Vehicle-mounted information declaration method and device and storage medium

Abstract

The embodiment of the invention provides a vehicle-mounted information declaration method and device and a storage medium, and the method comprises the steps: collecting a target picture of an environment where a vehicle is located, processing the target picture through employing a corresponding analysis method based on the picture type of the target picture, and obtaining a first analysis result of the target picture after obtaining the first analysis result of the target picture, and obtaining a second analysis result of the target picture based on the first analysis result and the vehicle information of the vehicle, and reporting the analysis result confirmed by the user. Different picture categories are processed through different analysis methods, and the vehicle information is combined with the first analysis result, so that the accuracy and the recognition efficiency of the picture analysis processing result can be improved, the picture analysis result is perfected, and the interactivity with external environment information interaction is improved.

Description

Technical Field

[0001] The present application relates to the technical field of vehicles, and particularly to a vehicle information declaration method, a vehicle information declaration device, and a corresponding computer-readable storage medium. Background Art

[0002] With the continuous advancement of the overall vehicle function design towards the electrification direction, the driving comfort and convenience have been improved accordingly, and the interaction frequency between the vehicle and the external environment information has gradually increased. In particular, in-vehicle cameras can identify the road environment and then construct a safe space during the vehicle driving process. As an important way of modern information acquisition, the camera recognition function can support a more comprehensive human-machine interaction method.

[0003] However, in the related art, the accuracy and recognition efficiency of identifying two-dimensional code pictures through the camera recognition function are relatively low, the picture parsing result cannot be improved based on the vehicle information, and the automatic declaration process cannot be performed based on the parsing result, resulting in poor interactivity with the external environment information. Therefore, there is still room for improvement in the interactivity of the existing camera recognition function in interacting with the external environment information after obtaining pictures. Summary of the Invention

[0004] In view of the above problems, embodiments of the present application are proposed to provide a vehicle information declaration method, a vehicle information declaration device, and a corresponding computer-readable storage medium that overcome the above problems or at least partially solve the above problems.

[0005] Embodiments of the present application disclose a vehicle information declaration method, and the method includes: Collect a target picture of the environment where the vehicle is located; Process the target picture based on an analysis method corresponding to the picture category of the target picture to obtain a first analysis result of the target picture; Obtain a second analysis result of the target picture based on the first analysis result and the vehicle information of the vehicle; After the user confirms the first analysis result and / or the second analysis result, declare the first analysis result and / or the second analysis result.

[0006] Optionally, the collecting a picture of the environment where the vehicle is located includes: Obtain the current actual speed of the vehicle and the environmental characteristics of the environment; Determine a shooting strategy according to the current actual speed and the environmental characteristics; Collect pictures based on the shooting strategy and determine the picture category of the pictures; When the collected picture category meets a preset condition, determine the picture as the target picture.

[0007] Optionally, the picture categories at least include QR code pictures and warning pictures; Processing the target picture by the parsing method corresponding to the picture category of the target picture to obtain a first parsing result of the target picture includes: When the picture category of the target picture is the QR code picture, determining a random factor and a random function according to the current actual speed of the vehicle; Dividing the QR code picture according to the random function to obtain picture data blocks; Parsing the picture data blocks according to the random factor to obtain parsing results of the picture data blocks; Filling the parsing results to obtain the first parsing result corresponding to the QR code picture; When the picture category of the target picture is the warning picture, obtaining feature information of the warning picture; Obtaining the first parsing result of the warning picture based on the feature information.

[0008] Optionally, the determining a random factor and a random function according to the current actual speed of the vehicle includes: Determining the average number of pixel points of the QR code picture; Determining a random factor according to the average number of pixel points and the current actual speed; Determining the random function based on the random factor.

[0009] Optionally, the QR code picture contains filling data of at least one batch, wherein the filling data of the first batch contains a first preset number of target pictures; The parsing the picture data blocks according to the random factor to obtain the parsing results of the picture data blocks includes: Encoding and marking the picture data blocks of each target picture according to the random factor; Determining a first encoding number of encoding marks that meet the parsing conditions within the first preset number of target pictures; When the first encoding number is equal to the random factor, parsing the picture data blocks corresponding to all the encoding marks to obtain corresponding parsing results; When the first encoding number is less than the random factor, determining the number of target pictures in the next batch of filling data according to the first encoding number, and dividing and determining corresponding encoding marks for the target pictures in the next batch of filling data; Determine whether the picture data blocks corresponding to the unparsed coding marks in the first batch meet the parsing conditions; when the picture data blocks corresponding to the unparsed coding marks in the first batch all meet the parsing conditions, parse the picture data blocks corresponding to all the unparsed coding marks to obtain the corresponding parsing results; otherwise, repeat determining the number of target pictures in the next batch of filling data and dividing them until all the picture data blocks corresponding to the unparsed coding marks meet the parsing conditions.

[0010] Optionally, the parsing conditions include: In the filling data of the same batch, the similarity between the picture data blocks with the same coding mark meets the similarity threshold.

[0011] Optionally, the obtaining the feature information of the warning picture and obtaining the first parsing result of the warning picture based on the feature information includes: Obtain the feature information of the warning picture; Judge the similarity between the feature information and the warning signs in the vehicle information database. If the similarity is greater than the preset threshold, use the warning sign corresponding to the feature information as the parsing result of the warning picture.

