An Efficient Region Based Object Detection method using Deep learning Algorithm

In the field of intelligent machine learning and computer vision techniques, the research of object detection has been improving step by step for several years. It also has been used across many of our real-time surveillance applications like Traffic detection, Vehicle detection, Road detection, face detection, Pedestrian detection, Fruit detection, Object tracking etc. The Roll of dataset and their size in all these detection plays a vital role. The proposed Smart-Region based detection method actively pre-process, classifies, validates and stores the new images. The SRBD method combining with YOLO-v3 algorithm achieves better results in the detection of small objects where the YOLO-v3 alone unable to give satisfactory results. Therefore the proposed SRBD method gives better results with good accuracy in vehicle detection over the other detection methods.