|Brabec, Daniel - Dan|
Submitted to: ASAE Annual International Meeting
Publication Type: Proceedings
Publication Acceptance Date: 6/29/2002
Publication Date: 7/29/2002
Citation: PEARSON, T.C., BRABEC, D.L. AUTOMATED DETECTION OF HIDDEN INTERNAL INSECT INFESTATIONS IN WHEAT KERNELS USING ELECTRICAL CONDUCTANCE. ASAE ANNUAL INTERNATIONAL MEETING. 2002. Interpretive Summary: Internal insect infestations in wheat kernels are difficult to detect because there is usually no visible indication of the infestation until the insect has emerged from the kernel as an adult. Internal infestations degrade the end use quality of wheat and international competitiveness of U.S. wheat. Detection of internally infested wheat is important so that loads of wheat that have infestations are not blended with non-infested wheat before milling and/or marketing. Previously developed methods to automatically detect internal insect infestations in wheat have not yet become economically viable, either because of cost or accuracy. The method developed in this study utilizes a commercially available instrument for evaluating wheat physical properties, the Perten SKCS 4100. Software was developed to analyze the signals generated when wheat is processed in the SKCS to detect live internal insects. The accuracy of this method was greater than 87% recognition for kernels infested with large sized larvae or pupae and there were no non-infested kernels classified as infested.
Technical Abstract: The wheat industry is in need of an automated, economical, and rapid means to detect whole wheat kernels internally infested with insects. The feasibility of the Perten Single-Kernel Characterization System (SKCS) to detect internal insect infestations was studied. The SKCS monitors compression force and electrical conductance as individual kernels are being crushed. Samples of hard red winter wheat (HRW) and soft red winter wheat (SRW) infested with rice weevil and lesser grain borer were run through the SKCS and the conductance/crush signals saved for post-run processing. It was found that a discontinuity is often present in the conductance signal of an insect-infested kernel. An algorithm was developed to classify kernels as infested, based on features of the conductance signal. Average classification accuracies for all wheat samples were 24.5% for small-sized larvae, 62.2% for medium-sized larvae, 87.5% for large-sized larvae, and 88.6% for pupae. There were no false positives (sound kernels classified as infested). The classification algorithm is robust for a wide range of moisture contents. Classification accuracy was somewhat better for kernels infested with rice weevils than for lesser grain borer, and classification accuracy was better for HRW than for SRW.