Region of Interest and Redundancy Problem in Migratory Birds Wild Life Surveillance

Wetlands or humid zones are one of the most vital areas where many species of birds that maintain the balance of ecological systems. Due to global warming and change climate, rare species are threatened with extinction. It is important to preserve track and monitor these species. Most wild life monitoring systems are expensive or sub-optimal in terms of performance, deployment and network overloading. In this paper, we explore the possibility of using image processing techniques to reduce the large amount of data transmitted in traditional audio/video streaming monitoring systems. We used the region of interest technique to convey only the occurrence of a moving object. Feature extraction and matching techniques are used to deal with the redundancy and counting problem. We believe that our results show the viability of employing and using these techniques to reduce the amount of data transmitted in wild life monitoring systems.