CN115810160A - Automatic desensitization processing method and system for sensitive information in road traffic video - Google Patents
Automatic desensitization processing method and system for sensitive information in road traffic video Download PDFInfo
- Publication number
- CN115810160A CN115810160A CN202211340842.2A CN202211340842A CN115810160A CN 115810160 A CN115810160 A CN 115810160A CN 202211340842 A CN202211340842 A CN 202211340842A CN 115810160 A CN115810160 A CN 115810160A
- Authority
- CN
- China
- Prior art keywords
- video frame
- sensitive
- key
- information
- key video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method and a system for automatically desensitizing sensitive information in a road traffic video, wherein the method comprises the following steps: extracting the acquired road traffic video frame data to acquire extracted key video frames; extracting character information in each key video frame, and matching in a sensitive information base to obtain sensitive character information in the key video frames; processing the sensitive character information in each key video frame and removing the front and rear video frames arranged according to the time sequence; and recombining the key video frames subjected to sensitive character information processing and the remaining video frames which are not removed into a new desensitized video frame file. Compared with the traditional video desensitization, the traditional road traffic video sensitive information desensitization needs manual interaction, filtering, screening and removing are carried out frame by frame, and manpower time is consumed; by extracting the video key frames, the processing quantity of the video frames is greatly reduced, and the processing speed is increased.
Description
Technical Field
The invention relates to the field of video data processing, in particular to a method and a system for automatically desensitizing sensitive information in a road traffic video.
Background
Countries have confidential requirements for map offerings, and public maps have to mark military facilities and sensitive content. And strict requirements are provided for confidential results. The map is examined before being published. The disclosure or leakage of the confidential map will cause serious harm, and may cause serious threat to the safety of national important military facilities and the safety of national security guard targets and facilities. Even the reliability of measures for protecting the national secrets is reduced or disabled, the local military defense capability and the effectiveness of the enemy of important weapons and equipment are weakened, and the national military facilities and the important engineering safety are threatened. More seriously, the national security, the complete territory and the national dignity can be seriously damaged, thereby causing serious disputes of external transactions, seriously threatening the safety of national defense strategies or weakening the overall military defense capability of the country. The video collected by the high-precision map making map collecting vehicle needs desensitization of sensitive information, and the video information after desensitization can be provided for map operators to operate. In order to reduce the consumption of manual operation, the invention provides an automatic desensitization system for sensitive information of road traffic videos in map making, which is used for directly desensitizing videos sent back by a map acquisition vehicle without manual interaction.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a method and a system for automatically desensitizing sensitive information in a road traffic video.
According to the first aspect of the invention, a method for automatically desensitizing sensitive information in road traffic video is provided, which comprises the following steps:
acquiring road traffic video frame data, extracting the road traffic video frame data, and acquiring an extracted key video frame;
extracting character information in each key video frame, matching in a sensitive information base to obtain sensitive character information in the key video frames and marking at least one key video frame comprising the sensitive character information;
processing the sensitive character information in each key video frame containing the sensitive character information, and removing the front and rear video frames arranged according to a time sequence;
and recombining the key video frames subjected to sensitive text information processing and the remaining video frames which are not removed into a new desensitized video frame file.
On the basis of the technical scheme, the invention can be improved as follows.
Optionally, the extracting the road traffic video frame data to obtain the extracted key video frame includes:
extracting road traffic video frame images with better quality based on a time sequence according to the brightness, the definition and the stain information of the video frame images;
according to the track point information of the collected videos during the running of the vehicle road, clustering video frame images with better quality between two adjacent track points, classifying similar video frame images with continuous time into the same class, and classifying non-similar video frame images into different classes;
and extracting key video frames from a plurality of video frame images in the same class, and forming the extracted key video frames and the non-similar video frame images into key video frames after first extraction.
