|Slaughter, David - UNIV CALIF DAVIS|
Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: August 1, 2003
Publication Date: January 1, 2004
Citation: Haff, R.P., Slaughter, D.C. 2004. Real-time x-ray inspection of wheat for infestation by the granary weevil, sitophilus granarius (l.). Transactions of the ASAE. 47(2):531-537. Interpretive Summary: An e-ray inspection system was assembled with the specific goal of imaging small agricultural commodities such as wheat for defects such as insect damage as a possible replacement for film for quality control sampling. This system produces digital x-ray images which can be displayed on a computer screen, as opposed to film which is generally observed on a light table. 1672 kernels of wheat, of which 773 were infested with granary weevils, were imaged both on film and with the new x-ray system. The resulting images were presented to human operators one kernal at a time to test how well they could identify the infestations. As expected, the recognition rates were better for the film images than the digital images. The total error for the film was only 3.0% vs 11.7% for the digital images. In both cases, the bulk of the errors were for infestations in which the weevil was in an early stage of development. In cases where the weevil was more developed, both methods resulted in nearly perfect detection. The lower recognition rates for the new system as compared to film are due to lower resolution and higher noise in the digital images than in the film. More research is called for to improve the detectors of the new system.
Technical Abstract: A high resolution x-ray imaging system was assembled using a low energy high current x-ray source, a low energy x-ray image intensifier, and a CCD camera interfaced to a PC. Overall system resolution was measured at 5 line pairs per millimeter, sufficient for identifying weevil infestations in kernels of wheat. A total of 1672 wheat kernels were imaged, both on film and with the system described above. Of the imaged kernels, 773 revealed infestations ranging from the egg to the adult life stage of the granary weevil, and 899 were non-infested. The film and the digital images were presented to human subjects to compare recognition of the infested kernels. Overall recognition error rates were 11.7% for the images from the intensifier system vs. only 3.0% for the film observations. However, when considering only infestations more advanced than the third larval instar, error rates for both methods fell below 1%.