Video local sensitive information processing method and system

Abstract

The invention provides a video local sensitive information processing method and system, and relates to the technical field of video desensitization, and the method comprises the steps: extracting each frame of video containing sensitive information as a to-be-processed frame, and carrying out the coding compression of the other frames of video to obtain a coded video frame; processing to obtain original information of each pixel point corresponding to the sensitive information, and encrypting by adopting a preset encryption password to obtain encrypted sensitive data; performing desensitization processing on each pixel point corresponding to the sensitive information to obtain a desensitized video frame, and performing coding compression to obtain desensitized video data; using an encrypted password and encrypted sensitive data obtained by encrypting a preset encryption password by using a public key of a digital certificate as desensitization auxiliary information, and binding the desensitized video data to obtain a desensitized video frame; and packaging the desensitized video frames and the coded video frames according to the frame sequence of the desensitized video frames and the coded video frames in the to-be-processed original video to generate a desensitized video file. The method has the beneficial effects that storage resource occupation is reduced; and hierarchical authority control is realized.

Description

Technical Field

[0001] This invention relates to the field of video desensitization technology, and in particular to a method and system for processing sensitive information in localized parts of a video. Background Technology

[0002] With the rapid development of information technology, video capture equipment has been widely applied in various scenarios, including urban public security monitoring, intelligent monitoring of residential communities, traffic violation capture systems, vehicle dashcams, drone aerial photography equipment, smartphone photography, smart home cameras, enterprise internal monitoring, law enforcement recorders, and video uploading to social media platforms. While these devices bring convenience to production and daily life and provide support for security, they also generate massive amounts of video content containing personal privacy information and sensitive data, such as facial features, vehicle license plates, ID card numbers, bank card information, home addresses, and trade secrets, which require special protection.

[0003] Traditional privacy desensitization methods, such as blurring, mosaicking, and color block overlay, essentially hide sensitive content by directly destroying the original pixel information of sensitive areas. This has an irreversible and fundamental flaw. If the original information needs to be restored due to legitimate needs (such as criminal investigations by public security organs, determination of liability for traffic accidents, evidence collection in commercial disputes, or tracing of emergency events), it cannot be recovered through technical means. This leads to a significant reduction or even complete loss of the evidentiary and tracing value of video data.

[0004] To balance privacy protection and the preservation of original information, existing technologies typically employ a dual-file storage solution, storing both the original and anonymized videos. This means simultaneously storing the unprocessed original video (for retrospective viewing in legitimate scenarios) and the anonymized video (for daily viewing and dissemination). However, dual-file storage effectively doubles the storage capacity of the video data. For large-scale video capture scenarios, the daily video data volume often reaches TB or even PB levels. A dual-file storage solution will drastically increase the procurement costs of storage devices, data center construction costs, and operation and maintenance costs. Furthermore, the dual-file storage solution requires the simultaneous transmission of both the original and anonymized videos. Whether it's data synchronization between local devices or uploading and downloading to cloud storage, this will consume twice the network bandwidth resources, resulting in significantly reduced data transmission efficiency and increased latency.

[0005] To prevent the leakage of sensitive information, some existing technologies employ overall encryption of video files. This involves encrypting the entire video using a preset password, allowing only users with the password to decrypt and view the complete video. However, the security of this overall encryption scheme relies on the confidentiality of a single password. If the password is illegally obtained due to human error, brute-force attacks, or system vulnerabilities, all content of the video will be completely exposed, causing serious privacy breaches and information security incidents. Furthermore, under this overall encryption scheme, users can only obtain either full access or complete inaccessibility, making it impossible to achieve precise access control that allows viewing only non-sensitive portions, thus failing to balance privacy protection and information availability.

