Detection of citrus canker using hyperspectral reflectance imaging with spectral information divergence

Abstract

Citrus canker is one of the most devastating diseases that threaten marketability of citrus crops. This research was aimed to develop a hyperspectral imaging approach for detecting canker lesions on citrus fruit. A hyperspectral imaging system was developed for acquiring reflectance images from citrus samples in the spectral region from 450 to 930 nm. Ruby Red grapefruits with cankerous, normal and other common peel diseases including greasy spot, insect damage, melanose, scab, and wind scar were tested. Spectral information divergence (SID) classification method, which was based on quantifying the spectral similarities by using a predetermined canker reference spectrum, was performed on the hyperspectral images of the grapefruits for differentiating canker from normal fruit peels and other citrus surface conditions. The overall classification accuracy was 96.2% using an optimized SID threshold value of 0.008, which was determined under the condition that the errors of false negative and false positive were weighted equally. Considering the high economic impact of missing a cankerous fruit, zero false negative error was achieved by using a threshold value of 0.009, under which the classification accuracy was 95.2%. This research demonstrated that hyperspectral imaging technique coupled with the SID based image classification method could be used for discriminating citrus canker from other confounding diseases.