| 摘要 |
Cells i corresponding to pixels are once initialized into a non-excitation state, to calculate coupling weights W ik between the eight cells k adjacent to the cells i, thereby determining leader cells (self-excitable cells candidate) p i =1 based on calculation results (a). Next, one leader cell yet to be excited is selected as a self-excitable cell (b). The selected self-excitable cell is put into the excitation state by self-excitation processing (c), the excitable cells are selected by excitable cell detection processing based on the coupling weights between the adjacent cells (d), and the selected cells are put into the excitation state by excitation as dependent processing (e). These operations are repeated until no excitable cell is detected any more and, if there no excitable cell is detected any more, inhibition processing is performed (f), thereby completing image segmentation of one region. These operations are repeated until there is no non-excited and non-inhibited leader cell any more, thereby pinpointing regions belonging to the same category from an input image and identifying them as an image segmentation regions. |