Optionally, the clustering the video frame images with better quality between two adjacent track points, classifying similar video frame images with continuous time into the same class, and classifying non-similar video frame images into different classes, includes:
calculating cosine similarity between two video frame images with better quality which are continuous in time, wherein if the similarity is greater than a set threshold value, the video frame images with better quality which are continuous in time are of the same type; otherwise, the classification is different;
for each type of the multiple video frame images, extracting a key video frame, including: calculating the distance characteristic average value of a plurality of video frame images of the same class, and extracting the video frame image with the distance characteristic value closest to the distance characteristic average value as a key video frame.
Optionally, the step of forming the extracted key video frame and the non-similar video frame image into a first extracted key video frame further includes:
calculating the similarity of two adjacent key video frames in the time sequence in the key video frames after the first extraction, reserving the key video frame of the previous frame for the two adjacent key video frames with the similarity, removing the key video frame of the next frame, and using the reserved key video frame as the key video frame after the second extraction.
Optionally, the extracting text information in each key video frame includes:
and extracting text information in each key video frame based on a transform neural network.
Optionally, the matching in the sensitive information base to obtain the sensitive text information in the key video frames and mark at least one key video frame including the sensitive text information includes:
sensitive character information needing desensitization is obtained, a Hash index is established based on the sensitive character information, and a sensitive information base is established;
and matching the text information extracted from the key video frames with the sensitive text information in the sensitive information base in a depth-first traversal mode in the sensitive information base, acquiring the sensitive text information in the key video frames and marking the key video frames comprising the sensitive text information.
Optionally, the removing the sensitive text information from each key video frame containing the sensitive text information and the video frames before and after the key video frame is processed according to a time sequence includes:
sensitive character information in each key video frame containing the sensitive character information is deleted, and video frames with a set number in front of and behind the key video frame containing the sensitive character information are provided.
According to a second aspect of the present invention, there is provided a system for automatic desensitization of sensitive information in road traffic video, comprising:
the extraction module is used for acquiring road traffic video frame data, extracting the road traffic video frame data and acquiring an extracted key video frame;
the acquisition module is used for extracting the text information in each key video frame, matching the text information in the sensitive information base, acquiring the sensitive text information in the key video frames and marking at least one key video frame comprising the sensitive text information;
the processing module is used for processing the sensitive character information in each key video frame containing the sensitive character information and removing the front and rear video frames which are arranged according to the time sequence;
and the generating module is used for recombining the key video frames subjected to sensitive text information processing and the remaining video frames which are not removed to generate a new desensitized video frame file.
According to a third aspect of the present invention, there is provided an electronic device comprising a memory, a processor for implementing the steps of the method for automatic desensitization processing of sensitive information in road traffic video when executing a computer management like program stored in the memory.
According to a fourth aspect of the present invention, there is provided a computer readable storage medium, on which a computer management-like program is stored, which when executed by a processor, implements the steps of a method for automatic desensitization processing of sensitive information in road traffic video.
According to the method and the system for automatically desensitizing sensitive information in the road traffic video, provided by the invention, the road traffic video frames are obtained, and key frames are extracted from the video frames by methods of screening high-quality video frames, clustering and the like, so that the number of processed video frames is greatly reduced; then, extracting and tracking the character information in the video frame so as to conveniently and quickly extract and match the character information in the video frame; the extracted text information is matched in the sensitive information, a video frame comprising the sensitive text information is found, the sensitive text information in the video frame is processed, front and back non-key frames in the sensitive key frame are removed to synthesize a new video, and the new video is returned to a high-precision map operation system, so that the automatic desensitization of the sensitive information of the road traffic video in the map making can be completed, the traditional road traffic video sensitive information desensitization needs manual interaction, filtering, screening and removing are needed for one frame, and the labor time is consumed.
Drawings
FIG. 1 is a flow chart of a method for processing automatic desensitization of sensitive information in a road traffic video according to the present invention;
FIG. 2 is a partially retained key frame pictorial illustration of a road traffic video after extraction;
FIG. 3 is a schematic diagram of a transform neural network;
FIG. 4 is a diagram illustrating an original image of a key frame in a video frame;
FIG. 5 is a diagram illustrating the extraction result of text information in key frames of a video sequence;
FIG. 6 is an overall flowchart of a method for automatic desensitization processing of sensitive information in road traffic video;
FIG. 7 is a schematic structural diagram of an automatic desensitization processing system for sensitive information in a road traffic video according to the present invention;
fig. 8 is a schematic diagram of a hardware structure of a possible electronic device provided in the present invention;
fig. 9 is a schematic diagram of a hardware structure of a possible computer-readable storage medium provided in the present invention.
