An automobile EDR data watermarking recording method, reading method and device

Abstract

This invention relates to a method, method, and apparatus for recording and reading automotive EDR data with watermarking. The recording method includes: data segmentation, storing EDR data in segments according to the length of a time window, with each segment containing an odd number of data items; midpoint differencing, retaining the median within each data segment and differencing the other data items with the median to obtain differential data; information embedding, embedding binary watermark information into differential data with a difference result of 0 or 1; incrementing the differential data by 1 for differential data with a difference result greater than 1; and decrementing the differential data by 1 for differential data with a difference result less than 0; and data storage, storing the data after information embedding. This invention embeds identifiable watermark information during data recording and extracts the identifiable watermark information during data reading, thus enabling the determination of whether the EDR recorded data has been tampered with.

Description

Technical Field

[0001] This invention relates to the field of vehicle safety technology, and in particular to a method, method and device for watermarking and recording automotive EDR data. Background Technology

[0002] An automotive Event Data Recorder (EDR) comprehensively records vehicle data related to a collision event. It records continuous data on vehicle operation, such as vehicle speed, engine speed, and coolant temperature, ultimately used for scientific analysis and evidence collection in legal proceedings. Automotive EDR data can objectively determine the cause of a collision, ensuring judicial fairness and protecting public interests. Furthermore, it can provide a basis for improving vehicle safety systems, preventing similar collisions, and enhancing vehicle safety. Because EDR data records have significant legal implications, they differ from ordinary data records, requiring stricter standards for legality and traceability. EDR systems must have robust means to verify the legality and tamper-proof nature of the data they read. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and propose a method, reading method and device for adding watermarks to vehicle EDR data that can switch power gears in advance according to predicted terrain. This method can add watermark information to EDR data and verify the legality of the source of EDR data and whether it has been tampered with during reading, without affecting the accuracy and correctness of data recording.

[0004] The technical solution adopted by this invention to solve its technical problem is:

[0005] Firstly, a method for watermarking automotive EDR data includes:

[0006] Data sharding involves dividing and storing EDR data into shards based on the length of the time window, with each shard containing an odd number of data items.

[0007] Midpoint differencing retains the median within a data slice, and for the remaining data, differs from the median to obtain the differencing data.

[0008] Information embedding: For difference data with a difference result of 0 or 1, embed binary watermark information; for difference data with a difference result greater than 1, increment the difference data by 1; for difference data with a difference result less than 0, decrement the difference data by 1.

[0009] Data storage involves embedding processed information into the data and storing it.

[0010] Preferably, the data stored in the data sharding step is represented as follows:

[0011] ;

[0012] The data obtained after midpoint differencing is represented as follows:

[0013] ;

[0014] in, All are integers; It is the median; It is an odd number.

[0015] Preferably, the specific implementation of the information embedding step is as follows:

[0016] ;

[0017] in, This represents the data after embedding. This is one bit in the watermark information, which can be 1 or 0.

[0018] Secondly, a vehicle EDR data watermarking recording device includes:

[0019] The data sharding unit is used to shard and store EDR data according to the length of the time window, and the number of data stored in each shard is set to an odd number;

[0020] The midpoint difference unit is used to retain the median within a data slice. For other data, the difference is obtained by subtracting the median from the remaining data.

[0021] The information embedding unit is used to embed binary watermark information for difference data with a difference result of 0 or 1; to increment the difference data by 1 for difference data with a difference result greater than 1; and to decrement the difference data by 1 for difference data with a difference result less than 0.

[0022] Data storage unit, used to store information embedded in processed data.

[0023] Preferably, the data stored in the data sharding unit is represented as follows:

[0024] ;

[0025] The data obtained after processing by the midpoint difference unit is represented as follows:

[0026] ;

[0027] in, All are integers; It is the median; It is an odd number.

[0028] The information embedding unit is specifically implemented as follows:

[0029] ;

[0030] in, This represents the data after embedding. This is one bit in the watermark information, which can be 1 or 0.

