A LASER OPTICAL SYSTEM FOR NONDESTRUCTIVE DETECTION OF CORN KERNEL DEFECTS (IMAGE-PROCESSING, ACOUSTIC, SPECTROPHOTOMETRY, SCANNING ELECTRON MICROSCOPE, QUALITY EVALUATION)

Corn is subjected to numerous physical and mechanical stresses from the time of harvest until delivery to the end-user inducing defects in the kernels. The defects such as surface splits, starch cracks, chip-offs, and breakage are external and are visible. However, stress cracks, are internal defects and are not readily identifiable. In this investigation size characteristics of stress cracks were studied with the help of scanning electron microscope (SEM). The SEM photographs of typical stress cracks indicated that stress cracks originate at the center of the fluory endosperm and propagate radially toward the kernel periphery along the boundary of starch granules. The stress cracks usually do not propagate through the strong aleurone layer, the outermost layer of endosperm; and hence may not open at the surface immediately underneath the pericarp. A typical stress crack was measured to be about 58 (+OR-) 14 (mu)m wide. The stress crack width was independent of the corn genotypes, and the amount of stress-cracked kernels present in the sample.; Several nondestructive testing techniques were considered for automatic evaluation of corn kernel defects. An optical method, based on the light reflectance difference between sound and defective kernel surfaces was found suitable for detecting most defect categories. Based on this principle, an instrument was developed to examine individual kernel surfaces for defects. A helium-neon laser (632.8 nm) was used as the light source. With this instrument the broken, chipped-off, and starch-cracked kernels were detected with nearly 100% accuracy. The minor cracks and surface splits were detected with about 80% accuracy. A kernel travel angle of 80(DEGREES) to the laser beam was most suitable for detecting various kernel defects.; Difference in light reflectance was not sufficient to detect stress cracks using the instrument developed. This is because the stress cracks are internal defects whereas the light reflectance is a surface phenomenon. Ultrasonic and optical imaging processes were studied for their applicability in stress crack evaluation. The ultrasonic method was not effective because the intercellular air spaces in corn kernels presumably blocked the ultrasonic wave transmission through the kernels. The optical imaging appeared to have potential, but further investigation would be required to determine optimal lighting and viewing conditions.