Detailed Description
In order to make the objects, 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 with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention. In addition, technical features of various embodiments or individual embodiments provided by the present invention may be arbitrarily combined with each other to form a feasible technical solution, and such combination is not limited by the sequence of steps and/or the structural composition mode, but must be realized by a person skilled in the art, and when the technical solution combination is contradictory or cannot be realized, such a technical solution combination should not be considered to exist and is not within the protection scope of the present invention.
Fig. 1 is a flowchart of a method for automatic desensitization of sensitive information in a road traffic video, as shown in fig. 1, the method includes:
s1, acquiring road traffic video frame data, extracting the road traffic video frame data, and acquiring an extracted key video frame.
Understandably, the data information of the collected road traffic video packet transmitted to the cloud end by the map making drive test collecting vehicle is obtained. And decoding the video, namely receiving the road traffic video packet at the cloud end, decoding the road traffic video packet into original video data, and acquiring video frame data.
As an embodiment, the extracting the road traffic video frame data to obtain the extracted key video frame includes: extracting road traffic video frame images with better quality based on a time sequence according to the brightness, the definition and the stain information of the video frame images; according to the track point information of the collected videos during the running of the vehicle road, clustering video frame images with better quality between two adjacent track points, classifying similar video frame images with continuous time into the same class, and classifying non-similar video frame images into different classes; and extracting key video frames from a plurality of video frame images in the same class, and forming the extracted key video frames and the non-similar video frame images into key video frames after first extraction.
The clustering of the video frame images with better quality between two adjacent track points, the grouping of similar video frame images with continuous time into the same category, and the grouping of dissimilar video frame images into different categories include: calculating cosine similarity between two video frame images with better quality which are continuous in time, wherein if the similarity is greater than a set threshold value, the video frame images with better quality which are continuous in time are of the same type; otherwise, it is not classified.
It can be understood that, because the amount of video frame data is large, the variation between frames is small, similar information is large, and the redundancy amount is large. In order to accelerate the processing speed of the algorithm, key frame extraction is carried out on the adopted road traffic video frames.
Firstly, preliminarily screening video frame data, removing video frame images with damaged quality according to information such as brightness, definition and stain of the video frame images, and reserving road traffic video frame images with better quality based on a time sequence.
For the video frame images after the preliminary screening, clear video frames between two adjacent track points are clustered according to track point information of a collected video when a vehicle runs on a road, the cosine similarity of the adjacent video frames on two time sequences of sequence frames is mainly calculated to measure the similarity of the time sequence video frames, the time-continuous similar video frames are fixed in the same class, and the non-similar video frames are classified in different classes.
As an embodiment, said extracting key video frames for each of the plurality of video frame images of each class includes: calculating the distance characteristic average value of a plurality of video frame images of the same class, and extracting the video frame image with the distance characteristic value closest to the distance characteristic average value as a key video frame.
It can be understood that for a plurality of video frame images of the same type, the distance characteristic value of each video frame image is calculated, the distance characteristic average value of the plurality of video frame images is obtained, and the video frame image with the distance characteristic value closest to the distance characteristic average value of the type in the plurality of video frame images is used as the key video frame. And for other video frame images in different classes, not extracting, and then taking the key video frame extracted from the same class and the video frames in different classes as the key video frame after the first extraction.
As an embodiment, the forming the extracted key video frame and the non-similar video frame image into a first extracted key video frame further includes: calculating the similarity of two adjacent key video frames in the time sequence in the key video frames after the first extraction, reserving the key video frame of the previous frame for the two adjacent key video frames with the similarity, removing the key video frame of the next frame, and using the reserved key video frame as the key video frame after the second extraction.
