Soft X-ray image analysis to detect wheat kernels damaged by Plodia interpunctella (Lepidoptera: Pyralidae)
Canada has a legally defined zero tolerance for stored-product insects, but infested grain is found in grain handling facilities and seed warehouses. In this study, the efficiency of a soft X-ray method to detect wheat kernels damaged by larvae of the Indianmeal moth, Plodia interpunctella (Hubner) was determined. Feeding damage by larvae results in loss of germination during storage. Undamaged Canada Western Red Spring wheat kernels and those damaged by P. interpunctella larvae were X-rayed at 15 kV and 65 µA for 3 to 5 s. Normalized histogram values were determined for wheat kernels from the X-ray images and were placed into 23 groups. The 23 histogram groups and the area of wheat kernels were used to classify undamaged and damaged wheat kernels using parametric and non-parametric classifiers. There was no significant difference between the parametric and non-parametric classifiers for the identification of undamaged and damaged wheat kernels from the independent test sets. The linear function parametric classifier was selected as the best classifier and it correctly identified more than 98% of sound and 97% of wheat kernels damaged by P. interpunctella feeding.
C. KARUNAKARAN, D. S. JAYAS, N. D.G. WHITE
Plodia interpunctella, X-ray imaging, grain damage, image analysis, insect infestation.