Zernike Based Scene Categorization Using Support Vector Machines

Natural scene categorization is a fundamental process of human vision that allows us to efficiently and rapidly analyze our surroundings. Thousands of images are generated every day, which implies the necessity to classify, organize and access them using an easy, faster and efficient way. Scene categorization, the classification of images into semantic categories (e.g., coast, mountains, highways and streets) is a challenging and important problem nowadays. Many different approaches concerning scene categorization have been proposed in the last few years. This paper presents a different approach to classify 'MIT-Street' and 'MIT-highway' scene categories using Zernike moments and Support Vector Machines. Radial Basis Function with p1-5 is used for the paradigm of Support Vector Machines. This work is implemented using various blocking models and the comparative results are proving the importance of blocking and the efficiency of classifier towards scene classification problems. This complete work is carried out using real world data set.