Image Component Analyses by Various Representations of Max-Plus Algebra Based Wavelet Transform

Image component analyses by various representation of max-plus algebra based wavelet transform is proposed, where analysis and synthesis operations of the wavelets are defined by max (or min) and standard sum. In the case of proposed wavelets, various component results can be generated by changing the size/shape of the sampling window used in the analysis/synthesis process. The proposed wavelets especially can reserve edge information on compressed images due to the non-linear operations (max and min); therefore, they are efficient for predictive frames used in video compression and reconstruction. In the experiments, various image component analyses results and video compression and reconstruction by the proposed wavelets are shown.