[0012] Optionally, the first parsing result at least includes: the parsing result of the QR code picture and the parsing result of the warning picture; The obtaining the second parsing result of the target picture based on the first parsing result and the vehicle information of the vehicle includes: When the first parsing result is the parsing result of the QR code picture, judge whether the vehicle information needs to be input for the parsing result of the QR code picture. If the vehicle information needs to be input, perform data filling on the vehicle information to be input based on the vehicle information database, and then obtain the second parsing result; When the first parsing result is the parsing result of the warning picture, judge whether the parsing result of the warning picture needs to be voice broadcast or traffic accident reporting based on the vehicle information. If the voice broadcast or traffic accident reporting based on the vehicle information is required, use the parsing result combined with the vehicle information as the second parsing result.

[0013] The embodiment of the present application also discloses a vehicle-mounted information reporting device, and the device includes: A target picture acquisition module, configured to acquire target pictures of the environment where the vehicle is located; A parsing module, configured to process the target pictures based on the parsing method corresponding to the picture category of the target pictures to obtain the first parsing result of the target pictures; A reporting module, configured to obtain the second parsing result of the target pictures based on the first parsing result and the vehicle information of the vehicle; An information declaration module, configured to declare the first parsing result and / or the second parsing result after the user confirms the first parsing result and / or the second parsing result. An embodiment of the present application also discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the vehicle information declaration method described in any one of the above is implemented.

[0014] The embodiments of the present application include the following advantages: In the embodiments of the present application, by collecting a target picture of the environment where the vehicle is located, and processing the target picture by using a corresponding parsing method based on the picture category of the target picture, after obtaining the first parsing result of the target picture, the second parsing result of the target picture is obtained based on the parsing result and the vehicle information of the vehicle, and then the parsing result after the user confirmation is declared. By processing different picture categories with different parsing methods and combining the vehicle information with the first parsing result, the accuracy of the picture parsing processing result can be improved, the picture parsing result can be improved, and the interactivity of the interaction with the external environment information can be improved. Description of the Drawings

[0015] Figure 1 is a flowchart of the steps of an embodiment of a vehicle information declaration method of the present application; Figure 2 is an implementation flowchart of a method for determining a target picture shooting strategy provided by an embodiment of the present application; Figure 3 is an implementation flowchart of a first parsing method for a target picture provided by an embodiment of the present application; Figure 4 is an implementation flowchart of a method for determining a second parsing result provided by an embodiment of the present application; Figure 5 is a structural block diagram of an embodiment of a vehicle information declaration device of the present application. Detailed Embodiment

[0016] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0017] With the continuous advancement of the whole vehicle function design towards the electrification direction, the driving comfort and convenience have also been improved, and the interaction frequency between the vehicle and the external environment information has gradually increased. Especially for in-vehicle cameras, they can identify the road environment and then construct a safe space during the vehicle driving process. The camera recognition function, as an important way of modern information collection, can support a more comprehensive human-machine interaction method.

[0018] However, the accuracy of identifying QR code pictures through the camera recognition function in the related art is relatively low, the picture parsing result cannot be improved based on vehicle information, and it cannot perform automatic declaration processing based on the parsing result, resulting in poor interactivity with external environmental information. Therefore, there is still room for improvement in the interactivity of the existing camera recognition function when interacting with external environmental information after obtaining pictures.

[0019] In the embodiments of the present application, by collecting a target picture of the environment where the vehicle is located, processing the target picture using a corresponding parsing method based on the picture category of the target picture, after obtaining the first parsing result of the target picture, obtaining the second parsing result of the target picture based on the parsing result and the vehicle information of the vehicle, and then declaring the parsing result after user confirmation. By processing different picture categories with different parsing methods and combining the vehicle information with the first parsing result, the accuracy of the picture parsing processing result can be improved, the picture parsing result can be improved, and the interactivity of interacting with external environmental information can be improved.

[0020] Refer to Figure 1 , which shows a step flowchart of an embodiment of a vehicle-mounted information declaration method of the present application, specifically including the following steps: Step 101, collect a target picture of the environment where the vehicle is located; When the vehicle is in the low-speed mode, the camera automatically enters visual recognition to scan and recognize pictures of the surrounding environment, combines the surrounding environmental conditions for picture collection, and then obtains the target picture.

[0021] Step 102, process the target picture using a parsing method corresponding to the picture category of the target picture to obtain the first parsing result of the target picture; When identifying pictures, different parsing methods need to be adopted for different categories of pictures. For example, for pictures of people, methods such as face detection, key point localization, and pose estimation need to be used for parsing, and for text pictures, methods such as Optical Character Recognition (OCR) need to be used for parsing. When the parsing method adopted does not correspond to the picture category, problems such as inaccurate parsing results and low efficiency will occur.

[0022] In the embodiments of the present application, when the camera combines the surrounding environmental conditions for picture collection and obtains the target picture, the category of the picture will be determined, and different parsing methods will be used to process the picture based on the picture category, so as to accurately and efficiently obtain the parsing result.

[0023] Step 103, obtain the second parsing result of the target picture based on the first parsing result and the vehicle information of the vehicle; Since the first parsing result is the content obtained through preliminary parsing, this content may be incomplete. In practical applications, after scanning a QR code picture, a preliminary first parsing result of the QR code picture will be obtained. When the QR code picture is a parking lot payment picture, the user needs to manually input and select vehicle information such as the license plate number to complete the first parsing result. When the user inputs and selects vehicle information, the user experience will be reduced, and the interaction experience with the external environment information is relatively poor.