[0006] Therefore, there is an urgent need for a video local sensitive information processing solution that can solve the above-mentioned technical defects, effectively protect sensitive privacy information, and take into account the traceability of information, the economy of storage and transmission, and the flexibility of access permissions. Summary of the Invention

[0007] To address the problems existing in the prior art, the present invention provides a method for processing sensitive local information in video, comprising:

[0008] Step S1: Obtain the original video to be processed, extract each frame of the original video containing sensitive information as the frame to be processed, and encode and compress the remaining frames to obtain the encoded video frame.

[0009] Step S2: Process the original information of each pixel corresponding to the sensitive information in each frame to be processed, and encrypt the original information using a preset encryption password to obtain encrypted sensitive data.

[0010] Step S3: Desensitize each pixel corresponding to the sensitive information in each frame to be processed to obtain desensitized video frames, and encode and compress each desensitized video frame to obtain desensitized video data.

[0011] Step S4: The preset encryption password is encrypted using the public key of the digital certificate to obtain the encrypted password. Then, the encrypted password and the encrypted sensitive data are used as desensitization auxiliary information and bound to the corresponding desensitized video data to obtain the desensitized video frame.

[0012] Step S5: Package the desensitized video frames and the encoded video frames according to their frame order in the original video to be processed to generate a desensitized video file.

[0013] Preferably, in step S2, processing to obtain the original information of each pixel corresponding to the sensitive information in each frame to be processed includes:

[0014] Step S21: Obtain the location information of the local region where the sensitive information is located in each of the frames to be processed;

[0015] Step S22: Extract the color information of each pixel contained in the local area based on the location information;

[0016] Step S23: Encode and compress the color information to obtain sensitive area information, and then concatenate the sensitive area information and the corresponding position information to form the original information.

[0017] Preferably, the location information includes the pixel coordinates of the top-left corner pixel of the local region where the sensitive information is located in the corresponding frame to be processed, as well as the length and width of the local region.

[0018] Preferably, in step S2, the original information is encrypted using a symmetric encryption algorithm to obtain the encrypted sensitive data.

[0019] Preferably, in step S4, using the encrypted password and the encrypted sensitive data as de-identification auxiliary information and binding them to the corresponding de-identified video data includes:

[0020] The encrypted password and the encrypted sensitive data are concatenated and written as the desensitization auxiliary information into the SEI type NALU unit of the desensitized video data.

[0021] Preferably, after performing step S5, the process further includes a de-identified file viewing process, including:

[0022] Obtain the de-identified video file, and call a general video player to play the de-identified video frames and the encoded video frames in the frame order.

[0023] Preferably, after performing step S5, the process further includes viewing the original file, including:

[0024] Obtain the desensitized video file, call a dedicated video player to restore the desensitized video frames to the frames to be processed, and play the original video to be processed in the order of the frames.

[0025] Preferably, the process of restoring the desensitized video frame to the frame to be processed includes:

[0026] Step A1: The dedicated video player verifies whether the digital certificate is installed on the currently running computer.

[0027] If not, play the desensitized video frames and the encoded video frames in the frame order, and then exit;

[0028] If so, proceed to step A2;

[0029] Step A2: The dedicated video player reads the desensitization auxiliary information in the desensitized video frames and decodes each of the desensitized video data to obtain decoded desensitized video data;

[0030] Step A3: The dedicated video player uses the private key in the digital certificate to decode the encrypted password in the de-identified auxiliary information to obtain the preset encrypted password;

[0031] Step A4: The dedicated video player uses the preset encryption password to decrypt the encrypted sensitive data to obtain the original information, and uses the original information to perform anti-desensitization processing on the corresponding desensitized video frames to obtain the frame to be processed containing the sensitive information.

[0032] Step A5: The dedicated video player plays the original video to be processed in the order of the frames.

[0033] Preferably, the original information includes the location information of the local region where the sensitive information is located in the corresponding frame to be processed, and its sensitive region information;

[0034] In step A4, the desensitization process involves filling the sensitive region information into the local region corresponding to the location information to restore the frame to be processed containing the sensitive information.

