Data security transmission method

Abstract

The present invention relates to the technical field of safe transmission of driving data, and in particular to a method for safe data transmission, comprising: when vehicle position data is located at a warning section and continues to approach, determining a first desensitization trigger time based on driving vector data, and when the first desensitization trigger time is reached, determining to trigger a desensitized driving operation based on current vehicle position information and vehicle speed, thereby obtaining desensitized data; uploading a hash value corresponding to the desensitized data to a backend server of a car company; on the one hand, the present invention desensitizes image data that is originally risky, thereby ensuring the security of the data from the source; on the other hand, uploading and backing up the hash value corresponding to each desensitized fragment, while backing up the data, preventing the car company from obtaining the real image data, so that it is impossible for the car company to repair the confidential desensitized image through image repair technology, thereby effectively preventing data leakage.

Description

Technical Field

[0001] The present invention relates to the technical field of safe transmission of driving data, and in particular to a method for safe transmission of data. Background Art

[0002] At present, with the popularization of new energy electric vehicles, the vision-based assisted driving system has brought great convenience to people's driving operations, but it also brings with it a large amount of driving privacy data (image data captured by the camera) obtained by the car company's back-end server, posing a risk of personal privacy leakage. For some high-density R&D bases or administrative office areas, although the confidentiality of the building images or internal road images in the area is not high, there are high confidentiality requirements for the behavior of the staff in the area, and the driving backup data uploaded by the vehicles in the area to the car company's back-end server cannot contain or involve the behavior image data of the staff in the area. Therefore, how to make new energy vehicles automatically desensitize confidential image data in specific areas to ensure the safe transmission of backup data has become an urgent problem to be solved. Summary of the invention

[0003] The object of the present invention is to provide a data security transmission method, device, equipment and readable storage medium to improve the problem of high training costs for relevant power maintenance personnel caused by the above-mentioned manual identification of discharge signal image data.

[0004] In order to achieve the above objectives, the present application provides the following technical solutions:

[0005] In one aspect, the present application provides a method for secure data transmission, the method comprising:

[0006] Obtain vehicle position data and driving vector data, and when the vehicle position data is located in the warning section, determine whether the current vehicle continues to approach the desensitized area corresponding to the warning section based on the driving vector data and the warning section. If it continues to approach, the time when the vehicle reaches the boundary of the desensitized area is predicted by the Kalman filter algorithm based on the vehicle motion vector trajectory in the past 10-30 seconds, and recorded as the first desensitization trigger time. The desensitized area is the area range where the image data obtained by the vehicle camera needs to be locally desensitized and uploaded to the background server for backup. The warning section is a road located near the desensitized area and connected to the desensitized area;

[0007] When the first desensitization trigger time is reached, it is determined whether to perform a desensitized driving operation based on the current vehicle position information and vehicle speed. If the vehicle position information is within the desensitized area and the vehicle speed is greater than the preset speed threshold, a position correction operation is performed. If the corrected vehicle position information is still within the desensitized area, the desensitized driving operation is performed to remove images of related human parts in the video data taken by the vehicle-mounted camera, thereby obtaining desensitized data.

[0008] The desensitized data is sliced ​​based on the time series to obtain multiple desensitized segments, and hash calculation is performed on each of the desensitized segments in turn to obtain the hash value corresponding to each desensitized segment, and the hash value is uploaded to the car company's backend server for backup processing.

[0009] Optionally, the determining whether to perform a desensitized driving operation based on the current vehicle position information and the vehicle speed, performing a position correction operation if the vehicle position information is within a desensitized area and the vehicle speed is greater than a preset speed threshold, and performing the desensitized driving operation if the corrected vehicle position information is still within the desensitized area, includes:

[0010] When the current vehicle position information is within the desensitized area and the vehicle speed is greater than a preset speed threshold, an identification code corresponding to the desensitized area is obtained, and a corresponding reference feature set is applied to the first server based on the identification code, where the reference feature set is linear contour features of buildings in the desensitized area at multiple different angles;

[0011] Acquire multiple driving environment photos taken by the vehicle-mounted camera within the recent 3-5 seconds, and detect the matching degree between the building outline features in each driving environment photo and the linear outline features in the reference feature set. If the matching degree between the building outline features in any driving environment photo and the linear outline features in the reference feature set is greater than a preset matching threshold, it is determined that the current vehicle is located in the desensitized area;

[0012] If there is no driving environment photo that matches the linear contour features in the reference feature set, it is determined that the current vehicle is not in the desensitized area.