[0024] In the embodiment of the present application, after scanning the QR code picture, the first parsing result can be completed based on the vehicle information stored in the vehicle. When the QR code picture is a parking lot payment picture, data such as the required license plate information can be automatically input, and the picture parsing result can be completed based on the vehicle information, improving the interactivity of the interaction with the external environment information.

[0025] It should be noted that completing the picture parsing result based on the vehicle information can be inputting the required license plate information when the QR code picture is a parking lot payment picture, inputting the required visitor information and other content when the QR code picture is an access registration QR code, or inputting the required accident location, accident vehicle and other content when the picture is a traffic accident warning picture. Corresponding vehicle information can be automatically input according to the specific category of the picture, and the embodiment of the present application does not limit this.

[0026] Step 104, after the user confirms the first parsing result and / or the second parsing result, declare the first parsing result and / or the second parsing result.

[0027] After completing the first parsing result based on the vehicle information, the first parsing result and / or the second parsing result can be automatically declared.

[0028] In practical applications, when the QR code picture is a parking lot payment picture, the required license plate information and other data can be automatically input, and the user is reminded to confirm. After confirmation, the payment declaration is automatically made; when the QR code picture is an access registration QR code, the required visitor information and other content can be automatically input, and the user is reminded to confirm. After confirmation, the access information declaration is automatically made; when the picture is a traffic accident warning picture, the required accident location, accident vehicle and other content can be automatically input, and the user is reminded to confirm. After confirmation, the traffic accident declaration is automatically sent.

[0029] It should be noted that automatic declaration can be performed according to the specific category of the picture and the content of the first parsing result and / or the second parsing result. The declaration object can be the parking lot QR code charging party, the access information registration party, or the traffic accident information registration party. The embodiment of the present application does not limit this.

[0030] In an embodiment of the present application, the vehicle information declaration method provided by the embodiment of the present application collects a target picture of the environment where the vehicle is located, processes the target picture by using a corresponding parsing method based on the picture category of the target picture, after obtaining a first parsing result of the target picture, obtains a second parsing result of the target picture based on the first parsing result and the vehicle information of the vehicle, and then declares the parsing result after user confirmation. By processing different picture categories with different parsing methods and combining the vehicle information with the first parsing result, the accuracy and recognition efficiency of the picture parsing processing result can be improved, the picture parsing result can be improved, and the interactivity of the interaction with the external environment information can be improved.

[0031] In an embodiment of the present application, in step 101, the collecting of the picture of the environment where the vehicle is located includes: Step 1011, obtaining the current actual speed of the vehicle and the environmental characteristics of the environment; Step 1012, determining a shooting strategy according to the current actual speed and the environmental characteristics; Step 1013, collecting a picture based on the shooting strategy and determining the picture category of the picture; Step 1014, when the collected picture category meets a preset condition, determining the picture as the target picture.

[0032] In order to better ensure the efficiency of the camera in collecting pictures and the quality of the pictures, the current actual speed of the vehicle and the environmental characteristics of the environment will be obtained, and then the shooting strategy will be determined according to the current actual speed and the environmental characteristics.

[0033] In a preferred embodiment of the present application, the shooting frequency of the picture can be determined according to the current speed of the vehicle. For example, when the vehicle speed is between 0-5 km / h, the shooting frequency of the photo can be set to 3 pictures per second. When the vehicle speed is between 5-10 km / h, the shooting frequency of the photo can be set to 7 pictures per second. When the vehicle speed is between 10-15 km / h, the shooting frequency of the photo can be set to 13 pictures per second. When the vehicle speed is between 15-20 km / h, the shooting frequency of the photo can be set to 20 pictures per second. Determining the shooting frequency of the picture by the current speed of the vehicle can intelligently balance the data collection efficiency and the system performance. By dynamically adjusting the shooting frequency, unnecessary picture collection can be reduced during high-speed driving to save storage space and computing resources, while the shooting frequency can be increased during low-speed or parking to capture more details, thereby improving the accuracy of data analysis.

[0034] In a preferred embodiment of the present application, environmental features can be obtained through signals from the vehicle's sunlight sensor, windshield wiper sensor, fog lamp, etc. For example, through the sunlight sensor, it can be determined whether the current environment is day or night; through the windshield wiper sensor, it can be determined whether it is raining in the current environment; through the fog lamp, the haze situation in the current environment can be obtained; after obtaining the environmental features, the shooting frequency of the pictures can be adjusted based on the environmental features. If the current weather condition is poor, the shooting frequency of the pictures can be adjusted on the basis of the shooting frequency corresponding to the current vehicle speed. For example, when the actual vehicle speed is 7 km / h and the current environment is night, the shooting frequency can be multiplied by 1.5 on the basis of the original shooting frequency, and the shooting frequency of the pictures can be adjusted from the original 7 pictures per second to 11 pictures per second. It should be noted that the specific adjustment ratio is adjusted according to the actual parsing effect, and the aforementioned sensors or vehicle devices such as fog lamps can be set according to the required environmental features. The embodiments of the present application do not limit this.