[0031] Thirdly, a method for reading automotive EDR data, based on the aforementioned method for adding watermarks to automotive EDR data, includes:

[0032] Embedded information extraction: Read the difference point information other than the median from each data segment, and determine whether there is embedded watermark information. If so, extract the corresponding watermark information.

[0033] Data authentication involves combining the watermark information extracted in the embedded information extraction step and determining whether it matches the embedded binary watermark information. If it does, the fragmented data is determined to be legitimate data.

[0034] Data recovery involves removing watermark information or adding / subtracting 1 to other differential data (excluding the median) before adding the median.

[0035] Preferably, the embedded information extraction step specifically includes:

[0036] From each data segment, read the difference point information other than the median. For data with a difference of 0 after embedding, extract the embedded watermark information as 0; for data with a difference of 1 after embedding, extract the embedded watermark information as 0; for data with a difference of -1 after embedding, extract the embedded watermark information as 1; for data with a difference of 2 after embedding, extract the embedded watermark information as 1.

[0037] Preferably, the data recovery step specifically includes:

[0038] Subtract 1 from the difference data greater than 1 and add the median; add 1 to the difference data less than 0 and add the median; add the corresponding watermark information to the difference data equal to 0 and add the median; subtract the corresponding watermark information from the difference data equal to 1 and add the median to restore the original EDR data.

[0039] Fourthly, a vehicle EDR data reading device, based on the aforementioned vehicle EDR data watermarking recording device, includes:

[0040] The embedded information extraction unit is used to read the difference point information other than the median from each data segment. For data with a difference of 0 after embedding, the embedded watermark information is extracted as 0; for data with a difference of 1 after embedding, the embedded watermark information is extracted as 0; for data with a difference of -1 after embedding, the embedded watermark information is extracted as 1; and for data with a difference of 2 after embedding, the embedded watermark information is extracted as 1.

[0041] The data identification unit is used to combine the watermark information extracted by the embedded information extraction unit and determine whether it matches the embedded binary watermark information. If it does, the fragmented data is determined to be legitimate data.

[0042] The data recovery unit is used to subtract 1 from difference data greater than 1 and then add the median; to add 1 to difference data less than 0 and then add the median; to add the corresponding watermark information to difference data equal to 0 and then add the median; and to subtract the corresponding watermark information from difference data equal to 1 and then add the median, thereby restoring the original EDR data.

[0043] According to embodiments of the present invention, the present invention has the following beneficial effects:

[0044] This invention provides a method, method, and apparatus for watermarking automotive EDR data, which can embed watermark information for identification during recording and extract the identifiable watermark information during data reading. By utilizing the integrity of the identification information and the identification information itself, it can determine whether the data has been tampered with and perform source tracing of the data record, while not affecting the precision and accuracy of the data record, thus effectively protecting the legal attributes of the EDR recorded data.

[0045] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments, but the method, method and apparatus for watermarking and recording automotive EDR data of the present invention are not limited to the embodiments. Attached Figure Description

[0046] Figure 1 This is a flowchart of the method for adding watermarks to automotive EDR data according to an embodiment of the present invention;

[0047] Figure 2 This is a structural block diagram of the vehicle EDR data watermarking recording device according to an embodiment of the present invention;

[0048] Figure 3 This is a flowchart of the vehicle EDR data reading method according to an embodiment of the present invention;

[0049] Figure 4 This is a structural block diagram of the automotive EDR data reading device according to an embodiment of the present invention. Detailed Implementation

[0050] The objects and functions of the present invention, as well as the methods for achieving these objects and functions, will be clarified by referring to exemplary embodiments. However, the present invention is not limited to the exemplary embodiments disclosed below; it can be implemented in various forms. The purpose of this specification is merely to help those skilled in the art to comprehensively understand the specific details of the invention.

[0051] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0052] It should be noted that the step labels in the specific implementation are only for making the document look clearer. In actual implementation, the order of the steps can be adjusted according to specific needs.

