A segmentation method for processing greenhouse vegetable foliar disease symptom images

Abstract

Uneven illumination and clutter background were the most challenging problems to segmentation of disease symptom images. In order to achieve robust segmentation, a method for processing greenhouse vegetable foliar disease symptom images was proposed in this paper. The segmentation method was based on a decision tree which was constructed by a two-step coarse-to-fine procedure. Firstly, a coarse decision tree was built by the CART (Classification and Regression Tree) algorithm with a feature subset. The feature subset consisted of color features that was selected by Pearson’s Rank correlations. Then, the coarse decision tree was optimized by pruning. Using the optimized decision tree, segmentation of disease symptom images was achieved by conducting pixel-wise classification. In order to evaluate the robustness and accuracy of the proposed method, an experiment was performed using greenhouse cucumber downy mildew images. Results showed that the proposed method achieved an overall accuracy of 90.67%, indicating that the method was able to obtain robust segmentation of disease symptom images.