Design and implementation of a low cost computer vision system for sorting of closed-shell pistachio nuts

摘要

Iran is one of the most important producers and exporters of pistachio nuts in the world. Certainly, the quality of pistachio is very important for customers. Sorting of pistachio nuts is a post-harvesting process which is currently performed using a mechanical apparatus called “pin picker”. Due to the possibility of kernel damage by the pin picker, mechanical methods of closed-shell pistachio removal are gradually being replaced by other appropriate systems. Low cost systems for post-harvesting processes such as computer vision-based sorting systems are experiencing rapid growth. In this paper, a low cost intelligent system is implemented for pistachio sorting using computer vision. The system uses a combination of two flat mirrors and one low cost camera to obtain suitable 3 dimensional images from pistachios which are processed to detect closed-shell nuts. The experimental results for the three varieties of pistachios nuts shows 92.7 and 86.7% average removal accuracy, respectively for open and closed shell pistachio nuts.