基于智能视觉识别的冬枣分选检测系统设计

针对目前我国冬枣人工分选效率低、工人劳动强度大和分级效果不准确等问题,研究一种冬枣智能分选系统。该系统基于Openmv视觉处理技术,以Arduino mega 2560单片机为控制核心,实现对冬枣快速统一的色选及大小分选。经过样机试验数据验证,该系统能够有效地实现冬枣分选功能,单通道分选速度达到60颗/s以上,分选准确率提高至96%以上,漏检和错检概率降低到4%以下,且能够减少用工数量和降低劳动强度,具有良好的推广应用前景。

Sorting and detecting system design of winter jujube based on intelligent vision recognition

In order to solve problems of low efficiency of artificial sorting, high labor picking intensity and inaccurate sorting effect of winter jujube, an intelligent sorting system of winter jujube is studied. The intelligent sorting systems is based on Openmv vision processing technology, and take Arduino Mega 2560 single chip computer as control core. Through prototype experimental data verification, the system can effectively realize the sorting function of winter jujube. The single channel sorting speed reaches more than 60 pieces per second. The sorting accuracy increased to more than 96%, the probability of missed and mistaken detection is reduced to less than 4%,and also reduce the number of workers and labor intensity, It has a good prospect of popularization and application.