MOVING OBJECT DETECTION USING SEMANTIC CONVOLUTIONAL FEATURES

Moving object detection from aerial images remains an unsolved problem in computer vision research domain. Detection results are not precise due to blurry images, thin edges and noise. Various methods were previously proposed for moving which could provide robust many challenges, i.e., noise, motion detection, lack of appropriate features, effective classification approach, complex background variations illumination. This proposes efficient method using convolutional semantic features VGG-16 use patterns facilitate each frame provides smaller area as region interest. Proposed reduces probability intensity information getting lost case same coloured the thus minimizes complexity. After that, performs distance measurement calculate linear distances frame. In this context, if there is any loss noise or illumination variation, uses Kalman filter process that by illuminating Finally, decision final determined random forest classifier feature vector generating a set probabilities class. Experimental show can detect objects efficiently, turn will decrease operating time increase rate compared previous methods.