基于机器视觉的自动收割机控制系统分析与研究

Analysis and Research on Automatic Harvester Control System Based on Machine Vision

收割是农业生产的最后一道工序,对作物最终的产量和品质有着直接的影响。我国大部分地区的稻麦都实现了机械化收割,但收割机的控制系统相比整体水平较为落后,包括对行走速度的控制。为此,设计了基于机器视觉的收割机自动控制系统,根据作物图像中的谷粒信息计算作物密度,依照设定的喂入量对收割机行走速度进行相应的调节,以保持较高的作业效率和质量。在试验中,收割机对水稻和高密度种植小麦的作业效率较高,对大麦和低密度种植小麦的作业效率较低,需要设定较大的喂入量值。结果表明:系统从拍摄作物图像到启动步进电机的整个过程耗时1 s,可以实现对收割机行走速度的实时调节。

The high efficiency of agricultural machinery can accomplish the target task in a short time,and provide guarantee for high and stable yield.Harvesting is the last step in agricultural production and has a direct impact on crop final yield and quality.Most of the rice and wheat in our country have achieved mechanized harvest,but the control system of the harvester is relatively backward compared to the overall level,including the control of the walking speed.In this paper,the automatic control system of the harvester based on machine vision is designed.The crop density is calculated according to the grain information in the crop image,and the walking speed of the harvester is adjusted according to the set feeding amount to maintain the high working efficiency and quality.In the actual experiment,the harvester had high working efficiency for rice and high-density wheat,and the operation efficiency of barley and low-density wheat was low,and a large amount of feed was needed.The system from the crop image to start the stepper motor the whole process takes 1 s,can achieve the speed of the harvester real-time adjustment.