A. Lorencs, J. Sinica-Sinavskis. One method of image processing and its numerical analysis. Electronics and Electrical Engineering. Kaunas; Technologija, 2010. No.7 (103), pp.25-29.
The paper discusses application of linear regression models to image transformation and pattern recognition, oriented towards creation of technology for fast processing of digital images. Theoretical and heuristic expectations are confronted with results of numerical analysis. Digital images are considered to be sets of samples taken from the surface of brightness levels. Adequate regression models, i.e. analytic expressions of functions for representation of these surfaces are sought. Images are represented by vectors of 30 coefficients of the model calculated using the least squares method. Classification of images further is based on classification results of coefficient vectors of linear regression models.