基于机器视觉的玉米收获机器人路径识别

摘要

China's corn acreage and yield are very large, occupies an important position in agriculture, but the cost of harvesting part of the amazing.Maize harvesting robot can improve the working efficiency, greatly reduce the production cost, has broad application prospects.Path recognition capability is an important aspect of robot environment adaptability.Machine vision can be used in path recognition of agricultural robots.In this paper, a path recognition method of corn harvest robot based on machine vision is designed, and a real-time image processing experiment is carried out in the field.The results show that the method has better adaptability and practicability in the field.The improvement and internal parameter optimization can provide technical support for the intelligence and informationization of corn harvest.%我国玉米种植面积和产量都很大,在农业中占有重要的地位,但收获环节耗费的成本惊人.玉米收获机器人能提高作业效率,极大地降低生产成本,具有广阔的应用前景.路径识别能力是机器人环境适应性的一个重要方面,而机器视觉主要用于农业机器人的路径识别.为此,设计了一种基于机器视觉的玉米收获机器人路径识别方法,并进行田间的实时图像处理试验.结果表明:该路径识别方法具有较好的田间适应性和实用性,经过载机结构改进和内部参数优化后能为玉米收获的智能化和信息化提供技术支撑.