[0035] The present invention also provides a system for processing sensitive information in a video location, which applies the above-described processing method. The processing system includes:

[0036] The video frame extraction module is used to acquire the original video to be processed, extract each frame of the original video containing sensitive information as the frame to be processed, and encode and compress the remaining frames to obtain the encoded video frame.

[0037] The sensitive information processing module is connected to the video frame extraction module. It is used to process the original information of each pixel corresponding to the sensitive information in each frame to be processed, and to encrypt the original information using a preset encryption password to obtain encrypted sensitive data.

[0038] The desensitization processing module is connected to the video frame extraction module. It is used to perform desensitization processing on each pixel corresponding to the sensitive information in each frame to be processed to obtain desensitized video frames, and to encode and compress each desensitized video frame to obtain desensitized video data.

[0039] An encryption module, connected to the sensitive information processing module and the desensitization processing module respectively, is used to encrypt the preset encryption password with the public key of the digital certificate to obtain an encrypted password. Then, the encrypted password and the encrypted sensitive data are used as desensitization auxiliary information and bound to the corresponding desensitized video data to obtain a desensitized video frame.

[0040] The video frame packaging module is connected to the video frame extraction module and the encryption module respectively, and is used to package each of the desensitized video frames and the encoded video frames according to their frame order in the original video to be processed to generate a desensitized video file.

[0041] Preferably, it also includes a video decryption and restoration module, used to obtain the desensitized video file, call a dedicated video player to restore the desensitized video frames to the frames to be processed, and play the original video to be processed in the order of the frames.

[0042] The above technical solution has the following advantages or beneficial effects:

[0043] 1) The desensitized video file of the present invention consists of encoded video frames, desensitized video frames, and bound desensitized auxiliary information. For unauthorized users without corresponding digital certificates, they can only directly view the desensitized screen composed of desensitized video frames and encoded video frames, and cannot obtain sensitive information. However, authorized users holding corresponding digital certificates can decrypt the encrypted password through the private key of the digital certificate, and then use the preset encryption password obtained from the decryption to unlock the encrypted sensitive data, thereby restoring the original pixel information of the sensitive information in the frame to be processed, and realizing the complete viewing of the original video content. This design not only prevents the illegal access of sensitive information, but also meets the legitimate access requirements of the original data, realizes hierarchical permission control, and ensures the balance between data security and legitimate use requirements.

[0044] 2) This invention extracts the original information of the pixels corresponding to the sensitive information in the original video frames containing sensitive information and encrypts and stores it. The remaining non-sensitive frames are directly encoded and compressed. The desensitized video frames are also encoded and compressed before storage. Moreover, the encrypted sensitive data is targeted data of local sensitive areas, rather than the complete original frame data. Compared with the drawback of the dual-file storage scheme which requires twice the storage space, this invention only adds a small amount of desensitization auxiliary information to the desensitized video file, which greatly reduces the storage resource consumption. At the same time, there is no need to carry the complete original video data during transmission. Only a single desensitized video file needs to be transmitted, which effectively reduces bandwidth consumption and transmission latency. It is more suitable for the application needs of large-scale video surveillance, vehicle cameras and other massive data scenarios.

[0045] 3) The system employs a dual encryption logic, using the public key of a digital certificate to encrypt a preset encryption password. The preset encryption password is not stored or transmitted in plaintext. Instead, it is encrypted using the public key of the digital certificate to form an encrypted password, which is then bound and stored together with the encrypted sensitive data as desensitized auxiliary information. Since the digital certificate is based on an asymmetric encryption mechanism, only authorized entities holding the corresponding private key can decrypt the encrypted password. Compared to traditional plaintext passwords or single-password encryption methods, this completely avoids the risk of the password being illegally stolen or cracked during storage, transmission, and sharing. It fundamentally ensures the security of the encrypted sensitive data and prevents sensitive information from being illegally obtained by third parties due to password leakage. Attached Figure Description

[0046] Figure 1 A flowchart illustrating a method for processing local sensitive information in a video, as described in a preferred embodiment of the present invention.

