Comparison of video image edge detection operators on red blood cells in microvasculature

This study focused on the step of preprocessing in the part of edge detection techniques to detect red blood cells (RBCs) boundaries. The objectives of the study were to make a comparison between gradient based edge detection and zero-crossing based edge detection and to find out the appropriate operator to use for edge detection of RBCs in capillary. Sobel, Robert, Prewitt and Canny were used as gradient based edge detectors whereas Laplacian of Gaussian was used as zero-crossing based egde detector. Our study used two criteria to consider the quality of edge detector by eye judgment: the probability of a false positive and the probability of a false negative. It was found that Canny and LOG are suitable to use as edge detectors for RBCs in microvasculature.