Description

The robust tracking of abrupt motion is a challenging task in computer vision. SVM classifier with Harr features is proposed for effective handling the local-trap problem. A new sampling based scheme is proposed to speed up object detection using adaboost feature selection process. The features were extracted using Harr like features extracted from the images. The performance of the process is measured. SVM classifier with Haar features is proposed for the tracking and identification of the moving objects in the video. The obtained features were then classified by employing SVM method and adaboost method is used for selection of the best features from the images. SVM is supervised learning method that classifies the input data vectors into two categories. In the proposed work the detected object is classified into person or Vehicle. The image pixels that were identified as background were eliminated and the other pixels were obtained. The obtained object pixels were then grouped and the identified positions were then represented in different color inorder to obtain them exactly. The object detection process is employed based on the optical flow based GMM process. The persons in the videos were tracked based on the multiple hypothesis tracking. Finally the performance of the process is measured by measuring the error rate and accuracy. The measure performance of the process shows that the proposed method is capable for the identification of the object positions in a accurate manner compared to the other existing methods used for the identification of the object positions in the video which is due to the intensity based feature values extracted from the image that denotes the object related information in a better manner.