It can be understood that the number of key video frames obtained after the first extraction is still huge, and therefore, the second extraction is performed. Specifically, for the key video frame after the first extraction, the similarity between two adjacent key video frames in the time sequence is determined, for the two frames with similarity, the key video frame in the previous frame is retained, the key video frame in the next frame is removed, the retained key video frame is used as the key video frame after the second extraction, and fig. 2 shows that the key frame is retained for the part after the road traffic video extraction.
And S2, extracting the text information in each key video frame, matching in a sensitive information base to obtain the sensitive text information in the key video frame, and marking at least one key video frame comprising the sensitive text information.
It can be understood that after the key video frames after the second extraction are obtained, the text information in each video frame image is extracted. In the invention, a transform neural network is mainly adopted to extract the key frame characters of the video. The method comprises the following steps:
firstly, a Backbone network (Backbone) is adopted to extract the multi-layer network characteristics of the video key frames which are related at the front and the back, and a transform coding form is mainly adopted to obtain the multi-layer network characteristic graphs of the original image. And (3) flattening the image column into a plurality of image 2D blocks, carrying out position coding, and sequentially extracting feature maps of a plurality of hierarchical networks of the video keyframe by adopting a self-attention mechanism and a feedforward neural network. Secondly, performing transform encoding (transform Encoder) on the sequence video key frames which are related in the front and back, and adopting a plurality of self-attention mechanisms and feedforward neural networks to mainly extract similarity related characteristics of the sequence video key frames which are related in the front and back. Then, the tracked Object box and the detected Object box are respectively matched by detection through two branch decoding modules (Object Encoder and Tracking Encoder). And finally, extracting the character real-column segmentation by adopting a CNN plus FPN coding and decoding network (Predict) to obtain a segmentation graph, and obtaining a text bounding box from the segmentation graph for prediction. The flow chart of the structure of video key frame sensitive text information extraction is shown in fig. 3. Fig. 4 is an original drawing of a video frame, and fig. 5 is a result of extracting text information of a key frame of the video.
And S3, processing the sensitive character information in each key video frame containing the sensitive character information, and removing the front and rear video frames arranged according to a time sequence.
As an embodiment, the matching in the sensitive information base to obtain the sensitive text information in the key video frames and mark at least one key video frame including the sensitive text information includes: acquiring sensitive text information needing desensitization, establishing a hash index based on the sensitive text information, and establishing a sensitive information base; and matching the text information extracted from the key video frames with the sensitive text information in the sensitive information base in a depth-first traversal mode in the sensitive information base, acquiring the sensitive text information in the key video frames and marking the key video frames comprising the sensitive text information.
Understandably, sensitive character information needing desensitization is obtained, key information of the sensitive character information needing desensitization is used as a key to establish a hash index, and a sensitive information base is established.
And matching and comparing the text information extracted from the key video frame in a sensitive information base to find the sensitive text information in the key video frame, wherein the text information extracted from the key video frame is matched with the sensitive text information in the sensitive information base in a depth-first traversal mode to obtain the sensitive text information in the key video frame and mark the key video frame comprising the sensitive text information.
As an embodiment, the removing the sensitive text information from the key video frames containing the sensitive text information to remove the front and rear video frames arranged according to the time sequence includes: sensitive character information in each key video frame containing the sensitive character information is deleted, and video frames with a set number in front of and behind the key video frame containing the sensitive character information are removed.
And for the key video frames containing the sensitive text information, the sensitive text information is scratched from the video frames, and the front and rear video frames of the key video frames containing the sensitive text information are removed. During the elimination, only non-key video frames in the original video frame data are eliminated, and some video frames are eliminated, so that the subsequent functions are not influenced.
And S4, recombining the key video frames subjected to sensitive text information processing and the remaining video frames which are not removed into a new desensitized video frame file.
It can be understood that the key video frames after the sensitive character information is scratched and the video frames which are not removed and remained are recombined into a video file, corresponding track information is returned and transmitted to a map making system, and then the automatic desensitization of the sensitive information in the road traffic video in the map making can be completed.