[0013] Optionally, detecting the matching degree between the building contour features in each driving environment photo and the linear contour features in the reference feature set includes:

[0014] Acquire a first driving environment photo, and when the current time zone is in a dark time zone, decompose the first driving environment photo into a first environment reflectance sub-map and a first illumination sub-map, and the first environment reflectance sub-map and the first illumination sub-map share a weight;

[0015] Performing illumination enhancement processing on the first illumination sub-image through the Enhance-Net model, and performing exposure correction on the enhanced first illumination sub-image through the current time node, thereby preventing over-exposure imaging caused by excessive illumination enhancement processing, and recording the corrected first illumination sub-image as the second illumination sub-image;

[0016] The dilated convolution in the first environment reflectance sub-image is used to fuse the shallow and deep information through the long path in the feature enhancement module to highlight the noise features in the first environment reflectance sub-image, and the noise in the first environment reflectance sub-image is removed by wavelet hard threshold filtering to obtain a second environment reflectance sub-image;

[0017] Recombining the second illumination sub-image and the second environment reflectance sub-image to obtain an enhanced first driving environment photo, recorded as a second driving environment photo;

[0018] Based on the feature extraction algorithm, when there are building image features in the second driving environment photo, the LBP features of the building strips are identified and depicted, and the depicted LBP features are matched with each linear contour feature in the reference feature set for similarity.

[0019] The beneficial effects of the present invention are:

[0020] The data security transmission described in the present invention is embodied in two aspects: on the one hand, the originally risky image data is desensitized, so that the risky data itself becomes non-confidential and secure data, and then transmitted, backed up and locally stored, thereby ensuring the security of the data from the root;

[0021] The second aspect is to slice the desensitized data based on time series, perform hash calculations on the desensitized segments after slicing, and upload and backup the hash value corresponding to each desensitized segment. While backing up the data, it prevents car companies from obtaining real image data, so that they cannot repair confidential desensitized images through image repair technology, effectively preventing data leakage.

[0022] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 It is a schematic flow chart of a method for secure data transmission described in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] It should be noted that similar reference numerals or letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0027] Example 1

[0028] like Figure 1 As shown, this embodiment provides a method for secure data transmission, which includes step S1, step S2 and step S3.

[0029] Step S1, obtaining vehicle position data and driving vector data, and when the vehicle position data is located in the warning section, determining whether the current vehicle continues to approach the desensitized area corresponding to the warning section based on the driving vector data and the warning section, if it continues to approach, then the time when the vehicle reaches the boundary position of the desensitized area is predicted by the Kalman filter algorithm based on the vehicle motion vector trajectory in the past 10-30 seconds, and recorded as the first desensitization trigger time, wherein the desensitized area is the area range where the image data acquired by the vehicle camera needs to be locally desensitized and uploaded to the background server for backup, such as government office areas, various military scientific research institutes, or corporate R&D centers with high confidentiality requirements, etc., and the warning section is a road located near the desensitized area and connected to the desensitized area. Prior to this, the GPS data of the areas of relevant enterprises and scientific research units and related connected reasons need to be marked in advance. This embodiment does not make relevant elaborations, and the implementation of this function can be based on existing satellite map data; it can be understood that when a new energy vehicle equipped with a visual intelligent assisted driving function enters the corresponding warning road, the vehicle triggers a warning measure for data transmission security;

[0030] Step S2: when the first desensitization trigger time is reached, determining whether to perform a desensitization driving operation based on the current vehicle position information and vehicle speed;

[0031] There are two bases for whether to perform desensitized driving operations: GPS location data and current vehicle speed. GPS location data is the first parameter to trigger the warning. Vehicle speed is a longitudinal correction to GPS location data, thereby preventing false triggering operations caused by the upper and lower position relationship. The vehicle speed in a closed place should be lower than the preset speed threshold. If it is higher than this threshold, it means that the current vehicle may not be in a closed place:

[0032] Case 1: The GPS location data corresponds to the desensitized area, and the vehicle speed is lower than the set speed threshold. At this time, it is determined to perform the desensitized driving operation;

[0033] Case 2: The GPS location data reflects that the vehicle speed is higher than the preset speed threshold. In this case, the vehicle may be traveling on the expressway below the desensitized area, which causes the detected speed to be higher than the preset threshold. However, it is also possible that the vehicle is indeed in the desensitized area. In this case, further judgment is required based on the data collected by the on-board camera.