[0035] Adjusting the shooting frequency of pictures based on environmental features can intelligently adapt to different scenario requirements. By identifying environmental factors such as light, weather, and road conditions, the shooting frequency is dynamically adjusted. When the light is sufficient, the weather is good, and the road is flat, the shooting frequency is reduced to save resources, while when the light is insufficient, the weather is bad, or the road conditions are complex, the shooting frequency is increased to capture more details, thereby improving the efficiency and accuracy of data collection and ensuring high-quality image data can be provided in various environments.

[0036] When collecting pictures based on the shooting strategy, the category of the collected pictures will be judged. Only when the category of the collected pictures meets the preset conditions, the pictures will be determined as target pictures, and then the target pictures will be processed using the corresponding parsing method to obtain the parsing result.

[0037] Referring to Figure 3 , in an embodiment of the present application, the picture categories at least include two-dimensional code pictures and warning pictures; the step 102 of processing the target pictures based on the parsing method corresponding to the picture category of the target pictures to obtain the first parsing result of the target pictures includes: Step 1021, when the picture category of the target picture is the two-dimensional code picture, determine a random factor and a random function according to the current actual speed of the vehicle; Step 1022, divide the two-dimensional code picture according to the random function to obtain picture data blocks; Step 1023, parse the picture data blocks according to the random factor to obtain the parsing results of the picture data blocks; Step 1024, fill the parsing results to obtain the first parsing result corresponding to the two-dimensional code picture; Step 1025, when the picture category of the target picture is the warning picture, obtain the feature information of the warning picture; Step 1026, obtain the first parsing result of the warning picture based on the feature information.

[0038] In the embodiment of the present application, the picture category can be divided into a two-dimensional code picture and a warning picture. When the target picture is a two-dimensional code picture, the two-dimensional code picture will be segmented and parsed according to the current actual speed of the vehicle and a random function, so as to obtain the picture data blocks of the two-dimensional code picture. Then, based on the picture data blocks, the first parsing result of the two-dimensional code picture is determined; when the target picture is a warning picture, the first parsing result is determined by obtaining the feature information of the warning picture.

[0039] Specifically, when the target picture is a two-dimensional code picture, a random factor is determined according to the current actual speed of the vehicle and the shooting quantity of the current batch of two-dimensional code pictures. A random function is determined according to the random factor. Based on the random function, the number of divisions of the two-dimensional code picture is determined to obtain the corresponding number of picture data blocks, and the picture data blocks are parsed based on the random factor to obtain the parsing results of the picture data blocks. According to the parsing results of each picture data block, the first parsing result corresponding to the final two-dimensional code picture is filled. By processing the two-dimensional code picture in the above recognition manner, the accuracy and recognition efficiency of the two-dimensional code picture can be improved, and then the more accurate parsing result obtained by recognition is declared to improve the interactivity of the information interaction with the external environment.

[0040] When the target picture is a warning picture, the first parsing result can be determined by obtaining the feature information of the warning picture.

[0041] Exemplarily, preprocessing such as normalization, denoising, grayscale conversion, and picture enhancement can be performed on the warning picture. Then, a pre-trained deep learning model is used to extract the picture features of the picture, the extracted picture features are converted into feature vectors, the distances between the feature vectors (such as Euclidean distance, cosine distance, etc.) are calculated, the similarity of the images is judged, and then according to the result of feature matching, the matching score is calculated to judge the similarity of the images.

[0042] In the embodiments of the present application, the warning pictures at least include: traffic warning pictures, vehicle scratch pictures, traffic accident pictures, etc. The embodiments of the present application can not only automatically identify and scan two-dimensional code pictures, but also identify warning pictures, and broadcast the content of the warning pictures to the user when necessary. For example, when the vehicle is parked in a no-parking section, the vehicle can use the photographed no-parking picture as a warning picture, and use the no-parking reminder or the remaining parking time as the first parsing result to prompt the user; when the vehicle is scratched by other vehicles, the scratched vehicle can be used as the first parsing result to remind the user; when a traffic accident occurs on the road section, the traffic accident scene picture can be used as a warning picture, and the traffic accident can be used as the first parsing result to remind the user. It should be noted that for the prompting method of the warning pictures, it can be broadcast to the user through devices such as the central display screen or in-vehicle voice of the vehicle, and the embodiments of the present application do not limit this.

[0043] In an embodiment of the present application, the determining the random factor and the random function according to the current actual speed of the vehicle includes: Determining the average number of pixel points of the two-dimensional code picture; Determining the random factor according to the average number of pixel points and the current actual speed; Determining the random function based on the random factor.

[0044] In practical applications, the styles of different parts of the two-dimensional code picture are different, and the contents contained in each part are also different. For example, the upper left, upper right, and lower left parts of the two-dimensional code picture are positioning patterns, which are composed of three concentric squares, with the outer layer being black, the middle layer being white, and the inner layer being black, and are used to position the position and direction of the two-dimensional code to ensure that the scanning device can correctly identify the two-dimensional code; while the central area of the two-dimensional code picture is data and error correction codes, which are composed of multiple modules, and the data and error correction codes are arranged alternately; store the actual data information and error correction codes to ensure that the data can still be correctly read in case of partial damage; the lower right and upper left parts of the two-dimensional code picture are composed of 18 modules, including 6 data bits and 12 error correction bits, and are used to store the version information of the two-dimensional code to help the scanning device determine the size and capacity of the two-dimensional code.

[0045] Since the vehicle is in motion, selecting only one or a small number of pictures for two-dimensional code picture recognition will result in low accuracy and poor recognition efficiency of the two-dimensional code picture, which is not conducive to interacting with external environmental information.

