A novel algorithm for detecting bacterial leaf scorch (BLS) of shade trees using image processing

Bacterial leaf scorch(BLS) is a common infection seen in different shade tree species such as American elm, red maple, sweet gum, sycamore and London plane, and a number of species of oak. The disease is caused by the bacterial pathogen Xylella fastidiosa. Spread of the disease may lead to early fall and determinate growth of the trees. The symptoms of the disease may first appear on branches and leaves and it spreads to other parts of the plant on a later stage. So detection of the infected areas of the plant leads to early analysis and elimination of disease spread and saves the plant from disease spread. Proposed scheme uses a novel approach to detect the infected areas of the plant by segmenting the plant leaves by clustering. K-means clustering is performed for different cluster centres to obtain different cluster groups of the region of interest (ROI).