CN-122027611-A - Video transmission method, system, equipment and medium based on narrowband condition

CN122027611ACN 122027611 ACN122027611 ACN 122027611ACN-122027611-A

Abstract

The application provides a video transmission method, a system, equipment and a medium based on a narrowband condition, and relates to the technical field of video encoding and decoding, wherein the method comprises the steps of carrying out frame type identification on original video data through a video preprocessing unit, generating a key frame and a non-key frame, identifying a scene switching frame in the original video data, and extracting a scene characteristic descriptor of the scene switching frame; comparing the scene feature descriptors with the historical scene feature descriptors, identifying new scene frames and scene multiplexing frames, processing the new scene frames, the key frames, the scene multiplexing frames and the non-key frames to generate preprocessed video data, compressing the preprocessed video data to generate compressed video data, transmitting the compressed video data to a decoding end, and restoring the decoded video data to generate output video data. The application has the technical effect of meeting the double requirements of video transmission quality and transmission speed in practical application.

Inventors

ZHANG JUN
TAN YUANYUAN

Assignees

湖南驰阳信息科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. A video transmission method based on narrowband conditions, the method comprising: Performing frame type identification on original video data through a video preprocessing unit, generating a key frame and a non-key frame, identifying a scene switching frame in the original video data, and extracting a scene characteristic descriptor of the scene switching frame; comparing the scene feature descriptors with historical scene feature descriptors, wherein the scene feature descriptors with similarity lower than a preset similarity threshold value are used as new scene frames, and the scene feature descriptors with similarity higher than the preset similarity threshold value are used as scene multiplexing frames; performing first preprocessing on the new scene frame, performing second preprocessing on the key frame, performing third preprocessing on the scene multiplexing frame and performing fourth preprocessing on the non-key frame to generate preprocessed video data; Transmitting the compressed video data to a decoding end through a narrowband network, so that the decoding end decodes the compressed video data through the decoding module to generate decoded video data; and inputting the decoded video data into the video preprocessing unit for restoration processing, generating output video data, and outputting the output video data to display equipment for display.
2. The video transmission method according to claim 1, wherein the performing frame type recognition on the original video data by the video preprocessing unit, generating key frames and non-key frames, and recognizing scene-switching frames in the original video data, extracting scene feature descriptors of the scene-switching frames, comprises: Calculating pixel difference values between adjacent video frames in original video data through a video preprocessing unit, marking the corresponding video frames as key frames when the pixel difference values are larger than a preset frame difference threshold value, and marking the corresponding video frames as non-key frames when the pixel difference values are smaller than or equal to the preset frame difference threshold value; calculating scene similarity between adjacent video frames in the original video data through a video preprocessing unit, and identifying the corresponding video frames as scene switching frames when the scene similarity is smaller than a preset scene switching threshold; and carrying out feature extraction on the scene switching frame, wherein the feature extraction comprises the steps of extracting color histogram features, texture features and edge contour features, and combining the color histogram features, the texture features and the edge contour features to generate the scene feature descriptor.
3. The video transmission method according to claim 1, wherein the scene feature descriptor having a similarity lower than a preset similarity threshold is used as a new scene frame, and the scene feature descriptor having a similarity higher than the preset similarity threshold is used as a scene multiplexing frame, comprising: Extracting historical scene feature descriptors from a historical scene feature library, the historical scene feature descriptors comprising scene feature descriptors in transmitted video data; calculating the similarity between the scene feature descriptors and the historical scene feature descriptors; And when the similarity is greater than or equal to the preset similarity threshold value, judging the current scene switching frame as a scene multiplexing frame.
4. A video transmission method based on narrowband conditions as defined in claim 3, wherein said calculating a similarity between the scene feature descriptor and the historical scene feature descriptor comprises: extracting color histogram features, texture features and edge contour features in the scene feature descriptors and the historical scene feature descriptors respectively; Calculating the color similarity between the color histogram features, calculating the texture similarity between the texture features, and calculating the contour similarity between the edge contour features; carrying out weighted summation on the color similarity, the texture similarity and the contour similarity to generate comprehensive similarity; and taking the comprehensive similarity as the similarity between the scene feature descriptor and the historical scene feature descriptor.
