Preprocessing Techniques for Image Analysis Applications
Hong Zhang, Atlantic State University, Georgia, hong@drake.armstrong.edu

The preprocessing of data (e.g., normalization, feature extraction, and custom kernel  function) before they are presented to a learning machine is often crucial in developing an  effective machine learning system. Even though modern machine learning methods such  as the support vector machine have the built-in mechanism of regularization,  preprocessing remains an important tool to facilitate effective training, improve inductive  bias, and incorporate prior knowledge. It is especially useful for problems with large  input dimensions and limited sample sizes such as the applications involving image  analysis. The raw image data are not a desirable form of input for learning machines.  Preprocessing is necessary in virtually any non-trivial problems of image analysis. In this  talk, examples of the preprocessing techniques and algorithms relevant to various image  analysis applications are discussed. Topics include statistical methods, clustering, image
processing techniques, and special invariant kernels.