Eastern black walnuts (Juglans nigra L.) were collected weekly in September and October to identify a method to determine the optimal harvest date and to develop a quantifiable color classification system for kernels. Husk hardness, measured by a durometer, decreased over a 5-week period. During the collection periods, the greatest percent kernel for ‘Emma K’, ‘Kwik Krop’, and ‘Sparrow’ was recorded when durometer values (i.e., force measurements) of husks were 5.2, 5.5, and 3.4 N, respectively. Husk softening was also detected using the denting method, but the perception of denting did not consistently correspond with a specific husk hardness value. Of the three cultivars, ‘Kwik Krop’ kernels were the most difficult to visually sort into light, medium, and dark colors. Kernel LCH sum (L* + chroma + hue angle values), measured by a handheld spectrophotometer, provided a reliable color classification for all black walnut cultivars in which light kernels had LCH sums ≥150, medium kernels had sums of 149 to 126, and dark kernels had values ≤125.
Increased Performace of FPGA-Based Color Classification System
