Subsea monitoring video big data storage method utilizing segmentation and compression

Abstract

The invention relates to a subsea monitoring video big data storage method using segmentation and compression. The method comprises the following steps: step 1, enabling a video data center to acquiresubsea monitoring video data; 2, enabling the video data center to perform video segmentation based on the macro block type on the seabed monitoring video data to obtain a key video and a blank video; 3, performing feature extraction on the key video; 4, performing redundancy compression on the blank video; and 5, organizing the video segments and the characteristic attributes into a big data table, and storing the big data table into a database. Compared with the prior art, the method has the advantages of realizing quick video segmentation, reducing video data volume, simplifying a video storage structure, providing better load balance, realizing quick retrieval and the like.

Description

Technical field [0001] The invention relates to the technical field of subsea monitoring video storage, in particular to a subsea monitoring video big data storage method using segmentation and compression. Background technique [0002] With the initial establishment of the seabed monitoring network at home and abroad, many ocean observation methods relying on the monitoring network have begun to be implemented. Among them, seabed video monitoring is one of the important means. It can transmit the most intuitive seabed image data to the shore in real-time and all-weather. Marine scientific research personnel provide valuable information. However, the ocean area is wide, and the video data generated by monitoring is huge. If it is not stored well, valuable seabed image information will be lost. At present, no one has proposed a storage solution specifically for videos in the marine monitoring field. Usually video data is stored in files or tapes, but only limited and small amoun...

AI Technical Summary

Problems solved by technology

If you still use files to store these video fragments one by one, it will cause huge difficulties in the organization and management of video data, and it will also cause problems such as file redundancy and difficult maintenance of data and metadata consistency.

[0004] In the paper "Hbase-Based Submarine Monitoring Video Big Data Storage Method", a method of using the Hbase database for submarine monitoring video big data storage is disclosed. In this method, background subtraction is used to calculate the movement amount change in the video. This method requires Performing grayscale conversion on each key frame and making the square difference with the background frame requires a large amount of calculation, and it is difficult to cope with the real-time processing requirements of massive and high-intensity video streams in submarine monitoring

Images