[0046] In the embodiments of the present application, after determining the two-dimensional code picture to be recognized, different numbers of the pictures will be taken in batches, and the average number of pixel points S of the two-dimensional code picture will be determined according to the number of pictures taken; S = ; Where: is the number of pixel points in the i-th photo taken. represents i photos.

[0047] After that, determine the random factor a according to the average number of pixel points S and the current actual speed V;

[0048] Where: S is the average number of pixel points of the QR code picture, and V is the current actual speed of the current vehicle.

[0049] After obtaining the random factor a, set the random function random(1:a), and determine the number of picture data blocks into which the QR code picture is divided according to the random function random(1:a). Among them, the random function random(1:a) represents randomly selecting a number from the integers from 1 to a. By adopting the above method, the division method of the QR code picture is linked to the vehicle speed and the quality of the photographed picture, so as to ensure that accurate parsing results can be obtained under different driving states of the vehicle and the camera shooting quality, and improve the accuracy and recognition efficiency of the QR code picture. In an embodiment of the present application, the QR code picture includes at least one batch of padding data, where the first batch of padding data includes a first preset number of target pictures; The parsing the picture data block according to the random factor to obtain the parsing result of the picture data block includes: Encoding and marking the picture data blocks of each target picture according to the random factor; Determine the first encoding quantity of the encoding marks that meet the parsing conditions within the first preset number of target pictures; When the first encoding quantity is equal to the random factor, parse the picture data blocks corresponding to all the encoding marks to obtain the corresponding parsing result; When the first encoding quantity is less than the random factor, determine the number of target pictures in the next batch of padding data according to the first encoding quantity, and divide and determine the corresponding encoding marks for the target pictures in the next batch of padding data; Judge whether the picture data blocks corresponding to the unparsed encoding marks in the first batch meet the parsing conditions; when the picture data blocks corresponding to the unparsed encoding marks in the first batch all meet the parsing conditions, parse the picture data blocks corresponding to all the unparsed encoding marks to obtain the corresponding parsing result; otherwise, repeat determining the number of target pictures in the next batch of padding data and performing the division until all the picture data blocks corresponding to the unparsed encoding marks meet the parsing conditions.

[0050] In the embodiments of the present application, pictures of different batches are taken, and the first batch of filled data includes the first preset number of target pictures. For the convenience of description, the first preset number can be set to 20. It should be emphasized that the first preset number is determined according to the picture quality captured by the camera mounted on the vehicle. If the picture quality captured by the camera mounted on the vehicle is higher, the first preset number is smaller; otherwise, it is larger. When the first preset number is set to 20, the i-th photo taken has a maximum value of .

[0051] Specifically, assume that the random factor a is 9, and the integer determined according to the random function is also 9. Then the QR code picture is equally divided into 9 picture data blocks, and the picture data blocks are encoded based on the random factor a to obtain encoding marks , where a represents the number of the picture data block, and i represents the number of the QR code picture. For example, the first picture data block of the first picture taken is encoded as ...... the first picture data block of the 20th is encoded as ; then, for each picture data block, a judgment is made on the parsing condition to determine the number of encoding marks of the picture data blocks that meet the parsing condition. When all the picture data blocks meet the parsing condition, that is, the first encoding number is 9, all the picture data blocks corresponding to the encoding marks are parsed to obtain the corresponding parsing result; When some picture data blocks meet the parsing condition, that is, the first encoding number is less than 9, then the number of unparsed encoding marks and the corresponding picture data blocks in the first batch are determined according to the difference between the first encoding number and the random factor a. Then, according to the number of picture data blocks corresponding to the unparsed encoding marks in the first batch, the number of the same QR code pictures in the filled data obtained by shooting in the second batch is determined. Then, for the same QR code pictures in the filled data obtained by shooting in the second batch, the QR code pictures are divided and encoded in the same way according to the random factor a and the integer determined by the random function in the first batch to obtain the picture data blocks that match the same encoding marks as the picture data blocks corresponding to the unparsed encoding marks in the first batch, and a secondary judgment is made to determine whether all the unparsed picture data blocks meet the parsing condition. If there are still picture data blocks that do not meet the parsing condition, then this step is repeated to determine the number of the same QR code pictures in the filled data obtained by shooting in the third batch, and the picture data blocks that do not meet the parsing condition are repeatedly determined until all the unparsed picture data blocks in the first batch meet the parsing condition, and the picture data blocks that meet the parsing condition are parsed to obtain the corresponding parsing result. Among them, the number j of the same QR code pictures in the filled data obtained by shooting in the second batch determined according to the number of picture data blocks corresponding to the unparsed encoding marks in the first batch can be determined according to the following formula: ; Wherein: a is a determined random factor, and the same random factor and the same number of divisions are adopted for the same QR code image; B is the number of image databases that meet the parsing conditions; is the number of image data blocks corresponding to the unparsed coding marks in the first batch; i is the first preset number preset in the first batch of filling data.

[0052] By repeatedly determining whether the image data blocks meet the parsing conditions, when the parsing results of all the image data blocks cannot be obtained for the QR code images in the filling data taken in the first batch, the number of QR code images in the filling data taken in the next batch is determined based on the number of unparsed image data blocks, which can reduce the number of images taken subsequently, reduce the data storage pressure, greatly improve the accuracy and recognition efficiency of the QR code images, and avoid problems such as incorrect recognition of QR code images.

