2A1-L12 Kana-kanji detection algorithm using Haar-like features(Robot Vision)

Abstract

We propose an algorithm to detect characters in images ． The proposed algorithm consists ofthree phases ； training ， candidate area detection ， and character detection ． In training Phase， a classifier to detect the ch 日racter is generated fセ om several haar − like feature − based weak classifiers by using Adaboost． In candidate area detection phase， one h馳ar − like feature is used to de丘ne the area to search charac 重 ers ． And in character detection phase， the generated Adaboost − based classi 行er is used to detect the character in the defined area ． We perfbrmed preliminary experiments to evaluate the proposed algorithm using 13 irnages ． The experimental results show that some character en homogeneQus b日ckground can be