DESIGN AND DEVELOPMENT OF ANIMAL DETECTION ALGORITHM USING IMAGE PROCESSING

Abstract

newline One serious problem that all the developed nations are facing today is death and injuries newlinedue to road accidents. The collision of an animal with the vehicle on the highway is one newlinesuch big issue apart from other issues such as vehicle over-speed, abrupt lane change, newlinedrunk-drive and others, which leads to such road accidents. In this thesis, a simple and a newlinelow-cost system for automatic animal detection on highways for preventing animal-vehicle newlinecollision using image processing and computer vision techniques is presented. A method newlinefor finding the distance of the animal in real-world units from the camera mounted vehicle newlineis also presented. The presented system is trained on more than 2200 images consisting of newlinepositive and negatives images and tested on various video clips of animals on highways newlinewith varying vehicle speed. As per the two-second rule, our developed system is able to newlinealert the driver when the vehicle speed is up to 35 kmph. Though the developed system has newlinebeen focused on automatic animal detection in context to Indian highways but it will work newlinein other countries also. The developed system can easily be implemented in any cars and newlinecan be extended to detect other animals also after proper training and testing to prevent newlinecollision of animals with vehicles on the road. newline