Recognition of elephants in infrared images using clustering-based image segmentation

Object recognition is a challenging task in image processing and computer vision. This paper proposes a clustering-based image segmentation approach for elephant recognition. An appreciable recognition rate was achieved by k-means clustering technique followed by feature extraction and K nearest neighbour (K-NN) classifier. The k-means clustering algorithm employs the concept of fitness and belongingness to provide a more adaptive and better clustering process as compared to several conventional algorithms. Elephant shape features are extracted for the recognition. The recognition rate for each class is calculated for performance evaluation. The recognition rate for different K values in K-NN classifier is calculated to find a proper K value for the proposed design.