Motion interpretation by analyzing difference images

Hyung–Il Choi

doi:10.1016/S1383-7621(96)00093-8

1997-03-01

PRE

Hyung–Il Choi *

Gye-Young Kim

DOI：10.1016/S1383-7621(96)00093-8

原文链接

原文求助

摘要

En 中文

We are concerned with extracting motion information from difference images. A difference image, formed by comparing gray-levels of two contiguous frames, contains moving areas. These moving areas are due to either COVering or UNCOVering a background by a moving object. We first describe how to label a moving area either COV or UNCOV. Especially, it will show how to subdivide a moving area formed by multiple moving objects, into COVs and UNCOVs of each object. We next extract motion characteristics of each COV and UNCOV region by examining distributions of intensities and topological structures. Experimental results show that the suggested method is very robust and accurate for natural as well as lab-generated images.

Keyword:

motion characteristics

intensity distribution

geometrical structure