Segmentation unsupervised

Although supervised and unsupervised methods have been in the past well-separated families, a number of hybrid approaches have appeared that combine both unsupervised and supervised classifiers for image segmentation purposes [61]. 

In the previous section, the classification problem was considered to be essentially that of learning how to make decisions about assigning cases to known classes. There are, however, different forms of classification problem, which may be tackled by unsupervised learning, or clustering. Unsupervised classification is appropriate when the definitions of the classes, and perhaps even the number of classes, are not known in advance, e.g., market segmentation of customers into similar groups who can then be targeted separately. 

This method are widely applied in medical imaging as a classifier, in which the weights are determined by using training data and the ANN is then used to segment new data. ANNs can also be used in unsupervised way as a clustering method. Because of the many interconnections used in a neural network, spatial information can be easily incorporated into its classification procedures [7]. 

Alfano B, Brunetti A, Covelli EM et al. (1997) Unsupervised, automated segmentation of the normal brain using a multispectral relaxometric magnetic resonance approach. Magn Reson Med 37(l) 84-93... 

Jain, A.K., Farrokhnia, F. 1991. Unsupervised texture segmentation using Gabor filter. Pattern Recognition 1991,24(12) 1167-1186. 

Hanke and Reich used Kohonen nets as a visualization tool for the analysis of protein sequence similarity (ISO). The proeedure eonverts sequenee (domains, aligned sequences, and segments of seeondary strueture) into a ehar-acteristie signal matrix. This eonversion depends on the property or replaee-ment seore vector selected by the user. The trained Kohonen network is functionally equivalent to an unsupervised nonlinear eluster analyzer. Protein families, or aligned sequences, or segments of similar seeondary strueture aggregate as clusters and their proximity may be inspeeted. 

Supervised segmentation methods work defining the different pixel classes beforehand with a series of well-identified pixels. Each different class is described by a model and these models are used to assign unknown pixels to the predefined classes. Reference pixels in and out of the classes can be selected by applying class membership masks in the image score scatter plots [74, 75], for instance, or by using results from unsupervised segmentation methods. 

---