The Study on Video Object Segmentation Based on Object Structure and Color Analysis in Rainy Situations

碩士 === 國立高雄應用科技大學 === 電子與資訊工程研究所碩士班 === 95 === Video segmentation is key role for developing technique (e.g. index-retrieval, compression or representation) of content-based video processing. In practically, it can be implemented in pre-processing for contend-based video system in order to separate...

Full description

Bibliographic Details
Main Authors: Yan-Ting Ye, 葉彥廷
Other Authors: Thou-Ho Chen
Format: Others
Language:zh-TW
Published: 2007
Online Access:http://ndltd.ncl.edu.tw/handle/01939034144937562967
Description
Summary:碩士 === 國立高雄應用科技大學 === 電子與資訊工程研究所碩士班 === 95 === Video segmentation is key role for developing technique (e.g. index-retrieval, compression or representation) of content-based video processing. In practically, it can be implemented in pre-processing for contend-based video system in order to separate the video into many video objects. Many proposed video segmentation algorithms which are aimed at specific sequence (e.g. indoor environment) or outdoor environment in clear day. However, there restrictions hardly make it to be invalid in bad situation. A novel video segmentation algorithm to be proposed in this dissertation, it can be applied to rainy situations. This algorithm is based on change detection and also combines with colors and structure of image. First of all, we separate the pixels of foreground and background from the video sequence by using change detection. Therefore, we can obtain the initial object mask which consist of shadow region and reflect region. The shadow region is caused by lighting change; the reflect region is caused by rainy weather. The above two regions can be reduced by image structure and image color analysis. In the end, to refine the boundary of moving object by using connect component labeling and morphological operation. Experimental result shows that the foreground objects can be extracted efficiently by the proposed algorithm, especially for rainy day.