[0047] Figure 2 In a preferred embodiment of the present invention, a flowchart is shown for processing the original information of each pixel corresponding to the sensitive information in each frame to be processed.

[0048] Figure 3 In a preferred embodiment of the present invention, a schematic diagram of the process of restoring the desensitized video frame to the frame to be processed is shown.

[0049] Figure 4 This is a schematic diagram of the structure of a video local sensitive information processing system, which is a preferred embodiment of the present invention. Detailed Implementation

[0050] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The present invention is not limited to this embodiment; other embodiments that conform to the spirit of the present invention may also fall within the scope of the present invention.

[0051] In a preferred embodiment of the present invention, based on the above-mentioned problems existing in the prior art, a method for processing local sensitive information in video is provided, such as... Figure 1 As shown, it includes:

[0052] Step S1: Obtain the original video to be processed, extract each frame of the original video containing sensitive information as the frame to be processed, and encode and compress the remaining frames to obtain the encoded video frame.

[0053] Step S2: Process the original information of each pixel corresponding to the sensitive information in each frame to be processed, and encrypt the original information using a preset encryption password to obtain encrypted sensitive data.

[0054] Step S3: Desensitize each pixel corresponding to the sensitive information in each frame to be processed to obtain desensitized video frames, and encode and compress each desensitized video frame to obtain desensitized video data.

[0055] Step S4: Use the public key of the digital certificate to encrypt the preset encryption password to obtain the encrypted password. Then, use the encrypted password and the encrypted sensitive data as desensitization auxiliary information and bind them to the corresponding desensitized video data to obtain the desensitized video frame.

[0056] Step S5: Package the desensitized video frames and encoded video frames according to their frame order in the original video to be processed to generate a desensitized video file.

[0057] Specifically, in this embodiment, to encrypt and desensitize the original video to be processed, it is first necessary to determine which frames in the original video contain sensitive information, including the time range and regional location information of the sensitive information in the original video. The time range is used to characterize which segment or multiple segments in the original video contain sensitive information, i.e., to filter out the frames to be processed. The regional location information specifically locates which specific area in the frame to be processed contains sensitive information, i.e., to locate the local sensitive area in the frame to be processed, and then desensitization processing is performed on this local sensitive area. The remaining video frames that do not contain sensitive information are directly encoded and compressed using mainstream video encoding algorithms (such as H.264 / H.265) to obtain the encoded video frames. This design abandons the redundant mode of the traditional approach of uniformly processing all video frames, and only performs targeted operations on frames containing sensitive information, reducing data processing volume and storage overhead from the source, while ensuring that the playback quality and storage efficiency of non-sensitive frames are not affected.

[0058] Specifically, different sensitive information location methods are preferred for different sensitive information scenarios to determine which frames in the original video to be processed contain sensitive information. These preferred sensitive information scenarios include:

[0059] 1. Sensitive information scenes in fixed screen areas

[0060] This includes, but is not limited to, fixed-location document display areas in surveillance footage. The location information of these areas can be preset directly, and each frame of the original video to be processed can be located using the same location information.

[0061] 2. Mobile sensitive information scenarios

[0062] This includes, but is not limited to, the faces of passersby and the license plates of passing vehicles. AI artificial intelligence recognition technology, such as deep learning-based target detection algorithms, can be used to track and identify the precise regional location information of the device in each frame of video in real time, ensuring that sensitive information is not missed or mislabeled in dynamic scenes.

[0063] After determining the frame to be processed, in a preferred embodiment of the present invention, such as Figure 2 As shown, the original information of each pixel corresponding to the sensitive information in each frame to be processed is obtained, including:

[0064] Step S21: Obtain the location information of the local region where the sensitive information is located in each frame to be processed;

[0065] Step S22: Extract the color information of each pixel in the local area based on the location information;

[0066] Step S23: Encode and compress the color information to obtain sensitive area information, and then concatenate the sensitive area information and the corresponding position information to form the original information.

