A Self Induced Warning System for Wild Animal Trespassing Using Machine Vision System

In India an increase in human population caused by industrial and agricultural growth has led to the conversion of the forest lands into human settlements. Animals that wait near forest-village borer for nightfall particularly to eat crops have been known to kill or harm people. Therefore there is a necessity to safeguard human livelihood without endangering the animal population by developing a suitable monitoring system. The development of a monitoring system for elephant intrusion can help to track the presence of animals over forest prone areas and avoid the chance of interference or harm of animal to human life. Animal is identified by matching the present captured image through vision based camera with the template image available. The pattern matching is done for identifying specific parts of the elephants such as ivory, trunk, ear etc. The platform used for processing is Machine Vision using LabVIEW based image processing algorithm. Cameras are fixed at places where animals usually used to enter the villages, and images are sent for processing through a wireless system. If the pattern matches with the template warning signals are produced. This system reduces the time required to detect animal presence. Thus this system becomes effective and preferable to implement. An early warning is sent to the forest officials as well as the villagers about the arrival of elephants towards forest-village prone areas. Thus, this system provides a solution for an unsupervised process for individual species identification specifically for elephants.