Referring to fig. 6, an overall flowchart of a processing method for automatic desensitization of sensitive information in a road traffic video mainly includes: firstly, decoding a video data packet at the cloud of a company to synthesize video frame data; secondly, key frame extraction is carried out on the video frames by methods of screening and clustering high-quality video frames and the like, so that the processing quantity of the video frames is greatly reduced. And then, extracting and tracking the character information in the video frame through a transform-based neural network so as to conveniently and quickly extract and match the character information in the video frame, and increasing the bloom capability of the algorithm by adopting a neural network method so as to provide sensitive information in traffic road videos in various scenes. And finally, establishing a hash index for the sensitive information needing desensitization according to the sensitive keyword so as to facilitate the quick matching of the extracted video frame and the sensitive information. After front and rear non-key frames in the sensitive key frames are removed and synthesized into a new video, the new video is returned to the high-precision map operating system, so that automatic desensitization of sensitive information of road traffic videos in map making can be completed, and the desensitized video is used for high-precision map making personnel to make and prevent leakage of national safety information.
Fig. 7 is a structural diagram of a system for automatic desensitization processing of sensitive information in road traffic video according to an embodiment of the present invention, and as shown in fig. 7, the system for automatic desensitization processing of sensitive information in road traffic video includes an extraction module 701, an acquisition module 702, a processing module 703, and a generation module 704, where:
an extraction module 701, configured to obtain road traffic video frame data, extract the road traffic video frame data, and obtain an extracted key video frame;
an obtaining module 702, configured to extract text information in each key video frame, perform matching in a sensitive information base, obtain sensitive text information in the key video frame, and mark at least one key video frame including the sensitive text information;
the processing module 703 is configured to process the sensitive text information in each key video frame that includes the sensitive text information, and remove video frames before and after the key video frame that are arranged according to a time sequence;
and a generating module 704, configured to recombine the key video frames after the sensitive text information is processed and the remaining video frames that are not removed to generate a new desensitized video frame file.
It can be appreciated that the present invention provides one. The system for automatically desensitizing sensitive information in a road traffic video corresponds to the method for automatically desensitizing sensitive information in a road traffic video provided in the foregoing embodiments, and relevant technical features of the system for automatically desensitizing sensitive information in a road traffic video may refer to relevant technical features of the method for automatically desensitizing sensitive information in a road traffic video, which are not described herein again.
Referring to fig. 8, fig. 8 is a schematic diagram of an embodiment of an electronic device according to an embodiment of the invention. As shown in fig. 8, an embodiment of the present invention provides an electronic device 800, which includes a memory 810, a processor 820, and a computer program 811 stored in the memory 810 and being executable on the processor 820, wherein the processor 820 implements the steps of the method for processing the automatic desensitization of the sensitive information in the road traffic video when executing the computer program 811.
Referring to fig. 9, fig. 9 is a schematic diagram of an embodiment of a computer-readable storage medium according to the present invention. As shown in fig. 9, the present embodiment provides a computer-readable storage medium 900, on which a computer program 911 is stored, where the computer program 911, when executed by a processor, implements the steps of the method for processing automatic desensitization of sensitive information in road traffic video.
The method and the system for automatically desensitizing sensitive information in the road traffic video, provided by the embodiment of the invention, have the following beneficial effects:
because the map acquisition vehicle acquires a lot of traffic road video files containing hundreds of kilometers of road traffic videos, in order to accelerate the processing speed of the map acquisition vehicle for acquiring the videos and reduce the memory consumption, the invention solves the problems of the synthesis of the videos acquired by the map acquisition vehicle and the extraction of key frames, and accelerates the processing speed.
The invention provides key frame sensitive text information extraction based on a transform neural network, and mainly aims to improve the generalization capability of video desensitization information extraction.
In order to quickly match desensitization information, the invention provides a method for establishing a hash table by using a sensitive keyword, searching the hash table by using a depth-first traversal algorithm, and quickly matching the extracted video text information with text information needing desensitization. The processing speed is increased.