[0067] The location information includes the pixel coordinates (x, y) of the top-left corner pixel of the local region where the sensitive information is located in the corresponding frame to be processed, as well as the length m and width n of the local region, which can be represented as (x, y, m, n).

[0068] Based on the aforementioned location information, the color information (such as RGB and YUV format data) of each pixel within the local area is precisely extracted, ensuring that only sensitive area data is extracted and avoiding redundant processing of non-sensitive information. The extracted color information is then encoded and compressed using an image encoding algorithm (such as JPEG / PNG) to obtain a smaller sensitive area information file. This sensitive area information is then concatenated with the corresponding location information (x, y, m, n) to form the complete original information. Through compression, the size of the sensitive area data is significantly reduced, laying the foundation for subsequent additional storage and avoiding extra storage pressure.

[0069] After processing the original information of each pixel corresponding to the sensitive information in each frame to be processed, a symmetric encryption algorithm, such as AES / SM4, is preferably used to encrypt the original information based on a preset encryption password to obtain the encrypted sensitive data. The selection of the symmetric encryption algorithm balances encryption strength and processing efficiency, resisting illegal cracking while being suitable for batch processing scenarios of large-scale video data (such as massive amounts of video from vehicle cameras and urban surveillance).

[0070] Taking a single-frame video as an example, based on the region information (x, y, m, n) obtained from the location, the color information of all pixels within that range is first copied (for subsequent restoration). Then, desensitization methods such as black pixel overlay and mosaic masking are used to replace the original sensitive region pixels, completely hiding the sensitive information, resulting in a desensitized video frame. Subsequently, mature video encoding algorithms such as H.264 / H.265 are preferred for compression of each desensitized video frame to ensure the smooth playback and compatibility of the desensitized video. The pixel information of sensitive areas is preferably compressed using JPEG / PNG image encoding algorithms to minimize data size while ensuring information integrity, so that the addition of auxiliary information to the SEI does not significantly increase the video file size.

[0071] Based on the above-mentioned desensitization processing method, the desensitization processing is directly applied to the pixels in the sensitive area, ensuring that even if unauthorized users obtain the video file, they cannot identify sensitive information. At the same time, the encoding and compression further reduce the data size, taking into account both privacy protection and storage efficiency.

[0072] Furthermore, the preset encryption password is encrypted using the public key of the digital certificate to obtain the encrypted password; then the encrypted password and the encrypted sensitive data are used as desensitization auxiliary information and bound to the corresponding desensitized video data.

[0073] The specific method for using the encrypted password and encrypted sensitive data as de-identification auxiliary information and binding them to the corresponding de-identified video data is as follows:

[0074] The encrypted password and the encrypted sensitive data are concatenated and written as desensitization auxiliary information into the SEI (Supplemental Enhancement Information) type NALU unit of the desensitized video data.

[0075] This design has dual technical advantages: First, as a non-core enhancement information unit in the video coding standard, the SEI does not affect the normal decoding and playback of the video after auxiliary information is written into it (ordinary players can directly ignore the SEI content and play the desensitized image), ensuring video compatibility; second, by using a dual encryption logic of digital certificate public key encryption password combined with symmetric encryption of sensitive data, the preset encryption password is not stored or transmitted in plaintext form, and only authorized users holding the corresponding digital certificate private key can decrypt it, thus fundamentally avoiding the risk of sensitive information leakage caused by password leakage.

[0076] Finally, the anonymized video frames (including SEI auxiliary information) and encoded video frames (non-sensitive frames) are packaged strictly according to their frame order in the original video to be processed using a video packaging library (such as an MP4 packaging tool) to generate the final anonymized video file. This file is a single video carrier, eliminating the need to store the original video or separate sensitive data files. Compared to the traditional dual-file storage scheme of original video and anonymized video, the storage space usage is significantly reduced, while also facilitating transmission, management, and distribution.

[0077] As can be seen, based on the above-mentioned local desensitization scheme, this invention can support both desensitized viewing and original restoration modes:

[0078] 1. Regarding the process of viewing de-identified files, including:

[0079] Obtain the de-identified video file, and use a general video player to play the de-identified and encoded video frames in frame order.

