Research On Classification Of Heart Sound Signal Based On Multi Scale Features

DOI

10.2991/iiicec-15.2015.268How to use a DOI?

Keywords

Heart Sound Signal; Multi Scale Features; Classification

Abstract

Since the near duplicated video detection in this study has obvious relevance, namely testing two videos are the same or nearly the same, thus the method in this article does not focus on the structural analysis of the video, but on two main points: 1) effective elimination of redundant video frames to improve the detection efficiency; 2) maintaining the granularity of the extracted key frames to make sure the accuracy of test results. Based on these two objectives, the thesis proposes an automatic dual-threshold method to eliminate redundant video frame, divides the video into several video segments, and takes the methods of adopting fixed extracted frame rate on each video segments to extract three key frames to represent the video segment on each video. This method can effectively eliminate the redundant frames of video, but also ensure extracting the granularity of key frame, which achieves good practice effect

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Lejuan Zhang
AU  - Shiyao Cui
AU  - Lu Zhang
AU  - Niqiang Li
PY  - 2015/03
DA  - 2015/03
TI  - Research On Classification Of Heart Sound Signal Based On Multi Scale Features
BT  - Proceedings of the 2015 International Industrial Informatics and Computer Engineering Conference
PB  - Atlantis Press
SP  - 1211
EP  - 1215
SN  - 2352-538X
UR  - https://doi.org/10.2991/iiicec-15.2015.268
DO  - 10.2991/iiicec-15.2015.268
ID  - Zhang2015/03
ER  -