Submitted to: Applied Engineering in Agriculture
Publication Type: Peer reviewed journal
Publication Acceptance Date: 11/1/2002
Publication Date: 3/1/2002
Citation: Pearson, T.C. 2002. Detection of Pistachio Nuts with Closed-Shells Using Impact Sounds. Applied Engineering in Agriculture. 17(2):249-253. Interpretive Summary: Pistachio nuts with closed-shells have low consumer acceptance because they are difficult to open and may contain immature kernels. Closed-shell nuts are currently separated from open-shell product by mechanical devices that can damage the kernel of open-shell nuts by inserting a needle into the kernel meat. The hole created by the needle can give the appearance of an insect tunnel, leading to rejection by the consumer. Furthermore, the mechanical separation devices used for closed-shell have a sorting error rate of about 5% to 10%, leading to good, open-shell, product being classified as closed. The development of a prototype acoustical based system to separate open-shell from closed-shell nuts is discussed. This system is able to sort nuts at a rate of 40 nuts per second and has a classification error rate of about 2%. Also, the machine vision system does not damage the kernel of open-shell nuts. The newly developed system has the opportunity to increase U.S. pistachio open-shell production by 3% to 5% resulting in increased revenue from $3.5 to $7.0 million a year while requiring a one time capital investment of approximately one million dollars.
Technical Abstract: An acoustical based sorting system was developed to separate pistachio nuts with closed-shells from those with open-shells. The system includes a microphone, digital signal processing hardware, material handling equipment, and an air reject mechanism. It was found that, upon impact with a steel plate, nuts with closed-shells emit sounds having higher amplitudes for the first 0.5 ms than nuts with open-shells. After this time, nuts with closed-shells emit sounds having lower amplitudes than those with open-shells. Linear discriminant analysis was used to classify nuts using three features extracted from the microphone signal during the first 1.4 ms after impact. The classification accuracy of this system is approximately 98%. The system has a throughput rate of approximately 40 nuts per second and cost of about $5,000 per channel. Currently, closed-shell pistachio nuts are removed by mechanical devices. These devices have a lower classification accuracy, 90% to 95%, and can damage kernels in open-shell pistachios. The newly developed system does not cause kernel damage. The increased sorting accuracy of the acoustic sorter, coupled with its low cost, enables a payback period of less than one year.