[0053] See Figure 1 As shown, a method for watermarking automotive EDR data includes:

[0054] S101, Data Sharding: EDR data is divided into shards according to the time window length, and the number of data stored in each shard is set to an odd number;

[0055] S102, Midpoint Differentiation: This method retains the median within the data slice and differs the remaining data with the median to obtain the differenced data.

[0056] S103, Information Embedding: For difference data with a difference result of 0 or 1, embed binary watermark information; for difference data with a difference result greater than 1, increment the difference data by 1; for difference data with a difference result less than 0, decrement the difference data by 1.

[0057] S104, Data storage, which involves embedding information into processed data and storing it.

[0058] In this embodiment, the data stored in the data fragmentation step is represented as follows:

[0059] ;

[0060] The data obtained after midpoint differencing is represented as follows:

[0061] ;

[0062] in, All are integers; It is the median; It is an odd number.

[0063] Specifically, It is generally set to greater than 32, which makes it easier to embed information.

[0064] It should be noted that the data on the vehicle's bus is generally in integer form, and is converted to actual values ​​using multiples and offsets. For example, for common vehicle speed information, if the bus-collected data is 100, and the multiple is 8, then the actual vehicle speed is 100 / 8 = 12.5 km / h. The recorded data above... All are original integers.

[0065] In this embodiment, the information embedding step is implemented as follows:

[0066] ;

[0067] in, This represents the data after embedding. This is one bit in the watermark information, which can be 1 or 0.

[0068] After the watermark information is embedded, the processed... Replace the original data with data of (i = 1, 2, ..., n). Store it.

[0069] Specifically, for continuous data on vehicle operation such as vehicle speed, engine speed, and coolant temperature, the probability of these values ​​remaining in a stable fluctuation state within a time slice is relatively high; therefore, the sequence... In this dataset, there are many values ​​close to 0, providing ample space for embedding watermarks. The identifiable watermark information is embedded into the 0s of the difference data, becoming 0 or -1; the identifiable watermark information is embedded into the 1s of the difference data, becoming 1 or 2; data with a difference greater than 1 is incremented by 1, and data with a difference less than 0 is decremented by 1. This ensures that for all unequal difference data, after watermark embedding or data processing, they will still not be equal, thus facilitating watermark identification and data recovery during reading.

[0070] The following example illustrates this. Assume the differenced data is: {0, -2, 1, 0, ..., 103, ...}

[0071] Recorded as , , , ……, ..., the binary watermark information to be embedded is 100, denoted as b1 = 1, b2 = 0, b3 = 0.

[0072] The first Therefore, it can be embedded in b1, and after embedding... The second one Therefore, it cannot be embedded, so according to the formula, we can obtain... The third Therefore, it can be embedded in b2, and after embedding... The fourth Therefore, it can be embedded in b3, and after embedding... The embedded difference data is {-1, -3, 1, 0, ..., 103, ...}.

[0073] It should be noted that if the embedded watermark length (e.g., if the binary watermark information to be embedded is 100, then the length is 3) is less than the length of the data that can be embedded (e.g., there are 32 data entries in the chip), a prefix and a suffix can be added for differentiation. For example, if the watermark information is b, use XX as the watermark prefix and OO as the watermark suffix to form XXbOO. After converting it into 01-bit binary, perform cyclic embedding until all the differential data that can be embedded has been embedded with watermark information.

[0074] See Figure 2 As shown, a watermarking recording device for automotive EDR data includes:

[0075] Data sharding unit 201 is used to shard and store EDR data according to the time window length, and the number of data stored in each shard is set to an odd number;

[0076] Midpoint difference unit 202 is used to retain the median within the data slice, and to obtain the difference data by differencing the other data with the median.

[0077] The information embedding unit 203 is used to embed binary watermark information for differential data with a difference result of 0 or 1; to increment the differential data by 1 for differential data with a difference result greater than 1; and to decrement the differential data by 1 for differential data with a difference result less than 0.