[0080] Specifically, after obtaining the anonymized video file, no special tools are needed; it can be played directly using a general video player (such as Potplayer or VLC). Since the anonymized auxiliary information is stored in SEI-type NALU units, the general player will automatically ignore this information, only decoding and playing the encoded and anonymized video frames, enabling quick viewing of the anonymized content. This is suitable for scenarios such as daily monitoring and public video dissemination where sensitive information is not required.

[0081] 2. Regarding the process of viewing the original file, including:

[0082] Obtain the de-identified video file, call a dedicated video player to restore the de-identified video frames to the frames to be processed, and play the original video to be processed in frame order.

[0083] Furthermore, such as Figure 3 As shown, the process of restoring the desensitized video frame to the frame to be processed includes:

[0084] Step A1: The dedicated video player verifies whether a digital certificate is installed on the currently running computer.

[0085] If not, play the desensitized video frames and the encoded video frames in frame order, then exit;

[0086] If so, proceed to step A2;

[0087] Step A2: The dedicated video player reads the desensitization auxiliary information in the desensitized video frames and decodes each desensitized video data to obtain the decoded desensitized video data.

[0088] Step A3: The dedicated video player uses the private key in the digital certificate to decode the encrypted password in the de-identified auxiliary information to obtain the preset encrypted password;

[0089] Step A4: The dedicated video player uses a preset encryption password to decrypt the encrypted sensitive data to obtain the original information, and uses the original information to perform anti-desensitization processing on the corresponding desensitized video frames to obtain the frames to be processed containing sensitive information.

[0090] Step A5: The dedicated video player plays the original video to be processed in frame order.

[0091] Specifically, in this embodiment, to view the original video containing sensitive information, a dedicated player must be used, and the following steps are required to achieve secure restoration, ensuring that sensitive information is only accessible to authorized users:

[0092] 1) Digital Certificate Verification: After the dedicated player starts, it first verifies whether the currently running computer (or device) has the corresponding digital certificate installed. The digital certificate serves as an authorization credential and is only issued to legitimate users (such as police officers handling cases or authorized enterprise administrators). If the certificate is not installed or is invalid, the player will only play the de-identified video and will refuse to restore sensitive information.

[0093] 2) Video frame decoding and auxiliary information extraction: After the digital certificate verification is passed, the player reads the video data frame by frame and sends the data other than the SEI unit to the decoding library of the video encoding algorithm (such as H.264 / H.265 decoder) to decode and obtain the desensitized video frame data; at the same time, the desensitized auxiliary information (including the encrypted password P2 and the encrypted sensitive data) stored in the SEI type NALU unit is extracted;

[0094] 3) Password Decryption: The player uses the private key corresponding to the digital certificate on the computer to decrypt the encrypted password, obtaining the original preset encrypted password. Due to the use of an asymmetric encryption mechanism, only authorized devices with the corresponding private key can successfully decrypt the password, effectively preventing unauthorized password theft.

[0095] 4) Sensitive data decryption: Using the decrypted password, the encrypted sensitive data is decrypted using the symmetric encryption algorithm (AES / SM4) selected in the encryption process to restore the location information (x,y,m,n) of the sensitive area and the compressed sensitive area image data;

[0096] 5) Sensitive information restoration: The decrypted sensitive area image data is sent to the image decoding library (JPEG / PNG decoder) to restore the original pixel color information. Then, according to the position information (x,y,m,n), the pixel information is accurately filled into the corresponding area of ​​the desensitized video screen to complete the restoration of sensitive information.

[0097] 6) Original video playback: The restored complete video data is displayed in real time on the player interface. By repeating the above steps, the original video containing sensitive information can be played smoothly frame by frame, meeting the needs of legal scenarios such as case investigation and accident identification.

