(11-13 oct. 2000)] 2000 ieee workshop on signal processing systems. sips 2000. design and implementation (cat. no.00th8528) - avoiding false alarms due to illumin..文档.pdf

AVOIDING FALSE ALARMS DUE TO ILLUMINATION VARIATION I N SHOT DETECTION Dalong Li, Hanqing Lu National Lab of Pattern Recognition Institute of Automation The Chinese Academy of Sciences P.O. Box 2728 100080,P. R. China dl@ Abstract - Automatic video partitioning is the first step for content-based parsing and indexing of video data. Many methods have been introduced to address this problem, e.g. pixel-by-pixel comparisons and histogram comparisons. Histogram is robust to object motion, therefore it is used widely, but it is sensitive to illumination variation inherent in the video production process especially in TV news reports. So false alarms are inevitable. Color Ratio Histogram, which is robust to illumination changes, is proposed as frame content measure to solve the problem. However it is computationally expensive. In this paper, efficient technologies are proposed to discard the false alarms due to illumination variation. Abrupt lighting condition change lasts a very short time and the content of the image will return to the previous one before the change. Therefore they can be recognized and discarded by comparing the histograms of a certain frames. Another illumination invariant measurement ABEM is used to deal with gradual illumination variation. Compared with color ratio histogram, ABEM is more economic regarding the computing cost while remaining a well performance. Keywords: Shot detection, Histogram, ABEM, Illumination Variation. 1.INTRODUCTION