[0078] Data storage unit 204 is used to store the information embedded in the processed data.

[0079] Specifically, the data stored in the data sharding unit is represented as follows:

[0080] ;

[0081] The data obtained after processing by the midpoint difference unit is represented as follows:

[0082] ;

[0083] in, All are integers; It is the median; It is an odd number.

[0084] The information embedding unit is specifically implemented as follows:

[0085] ;

[0086] in, This represents the data after embedding. This is one bit in the watermark information, which can be 1 or 0.

[0087] The specific implementation of a watermarking recording device for automotive EDR data and the same watermarking recording method for automotive EDR data will not be described again in this embodiment.

[0088] See Figure 3 As shown, a method for reading automotive EDR data, based on the aforementioned method for adding watermarks to automotive EDR data, includes:

[0089] S301, Embedded information extraction: Read the difference point information other than the median from each data segment, and determine whether there is embedded watermark information. If so, extract the corresponding watermark information.

[0090] S302, Data Authentication: Combine the watermark information extracted in the embedded information extraction step and determine whether it matches the embedded binary watermark information. If it does, determine that the fragmented data is legitimate data.

[0091] S303, Data Recovery: Except for the median, remove watermark information or add / subtract 1 to other differential data, and then add the median.

[0092] In this embodiment, the embedded information extraction step specifically includes:

[0093] From each data segment, read the difference point information other than the median. For data with a difference of 0 after embedding, extract the embedded watermark information as 0; for data with a difference of 1 after embedding, extract the embedded watermark information as 0; for data with a difference of -1 after embedding, extract the embedded watermark information as 1; for data with a difference of 2 after embedding, extract the embedded watermark information as 1.

[0094] For the stored differential data {-1, -3, 1, 0, ..., 103, ...,}, based on the above embedded information extraction steps, the watermark information 100 can be extracted and matched with the embedded binary watermark information. Therefore, it can be determined that the source of the EDR data record is legitimate and the data has not been tampered with.

[0095] In this embodiment, the data recovery step specifically includes:

[0096] Subtract 1 from the difference data greater than 1 and add the median; add 1 to the difference data less than 0 and add the median; add the corresponding watermark information to the difference data equal to 0 and add the median; subtract the corresponding watermark information from the difference data equal to 1 and add the median to restore the original EDR data. This is the reverse process of information embedding step 103.

[0097] See Figure 4 As shown, a vehicle EDR data reading device, based on the vehicle EDR data watermarking recording device, includes:

[0098] The embedded information extraction unit 401 is used to read the difference point information other than the median from each data segment. For data with a difference of 0 after embedding, the embedded watermark information is extracted as 0; for data with a difference of 1 after embedding, the embedded watermark information is extracted as 0; for data with a difference of -1 after embedding, the embedded watermark information is extracted as 1; and for data with a difference of 2 after embedding, the embedded watermark information is extracted as 1.

[0099] The data identification unit 402 is used to combine the watermark information extracted by the embedded information extraction unit and determine whether it matches the embedded binary watermark information. If it does, the fragmented data is determined to be legitimate data.

[0100] The data recovery unit 403 is used to subtract 1 from the difference data greater than 1 and then add the median; to add 1 to the difference data less than 0 and then add the median; to add the corresponding watermark information to the difference data equal to 0 and then add the median; and to subtract the corresponding watermark information from the difference data equal to 1 and then add the median, thereby restoring the original EDR data.

[0101] The specific implementation of an automotive EDR data reading device and the same automotive EDR data reading method will not be described again in this embodiment.

[0102] The specific embodiments described above further illustrate the purpose, technical solution, and effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and do not limit the present invention. Modifications, substitutions, and variations made by those skilled in the art under the guidance of the present invention without departing from the spirit and scope of the claims are all within the protection scope of the present invention.

