required to change the first a characters of str1 to an empty string is a. From the perspective of operation, this is equivalent to performing a deletion operation. Initialize the first row, and use the loop for(int a=0;a<=len2;a++) to initialize the first row element dif[0][a] of the two-dimensional array dif. The initialization value is a, which means that the number of insertion operations required to change the empty string to the first a characters of str2 is a. Use a double loop, the outer loop for(int i=1;i<=len1;i++) controls the traversal of the string str1 characters, and the inner loop for(intj=1;j<=len2;j++) controls the traversal of the string str2 characters. In this way, each character combination in the two strings can be processed. In each loop, compare the i-1th character of str1 (i.e. str1.charAt(i-1)) and the j-1th character of str2 (i.e. str2.charAt(j-1)). If the two characters are equal, then define a temporary variable temp with a value of 0, indicating that no replacement operation is required; if the two characters are not equal, then the value of temp is 1, indicating that a replacement operation is required, determine whether the characters are equal and determine the temporary value; calculate the edit distance dif[i][j] of the current position based on the temp value obtained previously and the edit distance related value calculated previously. The calculation method is to take the minimum value of the edit distance corresponding to the three operations (replacement, insertion, deletion). Specifically, dif[i-1][j-1]+temp represents the edit distance corresponding to the replacement operation (if temp is 0, it means no replacement is needed, and the previous edit distance is directly taken; if temp is 1, it means replacement is needed, and the edit distance is increased by 1); dif[i][j-1]+1 represents the edit distance corresponding to the insertion operation (insert a character into str2 based on str1, and the edit distance is increased by 1); dif[i-1][j]+1 represents the edit distance corresponding to the deletion operation (delete a character from str1, and the edit distance is increased by 1). Then the minimum of these three values ​​is assigned to dif[i][j].

[0083] Finally, we calculate the similarity operation. First, we find the element dif[len1][len2] in the lower right corner of the two-dimensional array dif. This value represents the minimum edit distance from the entire str1 to the entire str2. We calculate the maximum value of the two string lengths, that is, Math.max(str1.length(),str2.length()). Then we subtract 1 from dif[len1][len2] and divide it by this maximum value, and assign the resulting value to similarity. This similarity is the similarity between the two strings.

[0084] In a second aspect, the present invention discloses a license plate recognition system for a vehicle with multiple license plates, which includes the steps of the above-mentioned license plate recognition method for a vehicle with multiple license plates.

[0085] Specifically, the system implements the method steps provided in the first aspect to realize the recognition of cars with multiple license plates; wherein the method captures a first image through a first camera and extracts a first feature from the first image, then captures a second image through a second camera and extracts a second feature from the second image, performs comparison processing on the first feature and the second feature, and outputs the first feature or combines the first feature with the second feature according to the comparison processing result, so as to obtain an accurate result. When there is only one license plate, the result of a single license plate is directly output, and when there are two license plates, all license plates are output together to avoid misjudgment and improve the recognition accuracy.

[0086] Furthermore, two cameras are used to work together to perform license plate recognition and vehicle information recognition, and the final license plate recognition result is confirmed through a matching mechanism. This double verification mechanism can significantly improve the accuracy of license plate recognition and reduce the misjudgment rate.

[0087] In a third aspect, the present invention discloses an electronic device, which includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0088] Memory, used to store computer programs;

[0089] The processor is used to implement the steps of the license plate recognition method for multiple license plate vehicles when executing the program stored in the memory.

[0090] Specifically, the processor implements the computer program in the memory to implement the method provided in the first aspect, wherein the method captures a first image through a first camera and extracts a first feature from the first image, then captures a second image through a second camera and extracts a second feature from the second image, performs a comparison process on the first feature and the second feature, and outputs the first feature or combines the first feature with the second feature according to the comparison process result, thereby obtaining an accurate result. When there is only one license plate, the result of a single license plate is directly output, and when there are two license plates, all license plates are combined and output together to avoid misjudgment and improve the accuracy of recognition. Furthermore, two cameras are used to work together to perform license plate recognition and vehicle information recognition, and the final license plate recognition result is confirmed through a matching mechanism. This double verification mechanism can significantly improve the accuracy of license plate recognition and reduce the misjudgment rate.

[0091] In a fourth aspect, the present invention discloses a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-mentioned license plate recognition method for vehicles with multiple license plates.

[0092] Specifically, a computer program is stored in the storage medium for implementing the method of the first aspect, so that the computer can implement the method of the first aspect, wherein the method uses a first camera to shoot a first image and extract a first feature from the first image, then shoots a second image from a second camera and extracts a second feature from the second image, performs a comparison process on the first feature and the second feature, and outputs the first feature or combines the first feature with the second feature according to the comparison process result, so as to obtain an accurate result. When there is only one license plate, the result of a single license plate is directly output, and when there are two license plates, all license plates are output together to avoid misjudgment and improve the accuracy of recognition. Furthermore, two cameras are used to work together to perform license plate recognition and vehicle information recognition, and the final license plate recognition result is confirmed through a matching mechanism. This double verification mechanism can significantly improve the accuracy of license plate recognition and reduce the misjudgment rate.

