HEVC compressed domain video motion segmentation method and system based on diagonal regression classifier

A motion segmentation and compression domain technology, applied in instrumentation, image analysis, image data processing, etc., can solve problems such as insufficient consideration of temporal correlation, inability to handle lens motion, etc., to speed up video analysis and improve generalization capabilities , the effect of improving the accuracy

Inactive Publication Date: 2019-04-16
SHANGHAI UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this method makes full use of many compressed domain features, only one frame of data is used for judgment, and the correlation in the time domain is not fully considered.
In addition, these two methods only consider the situation of static background, they cannot deal with the scene where the camera moves, such as the outdoor surveillance camera will often be blown by the wind and shake slightly

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • HEVC compressed domain video motion segmentation method and system based on diagonal regression classifier
  • HEVC compressed domain video motion segmentation method and system based on diagonal regression classifier
  • HEVC compressed domain video motion segmentation method and system based on diagonal regression classifier

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0051] The present invention will be described in detail below in conjunction with specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0052] Such as figure 1 Shown, a kind of flow chart of the embodiment of the HEVC compressed domain video motion segmentation method based on the rate regression classifier, wherein the embodiment can be carried out with reference to the following steps:

[0053] Step S1: digitally convert the prediction mode of the prediction unit in the compressed domain code stream data; the compressed domain code stream data is compressed code stream data conforming to the HEVC standard;

[0054] In this step...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an HEVC compressed domain video motion segmentation method and system based on a diagonal regression classifier. The method comprises the following steps: carrying out numeralization on a prediction mode of a prediction unit in HEVC compressed code stream information; filtering compressed domain characteristics such as motion vectors and residual transform coefficients, removing isolated noise and filling up holes; detecting and removing global motion caused by lens motion; performing normalization on the preprocessed features; and for the normalized compressed domain features, carrying out foreground and background segmentation by adopting a classification algorithm based on diagonal regression. Meanwhile, the invention provides an HEVC compressed domain video motion segmentation system based on the diagonal regression classifier. According to the method and the system provided by the invention, the contour of the moving object can be quickly and accurately segmented for the monitoring video containing the lens jitter, the speed is 16.78 times that of an advanced pixel domain motion segmentation method, and meanwhile, the detection performance is almost thesame.

Description

technical field [0001] The present invention relates to the technical field of video motion segmentation, in particular, to a HEVC based on a rate regression classifier (HEVC / H.265 is a new video coding standard formulated by ITU-T VCEG after AVC / H.264) Compressed domain video motion segmentation method and system. Background technique [0002] Nowadays, massive amounts of video data are generated all the time, especially in surveillance systems. Although most video data is compressed, it is still a huge challenge to analyze and store these large amounts of data. To address this challenge, motion segmentation becomes a crucial technique as it is often the first step in a video analysis system. The purpose of video segmentation is to automatically obtain a binary image for video content frames, and segment moving objects and backgrounds. Based on the motion segmentation results, further analysis of the video content can run faster and with better results. [0003] There a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & AuthorityApplications(China)
IPC IPC(8): G06T7/246G06T7/269G06T7/215G06T7/194
CPCG06T7/246G06T7/194G06T7/215G06T7/269G06T2207/10016G06T2207/20081
Inventor陈立蔡春磊杨小康安平
OwnerSHANGHAI UNIV