AUTOMATED SORTING OF ALMONDS WITH EMBEDDED SHELL BY LASER TRANSMITTANCE IMAGING

Abstract

Shell fragments becoming embedded in kernels during hulling and shelling is a quality control problem for thealmond industry. While embedded shell is rare, with only about 0.1% of shelled kernels exhibiting this problem, the incidencehas been increasing over the past several years. The industry, therefore, needs a method to remove these kernels from theprocess stream. A prototype device was constructed which images laser light transmitted through the kernel and automaticallydetects and removes kernels with embedded shell fragments. A shell fragment blocks nearly all the transmitted light, forminga dark spot in the image that is detected by a computer algorithm. The computer then activates an air valve to divert thecorresponding kernel from the process stream. The sorting device has an inspection rate of approximately 40 kernels/s(100 kg/h). For a singlepass sorting operation, approximately 83% of the kernels with embedded shell were detected andremoved. Additionally, 11% of the clean kernels (no embedded shell) were incorrectly classified as having embedded shelland were also removed from the process stream. Running the rejects of the first sorting pass through the system a second timerecovered approximately 46% of the previously rejected clean kernels.