Claims

1. A method for watermarking and recording automotive EDR data, characterized in that, include: Data sharding involves dividing and storing EDR data into shards based on the length of the time window, with each shard containing an odd number of data items. Midpoint differencing retains the median within a data slice, and for the remaining data, differs from the median to obtain the differencing data. Information embedding: For difference data with a difference result of 0 or 1, binary watermark information is embedded; for difference data with a difference result greater than 1, the difference data is incremented by 1. For difference data where the difference result is less than 0, the difference data is decremented by 1; Data storage involves embedding information into processed data and storing it. The data stored in the data sharding step is represented as follows: ； The data obtained after midpoint differencing is represented as follows: ； in, All are integers; It is the median; It is an odd number; The specific implementation of the information embedding step is as follows: ； in, This represents the data after embedding. This is one bit in the watermark information, which can be 1 or 0.

2. A watermarking recording device for automotive EDR data, characterized in that, include: The data sharding unit is used to shard and store EDR data according to the length of the time window, and the number of data stored in each shard is set to an odd number; The midpoint difference unit is used to retain the median within a data slice. For other data, the difference is obtained by subtracting the median from the remaining data. The information embedding unit is used to embed binary watermark information for difference data with a difference result of 0 or 1; and to increment the difference data by 1 for difference data with a difference result greater than 1. For difference data where the difference result is less than 0, the difference data is decremented by 1; Data storage unit, used to store the processed data containing embedded information; The data stored in the data fragment unit is represented as follows: ； The data obtained after processing by the midpoint difference unit is represented as follows: ； in, All are integers; It is the median; It is an odd number; The information embedding unit is specifically implemented as follows: ； in, This represents the data after embedding. This is one bit in the watermark information, which can be 1 or 0.

3. A method for reading automotive EDR data, characterized in that, The method for adding watermarks to automotive EDR data as described in claim 1 includes: Embedded information extraction: Read the difference point information other than the median from each data segment, and determine whether there is embedded watermark information. If so, extract the corresponding watermark information. Data authentication involves combining the watermark information extracted in the embedded information extraction step and determining whether it matches the embedded binary watermark information. If it does, the fragmented data is determined to be legitimate data. Data recovery involves removing watermark information or adding / subtracting 1 to other differential data (excluding the median) before adding the median.

4. The method for reading automotive EDR data according to claim 3, characterized in that, The embedded information extraction step specifically includes: From each data segment, read the difference point information other than the median. For data with a difference of 0 after embedding, extract the embedded watermark information as 0; for data with a difference of 1 after embedding, extract the embedded watermark information as 0; for data with a difference of -1 after embedding, extract the embedded watermark information as 1; for data with a difference of 2 after embedding, extract the embedded watermark information as 1.

5. The method for reading automotive EDR data according to claim 4, characterized in that, The data recovery steps specifically include: Subtract 1 from the difference data greater than 1 and add the median; add 1 to the difference data less than 0 and add the median; add the corresponding watermark information to the difference data equal to 0 and add the median; subtract the corresponding watermark information from the difference data equal to 1 and add the median to restore the original EDR data.

6. A vehicle EDR data reading device, characterized in that, The vehicle EDR data watermarking recording device as described in claim 2 includes: The embedded information extraction unit is used to read the difference point information other than the median from each data segment. For data with a difference of 0 after embedding, the embedded watermark information is extracted as 0; for data with a difference of 1 after embedding, the embedded watermark information is extracted as 0; for data with a difference of -1 after embedding, the embedded watermark information is extracted as 1; and for data with a difference of 2 after embedding, the embedded watermark information is extracted as 1. The data identification unit is used to combine the watermark information extracted by the embedded information extraction unit and determine whether it matches the embedded binary watermark information. If it does, the fragmented data is determined to be legitimate data. The data recovery unit is used to subtract 1 from difference data greater than 1 and then add the median; to add 1 to difference data less than 0 and then add the median; to add the corresponding watermark information to difference data equal to 0 and then add the median; and to subtract the corresponding watermark information from difference data equal to 1 and then add the median, thereby restoring the original EDR data.

Images