[0093] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0094] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0095] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0096] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be connected, detachably connected, or integrated; it can be mechanically connected or electrically connected; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0097] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0098] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification.

[0099] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0100] The above is a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A license plate recognition method for vehicles with multiple license plates, characterized in that: The following steps are involved: The vehicle to be detected enters the detection area, the first camera and the second camera are started to shoot, a first image and a second image are obtained, and feature extraction processing is performed on the first image and the second image in turn to obtain a first feature and a second feature; the feature extraction processing includes vehicle feature extraction and license plate feature extraction, the first feature includes a first license plate feature and a first vehicle feature, the second feature includes a second license plate feature and a second vehicle feature, and the vehicle feature extraction includes obtaining the body color, vehicle brand, and vehicle model of the vehicle to be detected; The first feature is compared with the second feature, wherein the comparison includes a similarity calculation between the first feature and the second feature, and the first feature or a result of combining the first feature and the second feature is output according to the comparison result.

2. The method according to claim 1, characterized in that: The feature extraction process is performed on the first image and the second image in sequence to obtain the first feature and the second feature, specifically comprising the following steps: Extracting license plate features and vehicle features from the first image to obtain first features, where the first features include first license plate features and first vehicle features; Indexing the first feature and the timestamp of acquisition of the first image and temporarily storing them; Perform license plate feature extraction and vehicle feature extraction on the second image to obtain second features, where the second features include second license plate features and second vehicle features.

3. The method according to claim 1, characterized in that: The first feature is compared with the second feature, and the comparison process includes a similarity calculation process between the first feature and the second feature. The first feature or the result of combining the first feature and the second feature is output according to the comparison process result. Specifically, the following steps are performed: Calculating the similarity between the first license plate feature and the second license plate feature to obtain a first similarity result; Calculating the similarity between the first vehicle feature and the second vehicle feature to obtain a similarity result between the first vehicle feature and the second vehicle feature; The similarity between the first feature and the second feature is determined based on the first similarity result and the second similarity result, and the first feature or a result of combining the first feature and the second feature is output based on the determination result.

4. The method according to claim 3, characterized in that: The method of determining the similarity between the first feature and the second feature according to the first similarity result and the second similarity result, and outputting the first feature or the result of combining the first feature and the second feature according to the determination result, specifically includes the following steps: When the first similarity result is judged to be the same, the second feature is directly output; when the first similarity result is judged to be different, the second similarity result is judged; Calculating similarity one by one according to each parameter of the vehicle feature, combining the similarity results of each parameter to obtain a second similarity result, and judging whether the first vehicle feature and the second vehicle feature are the same according to the second similarity result; The second similarity result is judged to be the same, and the first license plate feature and the second license plate feature are combined and output; The second similarity result is judged to be different, and is output in combination with the first feature and the second feature.

5. The method according to claim 4, characterized in that: When the first similarity result is judged to be identical, the second feature is directly outputted, and when the first similarity result is judged to be different, the second similarity result is judged, which specifically includes the following steps: The first similarity result is greater than or equal to 9, the first license plate feature is the same as the second license plate feature, and it can be directly determined that the first feature is the same as the second feature; The first similarity result is less than 9, and the first license plate feature is different from the second license plate feature.

6. The method according to claim 4, characterized in that: Calculating similarity one by one according to each parameter of the vehicle feature and combining the similarity results of each parameter to obtain a second similarity result specifically includes the following steps: For the three parameter features of vehicle body color, vehicle brand, and vehicle model in the vehicle features, the first vehicle feature and the second vehicle feature are parameter-by-parameter feature for calculating similarity values ​​to obtain a body color similarity result, a vehicle brand similarity result, and a vehicle model similarity result; Take the average of the body color similarity results, vehicle brand similarity results, and vehicle model similarity results. When the average value is greater than or equal to 8, they are the same vehicle. If the average value is less than 8, they are different vehicles.

7. The method according to claim 1, characterized in that: The following steps are then included: The current license plate information is recorded, the first feature, the second feature, the first image, the second image and the acquired timestamp are stored, and uploaded to the upper system.

8. A license plate recognition system for vehicles with multiple license plates, characterized in that: The method comprises the steps of the license plate recognition method for a vehicle with multiple license plates as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the license plate recognition method for multiple license plate vehicles as described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the license plate recognition method for a vehicle with multiple license plates as described in any one of claims 1 to 7 are implemented.

Images