5. The method for video transmission based on narrowband conditions of claim 1, wherein the performing a first preprocessing on the new scene frame, a second preprocessing on the key frame, a third preprocessing on the scene multiplexed frame, and a fourth preprocessing on the non-key frame, generates preprocessed video data, comprises: Performing first preprocessing on the new scene frame, wherein the first preprocessing comprises resolution reduction processing and quality reduction processing by adopting a first compression parameter; Performing second preprocessing on the key frames, wherein the second preprocessing comprises resolution reduction processing and quality reduction processing by adopting second compression parameters; performing third preprocessing on the scene multiplexing frame, wherein the third preprocessing comprises resolution reduction processing and quality reduction processing by adopting third compression parameters; Performing fourth preprocessing on the non-key frames, wherein the fourth preprocessing comprises resolution reduction processing and quality reduction processing by adopting fourth compression parameters; wherein the compression strength of the first compression parameter is less than the compression strength of the second compression parameter, which is less than the compression strength of the third compression parameter, which is less than the compression strength of the fourth compression parameter; And combining the new scene frame, the key frame, the scene multiplexing frame and the non-key frame which are subjected to different preprocessing to generate preprocessed video data.
6. The video transmission method based on the narrowband condition according to claim 1, wherein the inputting the preprocessed video data into the encoding module for video encoding compression, generating compressed video data, comprises: Inputting the new scene frame and the key frame in the preprocessed video data into an encoding module for complete frame encoding to generate complete frame encoding data; for the scene multiplexing frame, according to the history scene index corresponding to the history scene feature descriptor, acquiring matched history scene frame data from a history scene database as a reference frame; Calculating inter-frame residual data between the scene multiplex frame and the reference frame; encoding the inter-frame residual data, and recording the historical scene index in a correlated way to generate scene multiplexing frame encoded data; Performing inter-frame predictive coding on the non-key frames to generate non-key frame coding data; and encapsulating the complete frame coding data, the scene multiplexing frame coding data, the non-key frame coding data and the historical scene index to generate compressed video data.
7. The video transmission method according to claim 6, wherein the decoding end decodes the compressed video data by a decoding module to generate decoded video data, comprising: the decoding end receives the compressed video data and analyzes a frame type identifier and a historical scene index in the compressed video data; decoding the complete frame coding data to generate a new decoded scene frame and a key frame; decoding the scene multiplexing frame coding data to obtain decoded inter-frame residual data; Extracting corresponding historical scene frame data from a historical scene database of the decoding end according to the historical scene index; overlapping and reconstructing the decoded inter-frame residual data and the historical scene frame data to generate a reconstructed scene multiplexing frame; Decoding the non-key frame coded data to generate a decoded non-key frame; and combining the decoded new scene frame, the key frame, the reconstructed scene multiplexing frame and the decoded non-key frame according to an original frame sequence to generate decoded video data.
8. A video transmission system based on a narrow-band condition is characterized by comprising an identification module, a comparison module, a processing module, an input module and an output module, wherein, The identification module is used for carrying out frame type identification on the original video data through the video preprocessing unit, generating key frames and non-key frames, identifying scene switching frames in the original video data, and extracting scene characteristic descriptors of the scene switching frames; The comparison module is used for comparing the scene feature descriptors with historical scene feature descriptors, wherein the scene feature descriptors with similarity lower than a preset similarity threshold value are used as new scene frames, and the scene feature descriptors with similarity higher than the preset similarity threshold value are used as scene multiplexing frames; the processing module is used for carrying out first preprocessing on the new scene frame, carrying out second preprocessing on the key frame, carrying out third preprocessing on the scene multiplexing frame and carrying out fourth preprocessing on the non-key frame, and generating preprocessed video data; The input module is used for inputting the preprocessed video data into the encoding module for video encoding and compression to generate compressed video data, and transmitting the compressed video data to a decoding end through a narrowband network so that the decoding end decodes the compressed video data through the decoding module to generate decoded video data; the output module is used for inputting the decoded video data into the video preprocessing unit for reduction processing, generating output video data, and outputting the output video data to a display device for display.
9. An electronic device comprising a processor, a memory, a user interface, and a network interface, the memory for storing instructions, the user interface and the network interface for communicating to other devices, the processor for executing the instructions stored in the memory to cause the electronic device to perform the method of any of claims 1-7.
10. A computer readable storage medium, characterized in that a computer program is stored which can be loaded by a processor and which performs the method according to any of claims 1-7.

