A Support Vector Machine with Gabor Features for Animal Intrusion Detection in Agriculture Fields

Abstract

Animal intrusion in agricultural fields has been a pestering problem for farmers, especially during monsoon when they try to maximize their yield. This paper puts forth an image processing and machine learning based approach to classify the animal as threat and hence alert the farmer. The image is segmented into parts using Watershed algorithm. The features are extracted from the training set by using 2D Gabor filter bank. Classification is done using Support Vector Machines algorithm. Percentage accuracy for each test image is analyzed. Training set has been increased in a step wise manner in order to find the minimum possible combination of test images and filter bank and hence increase the efficiency of the model compared to the existing models.