Method and system for non-frame-loss segmented storage of video record based on Android system

Abstract

The invention provides a method and a system for non-frame-loss segmented storage of video record based on Android system. The method comprises the following steps of: adding a video record segmentation interface function in a Stagefright frame, and adding a segmentation function external interface in Writer; splitting specific implementation of the Writer into an external interface class and an implementation class; monitoring that recording of one video record segment is completed, notifying the implementation class of the video record segment by the external interface class to stop recording video, and simultaneously initializing the implementation class of next video record segment; asking a coder for an IDR frame by the implementation class of the video record segment, quitting the video record segment when meeting the asked IDR frame after asking for the IDR frame, and storing data of the video record segment; firstly extracting the asked IDR frame by the implementation class of next video record segment, and then extracting coded frames from the coder to continue video recording. The method and the system have the advantages of being capable of realizing an aim of seamless segmented storage of video record, avoiding frame loss, improving quality of video record, and having very low technical requirements on upgrading and migration of later versions.

Description

technical field [0001] The invention relates to the field of video storage, in particular to a method and system for segmented storage of video without frame loss based on the Android system. Background technique [0002] Products based on surveillance are widely used in the market, such as driving recorders, surveillance cameras, etc. Although the requirements of different monitoring products are different, for example, driving recorders require high-definition, single / dual-channel video recording; surveillance cameras require more However, all surveillance products need to save video in segments, so it is a common requirement of all surveillance products to ensure that frames cannot be lost between segmented recordings, and it is very important. [0003] At present, there are two main categories of the function of not dropping frames based on the Android system: first, use Android’s existing MediaRecorder API to segment the application, that is, after the application saves...

AI Technical Summary

Problems solved by technology

[0002] Products based on surveillance are widely used in the market, such as driving recorders, surveillance cameras, etc. Although the requirements of different monitoring products are different, for example, driving recorders require high-definition, single / dual-channel video recording; surveillance cameras require more However, all monitoring products need to save the video in segments, so it is a common requirement of all monitoring products to ensure that frames cannot be lost between segmented recordings, and it is very important

[0003] At present, there are two main categories of the function of not dropping frames based on the Android system: first, use Android’s existing MediaRecorder API to segment the application, that is, after the application saves the previous recording, it immediately enables the next recording, and the other The entire switching process will have an overhead of about 200Ms when the system IO is smooth, and there may be a gap of a few seconds when the system IO is blocked, which will seriously affect the actual recording; Android standard video recording framework, this method has the following defects: the technology is difficult, it needs to be familiar with Camera, audio input, audio and video codec and other skills, and the portability is poor after implementation, it cannot be adapted to different chip platforms, and the core framework is not a standard API , the third party cannot develop based on this framework, and it is very difficult to maintain in the later stage. At the same time, it has to maintain the problem of how to share the resources related to Camera and audio with the system. There are too many unpredictable problems.

Images