基于机器视觉的果园喷药除草机器人视觉系统设计

Design of visual systems for orchard spraying weeding robot based on machine vision

果园环境路面条件差、路面崎岖、颠簸,且果园环境的光照条件变化明显,对机器人视觉系统的图像处理单元具有高运动噪音干扰和高光照条件变化干扰的特点。当前我国泛用型除草机器人视觉系统普遍不适用于果园环境,其识别效率较低、可靠性较差,针对这一问题,提出一种基于机器视觉的果园喷药除草机器人视觉系统。该视觉系统将2G-R-B模型、OTSU阈值分割和二值形态学滤波技术综合应用于果园杂草的图像处理与识别,解决喷药除草机器人在图像高运动噪音干扰及光照条件变化干扰的果园环境下的杂草识别问题,提高果园喷药除草机器人视觉系统的图像分割可靠性及识别精度。试验结果表明:应用该视觉系统的果园喷药除草机器人在日光及夜间照明环境下的喷药成功率分别为84.5%和80.5%,喷药效果良好,且受日光光照条件影响较小,在果园喷药除草作业等方面具有实际应用价值。

The road condition of the orchard environment is rugged and bumpy,and the illumination condition of the orchard environment changes significantly,which interferes with the image processing unit of the robot vision system.At present,the visual system of spraying weeding robot in general use in China is not suitable for orchard environment because its recognition efficiency is low and reliability is poor.A visual system of orchard spraying weeding robot based on machine vision is proposed to solve the problems.The visual system will use a 2 G-R-B model.OTSU threshold segmentation and binary morphological filtering technology is applied to image processing and recognition.Orchard weeds solved the weeding robot in the image of high noise and interference light conditions change under the environment of orchard weed identification problem,improve the image segmentation reliability and recognition accuracy of the visual system of orchard spraying weeding robot.The experimental results show that the success rate of the orchard spraying weeding robot using this vision system is 84.5%and 80.5%in daylight and night lighting environment,respectively.The spraying effect is good,and it is less affected by sunlight conditions.Therefore,it has practical application value in orchard spraying weeding operation and other aspects.