It should be noted that, in the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to relevant descriptions of other embodiments for parts that are not described in detail in a certain embodiment.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. An automatic desensitization processing method for sensitive information in road traffic video is characterized by comprising the following steps:
acquiring road traffic video frame data, extracting the road traffic video frame data, and acquiring an extracted key video frame;
extracting text information in each key video frame, matching in a sensitive information base to obtain sensitive text information in the key video frames and marking at least one key video frame comprising the sensitive text information;
processing the sensitive character information in each key video frame containing the sensitive character information, and removing the front video frame and the rear video frame which are arranged according to a time sequence;
and recombining the key video frames subjected to sensitive text information processing and the remaining video frames which are not removed into a new desensitized video frame file.
2. The method for automatic desensitization processing of sensitive information according to claim 1, wherein extracting the road traffic video frame data to obtain extracted key video frames comprises:
extracting road traffic video frame images with better quality based on a time sequence according to the brightness, the definition and the stain information of the video frame images;
according to the track point information of the collected videos during the running of the vehicle road, clustering video frame images with better quality between two adjacent track points, classifying similar video frame images with continuous time into the same class, and classifying non-similar video frame images into different classes;
and extracting key video frames from a plurality of video frame images in the same class, and forming the extracted key video frames and the non-similar video frame images into key video frames after first extraction.
3. The automatic desensitization processing method of sensitive information according to claim 2, wherein said clustering video frame images with better quality between two adjacent track points, categorizing similar video frame images that are continuous in time into the same category, categorizing non-similar video frame images into different categories, comprises:
calculating cosine similarity between two video frame images with better quality which are continuous in time, wherein if the similarity is greater than a set threshold value, the video frame images with better quality which are continuous in time are of the same type; otherwise, the classification is different;
for each type of the multiple video frame images, extracting a key video frame, including:
calculating the distance characteristic average value of a plurality of video frame images of the same class, and extracting the video frame image with the distance characteristic value closest to the distance characteristic average value as a key video frame.
4. The method for processing automatic desensitization of sensitive information according to claim 2, wherein said extracting key video frames and non-similar video frame images form key video frames after first extraction, and then further comprising:
calculating the similarity of two adjacent key video frames in the time sequence in the key video frames after the first extraction, reserving the key video frame of the previous frame for the two adjacent key video frames with the similarity, removing the key video frame of the next frame, and using the reserved key video frame as the key video frame after the second extraction.
5. The method of claim 1, wherein the extracting text information from each key video frame comprises:
and extracting text information in each key video frame based on a transform neural network.
6. The method of claim 1, wherein the matching in the sensitive information repository to obtain sensitive textual information in key video frames and to mark at least one key video frame including the sensitive textual information comprises:
sensitive character information needing desensitization is obtained, a Hash index is established based on the sensitive character information, and a sensitive information base is established;
and matching the text information extracted from the key video frames with the sensitive text information in the sensitive information base in a depth-first traversal mode in the sensitive information base, acquiring the sensitive text information in the key video frames and marking the key video frames comprising the sensitive text information.
7. The method for automatic desensitization of sensitive information according to claim 1, wherein said processing sensitive text information in each key video frame containing sensitive text information to remove the video frames before and after the key video frame according to the time sequence comprises:
and deleting the sensitive text information in each key video frame containing the sensitive text information, and providing a set number of video frames before and after the key video frame containing the sensitive text information.
8. An automatic desensitization processing system of sensitive information in road traffic video, characterized by comprising:
the extraction module is used for acquiring road traffic video frame data, extracting the road traffic video frame data and acquiring extracted key video frames;
the acquisition module is used for extracting the text information in each key video frame, matching the text information in the sensitive information base, acquiring the sensitive text information in the key video frames and marking at least one key video frame comprising the sensitive text information;
the processing module is used for processing the sensitive character information in each key video frame containing the sensitive character information and removing the front and rear video frames which are arranged according to the time sequence;
and the generating module is used for recombining the key video frames subjected to sensitive text information processing and the remaining video frames which are not removed to generate a new desensitized video frame file.
9. An electronic device comprising a memory and a processor, wherein the processor is configured to execute a computer management program stored in the memory to implement the steps of the method for automatic desensitization of sensitive information in road traffic video according to any of claims 1 to 7.