Description

Video transmission method, system, equipment and medium based on narrowband condition Technical Field The application relates to the technical field of video encoding and decoding, in particular to a video transmission method, a system, equipment and a medium based on a narrowband condition. Background With the rapid development of mobile internet and multimedia applications, video data transmission needs are growing increasingly, however, under a narrow-band network environment, due to serious bandwidth limitation, the traditional video transmission method often faces technical problems of low transmission speed, poor video quality, insufficient real-time performance and the like, and particularly under the conditions of remote areas, weak mobile network signals or network congestion, the problems are more prominent, and the video watching experience of users and the practical usability of applications are seriously affected. At present, the video transmission problem under the narrowband condition is mainly solved by reducing the video resolution, improving the compression ratio, adopting a more efficient coding algorithm and other modes, for example, advanced coding standards such as H.264, H.265 and the like are used for compressing the video, or an adaptive code rate adjustment technology is adopted for dynamically adjusting the video quality according to the network condition. However, the existing video transmission method adopts a unified compression and coding processing strategy for all video frames, and the processing mode of one cut cannot realize the optimal balance of transmission efficiency and video quality in a narrow-band network environment, so that the dual requirements of video transmission quality and transmission speed in practical application are difficult to meet. Disclosure of Invention The application provides a video transmission method, a system, equipment and a medium based on a narrowband condition, which are used for meeting the double requirements of video transmission quality and transmission speed in practical application. The video transmission method based on the narrow-band condition comprises the steps of carrying out frame type identification on original video data through a video preprocessing unit to generate a key frame and a non-key frame, identifying a scene switching frame in the original video data, extracting a scene characteristic descriptor of the scene switching frame, comparing the scene characteristic descriptor with a historical scene characteristic descriptor, taking the scene characteristic descriptor with similarity lower than a preset similarity threshold value as a new scene frame, taking the scene characteristic descriptor with similarity higher than the preset similarity threshold value as a scene multiplexing frame, carrying out first preprocessing on the new scene frame, carrying out second preprocessing on the key frame, carrying out third preprocessing on the scene multiplexing frame and carrying out fourth preprocessing on the non-key frame to generate preprocessed video data, carrying out video encoding compression on the preprocessed video data input encoding module to generate compressed video data, transmitting the compressed video data to a decoding end through a narrow-band network, enabling the decoding end to decode the compressed video data through the decoding module, carrying out decoding on the compressed video data to generate the decoded video data, carrying out third preprocessing on the scene multiplexing frame, carrying out fourth preprocessing on the non-key frame to generate preprocessed video data, and carrying out output processing on the video data to be output to display and display. By adopting the technical scheme, the scene characteristic descriptors can be accurately extracted and compared with the historical scene characteristic descriptors through frame type recognition and scene switching frame recognition on the original video data, so that intelligent distinction between the new scene frames and the scene multiplexing frames is realized, and further, differentiated preprocessing strategies are respectively adopted for the new scene frames, the key frames, the scene multiplexing frames and the non-key frames, thereby effectively avoiding the problems of important information loss and redundant data transmission caused by unified processing in the traditional method. According to the scheme, repeated transmission of similar scenes is reduced through a scene multiplexing mechanism, meanwhile, data volume distribution of frames with different importance is optimized through hierarchical preprocessing and coding compression, transmission efficiency under a narrow-band network condition is remarkably improved, the optimal balance of transmission speed and video quality is realized on the premise that complete transmission of key video information is ensured, and the dual requirements of video transmission quality and