[0053] In an embodiment of the present application, the parsing conditions include: In the filling data of the same batch, the similarity between the image data blocks with the same coding mark meets the similarity threshold.

[0054] Assume that the first preset number of QR code images in the filling data taken in the first batch is 20, the QR code images are divided into 9 image data blocks, and coding marks are made accordingly, obtaining coding marks , where a represents the a-th image data block, i represents the i-th QR code image, and the same coding mark refers to the image data blocks with the same a. For example, represents the coding mark of the first image data block, i is the number of the QR code image, and the coding mark at least includes: the coding mark of the first image data block of the 1st... ...... the coding mark of the first image data block of the 20th .

[0055] In an embodiment of the present application, after dividing and coding the QR code images in the filling data taken in the first batch, similarity judgment is performed on the image data blocks with the same coding mark within the shooting quantity range corresponding to the first batch. When the similarity between the image data blocks with the same coding mark meets the first similarity threshold, the content with the similarity meeting the similarity threshold is filled, and the filled data is compared with the next sequential image data block for a second time. If the similarity between the filled data and the next sequential image data block meets the second similarity threshold, it is considered that the image data block meets the parsing conditions, and the parsing result of the image data block is determined.

[0056] In practical applications, determining that the similarity between the picture data blocks of the coding marks meets the similarity threshold includes: within the range of 20 two-dimensional code pictures in the same batch, comparing the similarity between the picture data blocks with the coding mark in sequence, and comparing the data of the part where the similarity meets the first similarity threshold with the picture data blocks with the coding mark of the next picture. If the similarity threshold between the data that meets the first similarity threshold and the picture still meets the second similarity threshold, it is considered that the picture data blocks with the coding mark meet the parsing conditions, and the parsing result of the picture data blocks is determined.

[0057] Specifically, when the picture data blocks with the coding mark are compared with the picture data blocks with the coding marks and corresponding to the first picture, the second picture, and the third picture, and the data with a similarity exceeding the first similarity threshold (95%) is obtained, then the data with a similarity exceeding 95% is compared with the picture data blocks with the coding mark of the fourth picture. If the similarity between this part of the data and the picture data blocks with the coding mark meets the second similarity threshold (98%), it is considered that the picture data blocks with the coding mark meet the parsing conditions. According to the picture data blocks with the coding mark the content of the picture data blocks with the coding mark is improved to obtain the parsing result of the picture data blocks with the coding mark . Then, the parsing result is filled into the two-dimensional code picture, and the picture data blocks corresponding to the coding mark are deleted to optimize the storage efficiency and release the data cache space; determine whether the picture data blocks with the coding mark meet the parsing conditions in sequence or simultaneously.

[0058] Optionally, in this embodiment, it is possible to determine whether the picture data blocks meet the parsing conditions in sequence, or it is also possible to determine whether the picture data blocks with multiple coding marks meet the parsing conditions simultaneously. The first similarity threshold and the second similarity threshold can be adjusted according to the vehicle-mounted camera and the parsing effect.

[0059] In an embodiment of the present application, the obtaining the feature information of the warning picture and obtaining the first parsing result of the warning picture based on the feature information includes: Obtaining the feature information of the warning picture; Determine the similarity between the feature information and the warning signs in the vehicle information database. If the similarity is greater than a preset threshold, use the warning sign corresponding to the feature information as the parsing result of the warning picture.

[0060] In an embodiment of the present application, preprocessing such as normalization, denoising, grayscale conversion, and image enhancement can be performed on the warning picture. Then, use a pre-trained deep learning model to extract the picture features of the picture, convert the extracted picture features into feature vectors, calculate the distance between the feature vectors (such as Euclidean distance, cosine distance, etc.), determine the similarity of the images, and then calculate a matching score based on the feature matching result to determine the similarity of the images. When the similarity is greater than the preset threshold, the parsing result corresponding to the picture feature information is reported or reminded to the user.

[0061] In a preferred embodiment of the present application, when the similarity is less than the preset threshold, the second batch of pictures corresponding to the warning picture will be recognized and image enhancement will be performed. Preprocessing such as normalization, denoising, grayscale conversion, and image enhancement will be performed on the warning picture. Then, use a pre-trained deep learning model to extract the picture features of the picture, convert the extracted picture features into feature vectors, calculate the distance between the feature vectors (such as Euclidean distance, cosine distance, etc.), determine the similarity of the images, and then calculate a matching score based on the feature matching result to determine the similarity of the images. When the number of batch pictures with a similarity less than the preset threshold is greater than a certain number, the features of this picture will be extracted and stored in the picture library. For example, when the similarity of 20 consecutive batches of warning pictures is less than the preset threshold, the features of this warning picture will be stored in the picture library, and the content of the picture library will be continuously updated. When a similar warning picture appears next time, it will be parsed to remind the user.

[0062] In an alternative embodiment of the present application, the vehicle information at least includes: personal information of the driver of the vehicle, route information of the vehicle, and accident information of the vehicle.

[0063] In an embodiment of the present application, the vehicle information at least includes personal information of the driver, route information of the vehicle, and accident information, where: the personal information of the driver at least includes the home address and personal identity information; the route information of the vehicle at least includes the starting point of the route and the parking time of the current vehicle; the accident information at least includes the accident location, accident time, and license plate information of the accident vehicle.

