Title :
Pedestrian detection with an improved Adaboost
Author :
Tetik, Y.E. ; Bolat, B.
Author_Institution :
Eng. Dept., Tech. Univ., Istanbul, Turkey
Abstract :
This paper focuses on improving the performance of Adaboost (Adaptive Boosting) by using weak classifiers that make classification with a confidence score. Single thresholds and nearest neighbor classifiers are used as base classifiers. The proposed method is applied to the problem of pedestrian detection in still images. Haar-like basic features are used to construct weak classifiers.
Keywords :
Haar transforms; image classification; learning (artificial intelligence); object detection; pedestrians; Adaboost; Haar-like basic feature; base classifier; confidence score; nearest neighbor classifier; pedestrian detection; still image; Bagging; Boosting; Classification algorithms; Computer vision; Feature extraction; Prediction algorithms; Training; Adaboost; Confidence Score; Haar like basic features; Nearest Neighbour; Weak Classifiers;
Conference_Titel :
Innovations in Intelligent Systems and Applications (INISTA), 2013 IEEE International Symposium on
Conference_Location :
Albena
Print_ISBN :
978-1-4799-0659-8
DOI :
10.1109/INISTA.2013.6577626