[0034] This embodiment will further explain the relevant solutions around the second special case:

[0035] If the vehicle position information is within the desensitized area, and the vehicle speed is greater than the preset speed threshold, a position correction operation is performed; if the corrected vehicle position information is still within the desensitized area, the desensitized driving operation is performed to remove images of human bodies from the video data captured by the on-board camera, thereby obtaining desensitized data. That is, when the vehicle is traveling in the desensitized area, the driving video recorded by the on-board camera will automatically cut out the human body image, and then generate driving video data for storage, thereby desensitizing the data from the source of video generation, thereby effectively ensuring the security of the data source and providing underlying security architecture support for subsequent data security transmission;

[0036] Step S3: slice the desensitized data based on the time series to obtain multiple desensitized segments, and perform hash calculation on each of the desensitized segments in turn to obtain the hash value corresponding to each desensitized segment, and upload the hash value to the car company's backend server for backup processing. It should be noted that the processing method in step S3 is only for desensitized data, and the above processing method is not used for non-desensitized data.

[0037] In this embodiment, data security transmission is reflected in two aspects: on the one hand, the originally risky image data is desensitized, so that the risky data itself becomes non-confidential and secure data, and then it is transmitted, backed up and locally stored, thereby ensuring data security from the root;

[0038] The second aspect is to slice the desensitized data based on time series, perform hash calculations on the desensitized segments after slicing, and upload and backup the hash value corresponding to each desensitized segment. While backing up the data, it prevents car companies from obtaining real image data, so that they cannot repair confidential desensitized images through image repair technology, effectively preventing data leakage.

[0039] Example 2

[0040] This embodiment is based on Embodiment 1 and is used to further illustrate the specific method of the correction operation. Before the explanation, it should be noted that the driving data captured during the correction operation is stored in a temporary storage module. When the correction is completed, the specific processing logic of this part of the data is determined based on the correction result. That is, when the result after correction is still within the desensitized area, the data during the correction period is desensitized. If the result after correction is not within the desensitized area, the data is not processed but directly uploaded to the car company's backend server.

[0041] Step S21, when the current vehicle position information is within the desensitized area and the vehicle speed is greater than a preset speed threshold, obtain an identification code corresponding to the desensitized area, and apply to the first server for a corresponding reference feature set based on the identification code, wherein the reference feature set is a linear contour feature of a building in the desensitized area at multiple different angles;

[0042] Step S22, obtaining a plurality of driving environment photos taken by a vehicle-mounted camera within the recent 3-5 seconds, and detecting the matching degree between the building outline features in each driving environment photo and the linear outline features in the reference feature set; if there is a building outline feature in any driving environment photo and the linear outline feature in the reference feature set, the matching degree is greater than a preset matching threshold, then it is determined that the current vehicle is located in the desensitized area;

[0043] Step S23: If there is no driving environment photo that matches the linear contour feature in the reference feature set, it is determined that the current vehicle is not in the desensitized area.