[0098] In summary, during the original file viewing process, precise matching of location information ensures that the restored pixel data is accurately filled into the original sensitive areas, avoiding image misalignment or distortion. Furthermore, from extracting sensitive frames in video processing → encrypting sensitive data → desensitizing encoding → SEI storage → packaging generation, to certificate verification → decryption → restoration → playback during restoration viewing, a complete technical closed loop is formed. This ensures the security of sensitive information throughout the process and achieves data reversibility, solving the pain points of traditional desensitization schemes that either leak privacy or permanently lose information. In addition, the video encoding algorithm (H.264 / H.265), image encoding algorithm (JPEG / PNG), and encryption algorithm (AES / SM4) used in this invention are all industry-standard and compatible with most hardware devices and software environments. It also supports algorithm replacement according to actual needs (such as using the national cryptographic algorithm SM2 to replace digital certificate encryption, and H.266 encoding to improve compression efficiency), possessing good scalability.

[0099] This invention also provides a system for processing sensitive information in localized parts of a video, applying the above-described processing method, such as... Figure 4 As shown, the processing system includes:

[0100] Video frame extraction module 1 is used to acquire the original video to be processed, extract each frame of the original video containing sensitive information as the frame to be processed, and encode and compress the remaining frames to obtain the encoded video frame.

[0101] Sensitive information processing module 2 is connected to video frame extraction module 1. It is used to process the original information of each pixel corresponding to the sensitive information in each frame to be processed, and to encrypt the original information with a preset encryption password to obtain encrypted sensitive data.

[0102] The desensitization processing module 3 is connected to the video frame extraction module 1. It is used to desensitize each pixel corresponding to the sensitive information in each frame to be processed to obtain desensitized video frames, and to encode and compress each desensitized video frame to obtain desensitized video data.

[0103] Encryption module 4 is connected to sensitive information processing module 2 and desensitization processing module 3 respectively. It is used to encrypt the preset encryption password with the public key of the digital certificate to obtain the encrypted password. Then, the encrypted password and the encrypted sensitive data are used as desensitization auxiliary information and bound to the corresponding desensitized video data to obtain the desensitized video frame.

[0104] The video frame packaging module 5 is connected to the video frame extraction module 1 and the encryption module 4 respectively. It is used to package each desensitized video frame and the encoded video frame according to their frame order in the original video to be processed to generate a desensitized video file.

[0105] In a preferred embodiment of the present invention, a video decryption and restoration module 6 is further included, which is used to obtain the desensitized video file, call a dedicated video player to restore the desensitized video frames to the frames to be processed, and play the original video to be processed in frame order.

[0106] The above description is merely a preferred embodiment of the present invention and does not limit the implementation and protection scope of the present invention. Those skilled in the art should realize that any equivalent substitutions and obvious changes made using the content of this specification and illustrations should be included within the protection scope of the present invention.

Claims

1. A method for processing sensitive information in a video segment, characterized in that, include: Step S1: Obtain the original video to be processed, extract each frame of the original video containing sensitive information as the frame to be processed, and encode and compress the remaining frames to obtain the encoded video frame. Step S2: Process the original information of each pixel corresponding to the sensitive information in each frame to be processed, and encrypt the original information using a preset encryption password to obtain encrypted sensitive data. Step S3: Desensitize each pixel corresponding to the sensitive information in each frame to be processed to obtain desensitized video frames, and encode and compress each desensitized video frame to obtain desensitized video data. Step S4: The preset encryption password is encrypted using the public key of the digital certificate to obtain the encrypted password. Then, the encrypted password and the encrypted sensitive data are used as desensitization auxiliary information and bound to the corresponding desensitized video data to obtain the desensitized video frame. Step S5: Package the desensitized video frames and the encoded video frames according to their frame order in the original video to be processed to generate a desensitized video file.

