Balaji N.V.,Sri Ramakrishna Arts College for Women |
Punithavalli M.,Sri Ramakrishna Arts College for Women
ARPN Journal of Engineering and Applied Sciences | Year: 2012
Object detection is one of the typical difficulties in computer technology which has its usage to surveillance, robotics,multimedia processing, and HCI. The multi-resolution framework is utilized by the proposed technique for object detection. Inthis efficient object detection, the lower resolution features are first used to discard the majority of negative windows at comparatively small cost, leaving a relatively small amount of windows to be processed in higher resolutions and this helps toattain better detection accuracy. Then the frameworks on Histograms of Oriented Gradient (HOG) features are used to detect the objects. For training and detection, the classifier used previously is Support Vector Machine (SVM) and Extreme Learning Machine (ELM). Modified ELM is used in the proposed technique to reduce the time for detection and improve the accuracy of classification. The input weights and hidden biases are created with the help of integrated Analytic Network Process (ANP) and BayesianNetwork (BN) model. The experimental result shows that the proposed technique achieves better detection rate when compared to the existing techniques. © 2006-2012 Asian Research Publishing Network (ARPN).