[0044] Example 3

[0045] This embodiment is used to further explain the operation method of determining the matching degree between the building outline features in the driving environment photo and the linear outline features in the reference feature set, including:

[0046] Step S221, obtaining a first driving environment photo, and when the current time zone is in a dark light time zone, decomposing the first driving environment photo into a first environment reflectance sub-map and a first illumination sub-map, and the first environment reflectance sub-map and the first illumination sub-map share a weight;

[0047] Step S222: performing illumination enhancement processing on the first illumination sub-image through the Enhance-Net model, and performing exposure correction on the enhanced first illumination sub-image through the current time node, thereby preventing over-exposure imaging caused by excessive illumination enhancement processing, and recording the corrected first illumination sub-image as the second illumination sub-image;

[0048] Step S223, performing dilated convolution in the first environment reflectance sub-image through long path fusion of shallow and deep information in the feature enhancement module to highlight the noise features in the first environment reflectance sub-image, and removing the noise in the first environment reflectance sub-image through wavelet hard threshold filtering to obtain a second environment reflectance sub-image;

[0049] Step S224: recombining the second illumination sub-image and the second environment reflectance sub-image to obtain an enhanced first driving environment photo, which is recorded as a second driving environment photo;

[0050] Step S225: Based on the feature extraction algorithm, if there are building image features in the second driving environment photo, identify and depict the LBP features of the building strips, and perform similarity matching between the depicted LBP features and each linear contour feature in the reference feature set.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A data security transmission method, characterized in that: The method comprises: Obtain vehicle position data and driving vector data, and when the vehicle position data is located in the warning section, determine whether the current vehicle continues to approach the desensitization area corresponding to the warning section based on the driving vector data and the warning section. If it continues to approach, the time when the vehicle reaches the boundary of the desensitization area is predicted by the Kalman filter algorithm based on the vehicle motion vector trajectory in the past 10-30 seconds, and recorded as the first desensitization trigger time; When the first desensitization trigger time is reached, whether to perform a desensitized driving operation is determined based on the current vehicle position information and vehicle speed. If the vehicle position information is within the desensitization area and the vehicle speed is greater than a preset speed threshold, a position correction operation is performed. If the corrected vehicle position information is still within the desensitization area, the desensitized driving operation is performed to remove images of human body parts from the video data captured by the vehicle-mounted camera, thereby obtaining desensitized data, which is video data containing human body parts. Slice the desensitized data based on the time series to obtain multiple desensitized segments, perform hash calculations on each of the desensitized segments in turn, obtain the hash value corresponding to each desensitized segment, and upload the hash value to the car company's backend server for backup processing; The steps for determining whether the current vehicle position is within the desensitized area are as follows: When the current vehicle position information is within the desensitized area and the vehicle speed is greater than a preset speed threshold, an identification code corresponding to the desensitized area is obtained, and a corresponding reference feature set is applied to the first server based on the identification code, where the reference feature set is linear contour features of buildings in the desensitized area at multiple different angles; Acquire multiple driving environment photos taken by the vehicle-mounted camera within the recent 3-5 seconds, and detect the matching degree between the building outline features in each driving environment photo and the linear outline features in the reference feature set. If the matching degree between the building outline features in any driving environment photo and the linear outline features in the reference feature set is greater than a preset matching threshold, it is determined that the current vehicle is located in the desensitized area; If there is no driving environment photo that matches the linear contour feature in the reference feature set, it is determined that the current vehicle is not in the desensitized area; The detecting of the matching degree between the building outline features in each driving environment photo and the linear outline features in the reference feature set includes: Acquire a first driving environment photo, and when the current time zone is in a dark time zone, decompose the first driving environment photo into a first environment reflectance sub-map and a first illumination sub-map, and the first environment reflectance sub-map and the first illumination sub-map share a weight; Performing illumination enhancement processing on the first illumination sub-image through the Enhance-Net model, and performing exposure correction on the enhanced first illumination sub-image through the current time node, thereby preventing over-exposure imaging caused by excessive illumination enhancement processing, and recording the corrected first illumination sub-image as the second illumination sub-image; The dilated convolution in the first environment reflectance sub-image is used to fuse the shallow and deep information through the long path in the feature enhancement module to highlight the noise features in the first environment reflectance sub-image, and the noise in the first environment reflectance sub-image is removed by wavelet hard threshold filtering to obtain a second environment reflectance sub-image; Recombining the second illumination sub-image and the second environment reflectance sub-image to obtain an enhanced first driving environment photo, recorded as a second driving environment photo; Based on the feature extraction algorithm, when there are building image features in the second driving environment photo, the LBP features of the building strips are identified and depicted, and the depicted LBP features are matched with each linear contour feature in the reference feature set for similarity.

Images