Citrus greening detection using visible spectrum imaging and C-SVC

Citrus greening, also named citrus huanglongbing (HLB) is a very destructive disease in citrus production that has caused massive economy damage all over the world. Early and accurate detection of HLB infected tree is a critical management step to control the spread of this disease. However, the leaves of HLB infected citrus trees are alike the one of nutrient deficiency and other diseases. In view of this, a detection method of HLB based on visible spectrum image processing and C-SVC (cost-support vector classification) was investigated in the study. Different classes of visible images of citrus leaves (nutrition-deficient, healthy, HLB, etiolated) under natural light were collected and preprocessed to extract the texture and histograms of image color space (gray and HSI), followed by feature modeling and recognition of the existence of HLB based on C-SVC. The experimental results show that the proposed HLB recognition method achieved about 91.93% with low-cost and low-computation complexity.