Object Recognition using SVM-KNN based on Geometric Moment Invariant
Rangarajan Muralidharan, C. Chandrasekar · 2011
In this paper, a framework for recognizing an object from the given image is discussed. The proposed method is fusion of two popular methods in the literature, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM). We propose the use of KNN to find closest neighbors to a query image and train a local SVM that preserves the distance function on the collection of neighbors. The proposed method is implemented in two steps. The first one concerns KNN to compute distances of the query to all training and pick the nearest K neighbors. The second step is to recognize the object using SVM classifier. For feature vector formation, Hu's Moment Invariant is computed to represent the image, which is invariant to translation, rotation and scaling. Experimental results are shown for COIL-100 database. Comparative analysis of proposed method with SVM and KNN is also given for each experiment.