10. A computer-readable storage medium, characterized in that a computer management-like program is stored thereon, which, when being executed by a processor, implements the steps of the method for automatic desensitization processing of sensitive information in road traffic videos according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211340842.2A CN115810160A (en) | 2022-10-29 | 2022-10-29 | Automatic desensitization processing method and system for sensitive information in road traffic video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211340842.2A CN115810160A (en) | 2022-10-29 | 2022-10-29 | Automatic desensitization processing method and system for sensitive information in road traffic video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115810160A true CN115810160A (en) | 2023-03-17 |
Family
ID=85482934
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211340842.2A Pending CN115810160A (en) | 2022-10-29 | 2022-10-29 | Automatic desensitization processing method and system for sensitive information in road traffic video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115810160A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117278692A (en) * | 2023-11-16 | 2023-12-22 | 邦盛医疗装备（天津）股份有限公司 | Desensitization protection method for monitoring data of medical detection vehicle patients |
| CN117455751A (en) * | 2023-12-22 | 2024-01-26 | 新华三网络信息安全软件有限公司 | Road section image processing system and method |
-
2022
- 2022-10-29 CN CN202211340842.2A patent/CN115810160A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117278692A (en) * | 2023-11-16 | 2023-12-22 | 邦盛医疗装备（天津）股份有限公司 | Desensitization protection method for monitoring data of medical detection vehicle patients |
| CN117278692B (en) * | 2023-11-16 | 2024-02-13 | 邦盛医疗装备（天津）股份有限公司 | Desensitization protection method for monitoring data of medical detection vehicle patients |
| CN117455751A (en) * | 2023-12-22 | 2024-01-26 | 新华三网络信息安全软件有限公司 | Road section image processing system and method |
| CN117455751B (en) * | 2023-12-22 | 2024-03-26 | 新华三网络信息安全软件有限公司 | Road section image processing system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Deep matching and validation network: An end-to-end solution to constrained image splicing localization and detection | |
| CN115810160A (en) | Automatic desensitization processing method and system for sensitive information in road traffic video | |
| Luo et al. | Newsclippings: Automatic generation of out-of-context multimodal media | |
| CN107103314B (en) | A kind of fake license plate vehicle retrieval system based on machine vision | |
| US20090292701A1 (en) | Method and a system for indexing and searching for video documents | |
| CN111079816A (en) | Image auditing method and device and server | |
| CN106126680A (en) | A kind of video image reconnaissance method and system | |
| CN112329719A (en) | Behavior recognition method, behavior recognition device and computer-readable storage medium | |
| CN113450330B (en) | Image copying-pasting tampering detection method based on segmentation and depth convolution network | |
| Bulugu | Algorithm for license plate localization and recognition for tanzania car plate numbers | |
| CN111914648A (en) | Vehicle detection and identification method and device, electronic equipment and storage medium | |
| Khan et al. | Helmet and number plate detection of motorcyclists using deep learning and advanced machine vision techniques | |
| Jin et al. | Object-based video forgery detection via dual-stream networks | |
| CN117237418A (en) | Moving object detection method and system based on deep learning | |
| Aparicio-Díaz et al. | Temporal copy-move forgery detection and localization using block correlation matrix | |
| Zhong et al. | A fast forgery frame detection method for video copy-move inter/intra-frame identification | |
| CN114140674B (en) | Electronic evidence availability identification method combined with image processing and data mining technology | |
| CN113850284B (en) | Multi-operation detection method based on multi-scale feature fusion and multi-branch prediction | |
| CN114463755A (en) | Automatic sensitive information detection desensitization method in high-precision map-based acquired picture | |
| CN109657098B (en) | Video fingerprint extraction method and device | |
| Miao et al. | Multi-spectral Class Center Network for Face Manipulation Detection and Localization | |
| Kaur et al. | Digital Image Forgery and Techniques of Forgery Detection: A brief review | |
| CN114022869B (en) | Vehicle heavy identification method and device based on cascade network | |
| Harun et al. | Video structure extraction using shot boundary detection for authentication detection | |
| Al-Musawi et al. | Improving video streams summarization using synthetic noisy video data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |