Automatic Vehicle Detection from Aerial Images using Cascaded Support Vector Machine and Gaussian Mixture Model

This paper proposes a novel approach for automatic vehicle detection from aerial images using cascaded Gaussian Mixture Model (GMM) and Support Vector Machine (SVM) algorithm. The GMM based background removal technique eliminates the image background to achieve efficient color classification using SVM classifier. The GMM classifier followed by SVM classification to ensure better results. In the proposed algorithm, the color and local features are the main cues for vehicle detection. To evaluate the performance of the proposed vehicle detection system, the metrics such as hit rate, accuracy and precision valued are used. This paper analyses the system performance and compared it with other background removal methods and classifiers.