[0064] Refer to Figure 4 , in an embodiment of the present application, the first parsing result at least includes: the parsing result of the QR code picture and the parsing result of the warning picture; step 103, obtaining the second parsing result of the target picture based on the first parsing result and the vehicle information of the vehicle, includes: Step 1031, when the first parsing result is the two-dimensional code picture parsing result, determine whether the vehicle information needs to be input for the two-dimensional code picture parsing result. If the vehicle information needs to be input, perform data filling on the vehicle information to be input based on the vehicle information database, and then obtain the second parsing result. Step 1032, when the first parsing result is the warning picture parsing result, determine whether the warning picture parsing result needs to be voice broadcast or traffic accident reported based on the vehicle information. If it needs to be voice broadcast or traffic accident reported based on the vehicle information, use the parsing result combined with the vehicle information as the second parsing result.

[0065] In the embodiment of the present application, when the first parsing result is the two-dimensional code picture parsing result, determine whether vehicle information needs to be input. If vehicle information needs to be input, perform data filling on the vehicle information to be input based on the vehicle information database to obtain the second parsing result, and display the second parsing result to the user. If not, directly display the first parsing result to the user.

[0066] In a preferred embodiment of the present application, when there is no corresponding data in the vehicle information database for filling, the first parsing result will be displayed to the user, and the user will be reminded to input data. The data input by the user will be stored to update the content of the database. Then, when the same two-dimensional code picture is recognized next time, the required data can be automatically filled.

[0067] Specifically, when the two-dimensional code picture is a parking lot payment picture, the required license plate information can be input on the payment interface of the first parsing result to obtain the second parsing result after inputting the license plate. After the user confirms the second parsing result, a parking fee payment declaration is made. When the two-dimensional code picture is an access registration two-dimensional code, the required visitor information and visitor address can be input on the visitor registration interface of the first parsing result to obtain the second parsing result after inputting the visitor information. After the user confirms the second parsing result, a visitor registration declaration is made, which can improve the user experience, avoid manually inputting vehicle information after recognizing the two-dimensional code picture, and reduce the interactivity of interacting with the external environment information.

[0068] In the embodiment of the present application, when the first parsing result is the warning picture parsing result, determine whether the warning picture parsing result needs to be voice broadcast or traffic accident reported based on the vehicle information. If it needs to be voice broadcast or traffic accident reported based on the vehicle information, use the parsing result combined with the vehicle information as the second parsing result and display it to the user. If not, directly display the first parsing result to the user.

[0069] Specifically, when the vehicle is parked in a no-parking section, the vehicle can obtain the first parsing result of the no-parking reminder based on the captured no-parking pictures, or it can also combine the parking time in the vehicle information to determine the remaining parking time, and use the remaining parking time as the second parsing result to broadcast to the user. For example, if the current section has a 10-minute parking limit and it is obtained from the vehicle information that the vehicle has been parked for 5 minutes, then the user can be broadcast "The parking limit time is approaching. There are currently five minutes remaining. Please pay attention to leaving!"; when the picture is a warning picture of a traffic accident, it can directly broadcast a reminder that there is a traffic accident in this section, or it can extract the license plate information of the accident vehicle based on the captured warning picture and the location of the driving section in the vehicle information, and report the specific information of the traffic accident for traffic accident reporting to help the user call the police, thereby increasing the interactivity of the interaction with the external environment information.

[0070] The embodiment of the present application provides a method for reporting vehicle information. By collecting target pictures of the environment where the vehicle is located, using a corresponding parsing method to process the target pictures based on the picture category of the target pictures, after obtaining the first parsing result of the target pictures, obtaining the second parsing result of the target pictures based on the first parsing result and the vehicle information of the vehicle, and then reporting the parsing result after user confirmation. By processing different picture categories with different parsing methods and combining the vehicle information with the first parsing result, the accuracy and recognition efficiency of the picture parsing processing result can be improved, the picture parsing result can be improved, and the interactivity of the interaction with the external environment information can be improved.

[0071] Refer to Figure 5 , which shows the structural block diagram of an embodiment of a vehicle information reporting device of the present application. Specifically, it may include the following modules: The target picture acquisition module 501 is used to acquire target pictures of the environment where the vehicle is located; The first parsing module 502 is used to process the target pictures using the parsing method corresponding to the picture category of the target pictures to obtain the first parsing result of the target pictures; The second parsing module 503 is used to obtain the second parsing result of the target pictures based on the first parsing result and the vehicle information of the vehicle; The information reporting module 504 is used to report the first parsing result and / or the second parsing result after the user confirms the first parsing result and / or the second parsing result.

[0072] In an embodiment of the present application, the in-vehicle information declaration device provided in the embodiment of the present application collects a target picture of the environment where the vehicle is located, processes the target picture by using a corresponding parsing method based on the picture category of the target picture, after obtaining a first parsing result of the target picture, obtains a second parsing result of the target picture based on the parsing result and the vehicle information of the vehicle, and then declares the parsing result after user confirmation. By processing different picture categories with different parsing methods and combining the vehicle information with the first parsing result, the accuracy of the picture parsing processing result can be improved, the picture parsing result can be perfected, and the interactivity of the information interaction with the external environment can be improved.

[0073] For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiment.

[0074] The embodiment of the present application further provides a vehicle, including: a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, it implements each process of the above-mentioned in-vehicle automatic information declaration method embodiment and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0075] The embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor, it implements each process of the above-mentioned in-vehicle information declaration method embodiment and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0076] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, refer to each other.

