Authors

Abstract

An automatic potato seedling film-breaking device based on machine vision was designed to reduce the labor intensity of potato film-breaking. The device mainly consists of a motion mechanism, a seedling recognition mechanism, a film-breaking mechanism, etc. Based on single-ridge and single-row potato planting agronomy, the whole structure and key component dimensions of the film-breaking device are determined, the YOLOv5s model is improved to build a potato seedling recognition system, and the film-breaking mechanism based on the parallel CoreXY structure is designed. Potato seedlings at the emerging stage are used as research objects for field film-breaking tests. The results show that the recognition rate of potato seedlings is 89.7%, the qualified rate of film-breaking is 83.5%, and the damage rate is 2.3%. This paper can provide technical support for the development of intelligent film-breaking equipment for potato seedlings.

Abstract in Chinese

为减轻马铃薯人工破膜劳动强度，设计了一种基于机器视觉的马铃薯幼苗自动破膜装置。该装置主要由运动机构、幼苗识别机构、破膜机构等组成。基于单垄单行马铃薯种植农艺，确定破膜装置的整体结构和关键部件尺寸，改进YOLOv5s模型搭建马铃薯幼苗识别系统，设计基于并联CoreXY结构的破膜机构。以出苗期马铃薯幼苗为研究对象进行田间破膜试验，结果表明：马铃薯幼苗识别率为89.7%，破膜合格率为83.5%，幼苗损伤率为2.3%。本文可为马铃薯幼苗智能化破膜装备研制提供技术支持。