Segmentation of plant disease spots using automatic SRG algorithm: a look up table approach

Image segmentation is the key component of identifying plant leaf diseases. Most of the available techniques for leaf disease segmentation use grayscale values. In this paper, an automatic seeded region growing (SRG) algorithm for coloured images proposed by Y. Shih and S. Cheng is modified for segmentation of plant leaf diseases. The colour difference between adjacent regions is computed using Euclidean distance metric in the algorithm. This paper proposes a novel two dimensional look up table for labeling the neighbours for region merging. The look up table is created by traversing the image vertically and horizontally and any change in the labels of pixel is noted in the table. The incorporation of the table helps in better organization in region merging step and helps in further segmentation of the image. It must be noted that the performance of coloured image segmentation largely depends on the colour space chosen. The algorithm is first implemented in the YCbCr colour space and then implemented in other colour spaces like YCgCr, CIELAB and RGB to check for the best performance of the segmentation algorithm. Experimental results show that the SRG algorithm along with the proposed modification for region merging give good results in the YCbCr compared to other colour spaces for plant leaf disease segmentation.