A quality measurement and sorting system for Kiwi fruit based on automatic identification technologies

Automatic identification by machine vision is a promising technology which has been widely used in various field nowadays. Kiwi fruit is an important part of the fruit industry in Guizhou Province, Southwest China. When kiwi fruit grows ripe in large quantities, a manual sorting procedure will waste too much time to miss the best period of kiwi fruit on the market. Moreover, the cost of manual sorting is high. To solve those problems over Kiwi fruit harvest with inconsistent standard and low efficiency of macaque picking, this paper has proposed and designed a set of quality detection and sorting system for kiwi fruit based on vision sensors as well as pressure sensors. In the study, an STM32 micro-processor is as a control unit. A pressure sensors is to obtain fruit quality of kiwi and the image sensor is to obtain the size of kiwi. Afterwards, the obtained data return to the main controller to realize the classification of kiwi. The sorting system can save labor costs and improve the sorting efficiency and accuracy. The experimental results show that the proposed system can effectively complete the sorting tasks by using its cross-section area and weight, with a 98% accuracy.