Unsupervised image classification, segmentation, and enhancement using ICA mixture models

• Published in: IEEE transactions on image processing Vol. 11; no. 3; pp. 270 - 279
• Main Authors: Te-Won Lee, Lewicki, M.S.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.03.2002; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Classification; Classification algorithms; Code standards; Density; Exact sciences and technology; Filling; Image classification; Image processing; Image segmentation; Image texture; Independent component analysis; Information, signal and communications theory; Layout; Mathematical models; Noise reduction; Pixel; Segmentation; Signal processing; Studies; Telecommunications and information theory; Texture; Image segmentation; Image coding; Image processing; Noise reduction; Independent component analysis; Maximum likelihood; Algorithm; Modeling; Image enhancement; Unsupervised learning; Image classification
• ISSN: 1057-7149; 1941-0042
• DOI: 10.1109/83.988960

An unsupervised classification algorithm is derived by modeling observed data as a mixture of several mutually exclusive classes that are each described by linear combinations of independent, non-Gaussian densities. The algorithm estimates the data density in each class by using parametric nonlinear functions that fit to the non-Gaussian structure of the data. This improves classification accuracy compared with standard Gaussian mixture models. When applied to images, the algorithm can learn efficient codes (basis functions) for images that capture the statistically significant structure intrinsic in the images. We apply this technique to the problem of unsupervised classification, segmentation, and denoising of images. We demonstrate that this method was effective in classifying complex image textures such as natural scenes and text. It was also useful for denoising and filling in missing pixels in images with complex structures. The advantage of this model is that image codes can be learned with increasing numbers of classes thus providing greater flexibility in modeling structure and in finding more image features than in either Gaussian mixture models or standard independent component analysis (ICA) algorithms.