2. The processing method according to claim 1, characterized in that, In step S2, the original information of each pixel corresponding to the sensitive information in each frame to be processed is obtained by processing the following methods: Step S21: Obtain the location information of the local region where the sensitive information is located in each of the frames to be processed; Step S22: Extract the color information of each pixel contained in the local area based on the location information; Step S23: Encode and compress the color information to obtain sensitive area information, and then concatenate the sensitive area information and the corresponding position information to form the original information.

3. The processing method according to claim 2, characterized in that, The location information includes the pixel coordinates of the top-left corner pixel of the local region where the sensitive information is located in the corresponding frame to be processed, as well as the length and width of the local region.

4. The processing method according to claim 1, characterized in that, In step S4, using the encrypted password and the encrypted sensitive data as desensitization auxiliary information and binding them to the corresponding desensitized video data includes: The encrypted password and the encrypted sensitive data are concatenated and written as the desensitization auxiliary information into the SEI type NALU unit of the desensitized video data.

5. The processing method according to claim 1, characterized in that, After performing step S5, the process also includes viewing the de-identified file, including: Obtain the de-identified video file, and call a general video player to play the de-identified video frames and the encoded video frames in the frame order.

6. The processing method according to claim 1, characterized in that, After performing step S5, the process also includes viewing the original file, including: Obtain the desensitized video file, call a dedicated video player to restore the desensitized video frames to the frames to be processed, and play the original video to be processed in the order of the frames.

7. The processing method according to claim 6, characterized in that, The process of restoring the desensitized video frame to the frame to be processed includes: Step A1: The dedicated video player verifies whether the digital certificate is installed on the currently running computer. If not, play the desensitized video frames and the encoded video frames in the frame order, and then exit; If so, proceed to step A2; Step A2: The dedicated video player reads the desensitization auxiliary information in the desensitized video frames and decodes each of the desensitized video data to obtain decoded desensitized video data; Step A3: The dedicated video player uses the private key in the digital certificate to decode the encrypted password in the de-identified auxiliary information to obtain the preset encrypted password; Step A4: The dedicated video player uses the preset encryption password to decrypt the encrypted sensitive data to obtain the original information, and uses the original information to perform anti-desensitization processing on the corresponding desensitized video frames to obtain the frame to be processed containing the sensitive information. Step A5: The dedicated video player plays the original video to be processed in the order of the frames.

8. The processing method according to claim 7, characterized in that, The original information includes the location information of the local region where the sensitive information is located in the corresponding frame to be processed, and its sensitive region information. In step A4, the desensitization process involves filling the sensitive region information into the local region corresponding to the location information to restore the frame to be processed containing the sensitive information.

9. A system for processing sensitive local information in video, characterized in that, The processing system, employing the processing method as described in any one of claims 1-8, comprises: The video frame extraction module is used to acquire the original video to be processed, extract each frame of the original video containing sensitive information as the frame to be processed, and encode and compress the remaining frames to obtain the encoded video frame. The sensitive information processing module is connected to the video frame extraction module. It is used to process the original information of each pixel corresponding to the sensitive information in each frame to be processed, and to encrypt the original information using a preset encryption password to obtain encrypted sensitive data. The desensitization processing module is connected to the video frame extraction module. It is used to perform desensitization processing on each pixel corresponding to the sensitive information in each frame to be processed to obtain desensitized video frames, and to encode and compress each desensitized video frame to obtain desensitized video data. An encryption module, connected to the sensitive information processing module and the desensitization processing module respectively, is used to encrypt the preset encryption password with the public key of the digital certificate to obtain an encrypted password. Then, the encrypted password and the encrypted sensitive data are used as desensitization auxiliary information and bound to the corresponding desensitized video data to obtain a desensitized video frame. The video frame packaging module is connected to the video frame extraction module and the encryption module respectively, and is used to package each of the desensitized video frames and the encoded video frames according to their frame order in the original video to be processed to generate a desensitized video file.

10. The processing system according to claim 9, characterized in that, It also includes a video decryption and restoration module, which is used to obtain the desensitized video file, call a dedicated video player to restore the desensitized video frames to the frames to be processed, and play the original video to be processed in the order of the frames.

Images