Figure 8

Image segmentation results. The sub-images present overlapped results of adaptive threshold methods in the left column for: the Niblack method (in red), the Sauvola method (in blue), the Yasuda method (in green) and the Palumbo method (in gray) and in the right column for: the White method (in red) and the Bernsen method (in green). The top row (A and B) presents results calculated on B-channel of RGB, the middle row (C and D) presents results calculated for the brown map after colour deconvolution while the bottom row (E and F) for the “brown axis” in RGB. The other colours appearing in image should be identified according to the law of primary colour adding as overlapping outlines. The left part of each sub-image is imposed on the template what causes that inside of object there is white colour while the right part, shows the mutual localization of the detected lines on dark gray instead of black background.