Video hierarchical and partitioned storage system based on content features

Abstract

The invention discloses a video hierarchical and partitioned storage system based on content features. The storage system comprises a video source module, a CA-SDG module and a plurality of hierarchical storage rear-end modules. A video source end management dispatch unit is a hinge between the video source module and a hierarchical storage decision module, is in charge of data receiving and data distinction, and transmits distinguished data to the hierarchical storage decision module, and the distinguished data are distributed and stored in the hierarchical storage rear-end modules. The video hierarchical and partitioned storage system is designed on the base of the essential content features of data, and not only stores information of original video, but also stores texts, pictorial information, video behavioral feature information and the like extracted from the original video. The video data are marked with media flags, the degree of accuracy of data information is improved, search conditions are wide in range and various, search is simpler, more convenient and more humanized, the reusing rate of the data is increased, management and archiving are convenient, storage cost is lowered, and manual screening efficiency of the video data is improved.

Description

technical field [0001] The invention relates to a video storage system, in particular to a system for classifying and partitioning video storage based on content features. Background technique [0002] With the development and wide application of network technology, video technology has also been widely used. When people search for videos, they usually put the content of the video in the first place, such as the faces involved, the clothing of the characters, the behavior of the characters, and others. Image features related to people and vehicles, the region where the content occurs, body color, license plate recognition, car logo, model, subtitle analysis, etc. A series of feature retrieval; the above features are stored separately, which can be used as direct evidence or for video search. Therefore, grading and partitioning storage of video content is particularly important for video retrieval. [0003] In the existing video hierarchical storage system, it is mainly bas...

AI Technical Summary

Problems solved by technology

[0003] In the existing video hierarchical storage system, it is mainly based on comprehensive factors such as system hardware performance, cost, importance of storage content, security, popularity, and timeliness to consider layering and hierarchical storage of storage content, even if it enters cloud storage In the current era, the hierarchical storage system involved is still limited to traditional hierarchical storage; (patent name: a hierarchical storage in a distributed environment, patent application number: 201010540729.X), this hierarchical storage system is to store data according to In many cases, a lot of completely different storage content will be stored in one place according to the performance parameters and cost of the hardware, resulting in unnecessary network transmission overhead; it also achieves low-cost storage by reducing access performance. More importantly, it cannot correctly reflect the essence of the data, resulting in high cost and difficulty in managing archives

However, the content-aware storage system (Content Aware Storage System) is currently limited to dividing content according to the nature of the content, such as video, audio, documents, images, etc., and cannot be separated or hierarchically stored according to specific content characteristics. visitor needs

Images