Segmentation of Medical Images Using Fuzzy Technique

Fuzzy logic has proved to yield promising results in image processing when little information about the image is available. On the other hand, fully automated segmentation is still unsolved and existing semi-automated segmentation techniques demand significant user input. Thus, an edge-based segmentation approach using fuzzy logic is hereafter presented, which does not require much user input. In comparison with other fuzzy techniques and with well-known conventional techniques, the fuzzy method has shown good results in the segmentation of Computed Tomography images of kidneys. Furthermore, a postprocessing contour-sorting technique using fuzzy rules is introduced, which avoids manual enhancement after segmentation of approximately round objects.