Comparison of Advanced Techniques of Image Classification

Abstract

Image classification is an important task for many aspects of global change studies and environmental applications. Digital image classification is the process of sorting all the pixels in an image into a finite number of individual classes. The conventional statistical approaches for image classification use only the gray values. Different advanced techniques in image classification like Artificial Neural Networks, Support Vector Machines, Fuzzy methods, Genetic Algorithms are being developed. A comparative study of some of these techniques for image classification is made to identify relative merits. Artificial neural networks can handle non-convex decisions. The use of textural features helps to resolve misclassification. The support vector machine (SVM) is a group of theoretically superior machine learning algorithms. It was found competitive with the best available machine learning algorithms in classifying high-dimensional data sets. Fuzzy measures show the detection of textures by analyzing the image by stochastic properties. The fundamental stochastic properties of the image will be isolated by different kinds of stochastic methods, by non-linear filtering and by non-parametric methods. In conventional support vector machines (SVMs), an n-class problem is converted into n two-class problems. To overcome n-class problem, fuzzy support vector machines (FSVMs) was proposed. Using the decision functions obtained by training the SVM, for each class, a truncated polyhedral pyramidal membership function was defined. The genetic algorithm searches a space of image processing operations for a set that can produce suitable feature planes, and a more conventional classifier which uses those feature planes to output a final classification. A comparative study of some of these techniques for image classification is made to identify relative merits. Finally the paper depicts the comparative analysis of different classification techniques with respect to several parameters.