Research On Scene Image Classification Based On MSER Algorithm

With the advent of the information age, more and more images have been used to transport information. The images are the most intuitive way to describe the objects. So now researchers focus on computer vision such as object recognition and image retrieval. Since the last century, with the rapid development of digital image processing technology, a large number of digital images have emerged and spread what lead to a large amount of tedious work in image classification and retrieval. Traditional manual annotation way to manage and classify images now can’t meet people’s requirements. How to use computers to retrieve and classify images automatically and efficiently is now becoming more and more important. Scene image classification is one kind of image classification, it has become an indispensable and powerful tool in image retrieval, robots and other areas.Firstly, the background, significance and application of machine vision are introduced, and the status of scene image classification is detailed, moreover, algorithms used during the process are briefly introduced and analyzed. Secondly features often be used in the process of scene image classification are introduced, especially the color feature, texture feature and edge feature. Choose to use the texture feature to classify the scene images in this paper. Thirdly, SURF (Speeded Up Robust Features) algorithm is introduced in detailedand in order to meet the real-time requirements, a classification of feature points obtained by SURF (Speeded Up Robust Features) algorithm and combining with which BBF (best bin first) search algorithm used for feature points matching are implemented. Furthermore, the programming experimental results of the object recognition based on the improved SURF combined with BBF show that the algorithm proposed is fast and efficient in object recognizing. Fourthly, in this paper we proposed a method based on MSER-SURF algorithm to classify the scene images. Experiments show that the new method based on MSER-SURF algorithm is useful and effective.The last but not the least, a conclusion of the paper is given, and a brief analysis of the follow-up study of scene image classification is added.