Animal Detection Using Image Processing For Agri

  • Dr.K.Senathipathi , Dr.A.Shobana , Alfred E Jose, K..Jayaprakash, R.Gopinath


The interaction between wild animals and people and the resultant negative impact on people or their resources, or wild animals or their habitat is “Human–wildlife conflict”. Increasing growth of human populations overlap with wildlife territory, creating reduction of resources. Due to conflict many forms of  loss of life or injury to humans, animals both wild and domesticated, occurs in particular habitat. Invade of animals in the farmland cause great loss to farmers and also to the animal. The main aim of our project is to reduce the loss and conflict between man and animal. The project uses image processing technique and embedded system . image processing technique is done using matlab which uses expectation maximization for segmentation, statistical and glcm for feature extraction and classification is done using ANN. classification is for finding empty land and presence of animal in the land. When the animal invade the agricultural land image is recorded and processed according to it result is serially sent to the controller broad from the control board a notification message is sent to the farmer regarding the entry of the animal in the farm land, buzzer sound and mild electric current is passed to the fence so that the animal will not come to the land and also land owner will take proper action for the animal to returning to the forest.