Robust object detection under various environmental conditions
Eisaku Maeda, Kenichiro Ishii, Yasuko Takahashi · Systems and Computers in Japan · 1993
Abstract An algorithm for object detection is presented. This method is robust against brightness variation and does not depend on the position, size, background pattern, shape, or color of target objects. The method consists of three processes: (1) normalize the brightness of the target and reference images; (2) calculate features, called the normalized principal component features; and (3) construct a classifier by statistical learning. The method is applied to car detection in a parking lot. Under various conditions, at least 98 percent of the cars were detected correctly. Application to moving‐car detection and person detection also is presented.