[0077] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a device, or a computer program product. Therefore, the embodiments of the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0078] Embodiments of the present application are described with reference to the flowcharts and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate means for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0079] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0080] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0081] Although the preferred embodiments of the embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0082] Finally, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions, and corresponding operation entrances are provided for users to choose to authorize or refuse.

[0083] It should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising the said element.

[0084] The above has introduced in detail a vehicle-mounted information declaration method, a vehicle-mounted information declaration device, a corresponding vehicle and a corresponding computer-readable storage medium provided by the present application. Specific examples are used in this text to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A vehicle information reporting method, characterized in that: The method comprises: Collect target images of the vehicle’s environment; Processing the target image based on the parsing method corresponding to the image category of the target image to obtain a first parsing result of the target image; Obtaining a second parsing result of the target image based on the first parsing result and the vehicle information of the vehicle; After the user confirms the first analysis result and / or the second analysis result, the first analysis result and / or the second analysis result is reported.

2. The method according to claim 1, characterized in that The image of the environment where the vehicle is located is collected, including: Acquiring the current actual speed of the vehicle and environmental characteristics of the environment; Determining a shooting strategy according to the current actual speed and the environmental characteristics; Collecting pictures based on the shooting strategy and determining the picture category of the pictures; When the category of the collected picture meets a preset condition, the picture is determined as the target picture.

3. The method according to claim 1, characterized in that The image categories include at least QR code images and warning images; The parsing method corresponding to the image category of the target image processes the target image to obtain a first parsing result of the target image, including: When the image category of the target image is the QR code image, determining a random factor and a random function according to the current actual speed of the vehicle; Dividing the two-dimensional code image according to the random function to obtain image data blocks; Parsing the picture data block according to the random factor to obtain a parsing result of the picture data block; Fill the parsing result to obtain a first parsing result corresponding to the QR code image; When the picture category of the target picture is the warning picture, acquiring feature information of the warning picture; The first analysis result of the warning image is obtained based on the feature information.

4. The method according to claim 3, characterized in that The determining of the random factor and the random function according to the current actual speed of the vehicle comprises: Determine the average number of pixels of the QR code image; Determine a random factor according to the average number of pixels and the current actual speed; The random function is determined based on the random factor.

5. The method according to claim 3, characterized in that: The two-dimensional code image includes at least one batch of filling data, wherein the first batch of filling data includes a first preset number of target images; The step of parsing the picture data block according to the random factor to obtain a parsing result of the picture data block includes: Encoding and marking the picture data blocks of each target picture according to the random factor; Determining a first number of codes of code marks meeting the parsing condition in the first preset number of target pictures; When the first code quantity is equal to the random factor, parsing the picture data blocks corresponding to all the code marks to obtain corresponding parsing results; In a case where the first code quantity is less than the random factor, determining the number of target images in a next batch of filling data according to the first code quantity, and dividing the target images in the next batch of filling data and determining corresponding code marks; Determine whether the picture data blocks corresponding to the unresolved coding tags in the first batch meet the parsing conditions; if the picture data blocks corresponding to the unresolved coding tags in the first batch all meet the parsing conditions, parse all the picture data blocks corresponding to the unresolved coding tags to obtain corresponding parsing results; otherwise, repeatedly determine the number of target pictures in the next batch of filling data and divide them until all the picture data blocks corresponding to the unresolved coding tags meet the parsing conditions.

6. The method according to claim 5, characterized in that The analysis conditions include: In the same batch of padding data, the similarity between the image data blocks with the same encoding mark meets the similarity threshold.

7. The method according to claim 1, characterized in that The obtaining of the characteristic information of the warning picture, and obtaining the first analysis result of the warning picture based on the characteristic information, includes: Obtaining characteristic information of the warning image; The similarity between the characteristic information and the warning sign in the vehicle information database is determined. If the similarity is greater than a preset threshold, the warning sign corresponding to the characteristic information is used as the warning picture analysis result.

8. The method according to claim 1, characterized in that The first analysis result at least includes: a QR code image analysis result and a warning image analysis result; The obtaining a second parsing result of the target image based on the first parsing result and the vehicle information of the vehicle includes: When the first analysis result is the analysis result of the two-dimensional code image, determining whether the two-dimensional code image analysis result needs to input the vehicle information, and if the vehicle information needs to be input, filling the vehicle information that needs to be input based on the vehicle information database, thereby obtaining the second analysis result; When the first analysis result is the analysis result of the warning picture, determine whether the analysis result of the warning picture needs to be voice broadcast based on the vehicle information, or a traffic accident report is required. If a voice broadcast based on the vehicle information, or a traffic accident report is required, the analysis result combined with the vehicle information will be used as the second analysis result.

9. A vehicle information reporting device, characterized in that: The device comprises: A target image acquisition module is used to acquire target images of the environment where the vehicle is located; A parsing module, configured to process the target image based on a parsing method corresponding to the image category of the target image to obtain a first parsing result of the target image; A declaration module, configured to obtain a second analysis result of the target image based on the first analysis result and the vehicle information of the vehicle; An information reporting module is used to report the first analysis result and / or the second analysis result after the user confirms the first analysis result and / or the second analysis result.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the vehicle-mounted information reporting method according to any one of claims 1 to 8 is implemented.

Images