The study of image feature extraction and classification

Abstract

As we all know, research continues in the areas of image classification development in computer vision. People eagerly hope to achieve a perfect classification accuracy, however, many results of these experiments are less than satisfactory because of many complex factors. Therefore, in order to find these factors and improve the classification accuracy, we describe the details of classification methods with logistic regression and support vector machine algorithm, then discuss the impact of different methods on classification results and the factors that affect the classification accuracy in one method. Before that we briefly introduce the image feature extraction which plays a necessary role of image classification.