Research and development of strawberry quality sorting machine with image processing

Abstract

The objective of this project was to create a sorting machine that could sort strawberries into five different grades based on the Royal Project Foundation’s standards, while also identifying and separating misshapen, bruised, and overripe fruits. The machine uses a combination of image processing, automation, and air-based sorting. The prototype dimensions are 1.60 x 4.0 x 1.5 m, which is suitability for industrial-scale operations. The machine categorizes strawberries based on shape, size, and quality, and it uses a webcam with a resolution of 640 x 320 pixels for image capture during sorting. The image processing software was developed by using LabVIEW 2018 as a system design software commonly used for data acquisition, control systems, and image processing tasks. The strawberries are sorted using air blowing, a non-invasive technique that ensures minimal damage to the fruit. The sorting system is controlled by a Programmable Logic Controller (PLC), allowing for automatic operation of the sorting process. The testing was conducted at three different belt speeds: 0.08, 0.10, and 0.13 m/s. The test results showed the best performance at 0.08 m/s, the machine achieved an average sorting accuracy of 93.78%. The machine’s working capacity was measured at 3,273 fruits/h, which is significantly faster than manual grading 2.17 times. The machine is notably more efficient than manual sorting, with no damage to the strawberries that ensuring quality preservation. The cost to deploy the sorting machine was 250,000 THB and the expected operational lifespan was 7-year, its break-even point in 1.94 years.