基于视觉识别的苹果采摘机械仿真

Simulation of Apple Picking Machinery Vased on Visual Recognition

为了解决当前苹果采摘设备识别精度差、人工采摘效率低等问题,设计了苹果采摘机械,可以显著提高苹果识别精度,加快苹果采摘效率,使经济效益最大化。通过对苹果树生长环境及生长特性进行分析研究,依据深度学习原理,选择YOLOv5-s算法对苹果进行目标检测,研制适合苹果采摘的机械设备。采摘机械设备主要由识别装置、伸缩装置、分类装置、传送装置、采摘通道、主体钢结构、采摘机械手及动力源装置等组成。Python仿真实验验证表明:通过YOLOv5-s目标检测后苹果采摘机械对苹果的识别精确度在75%以上,与RCNN、SSD等图像检测模型相比,检测精度提高了10%以上,且占用内存小、识别速度快。

In order to solve the current problems of poor recognition accuracy and low manual picking efficiency of apple picking equipment,the apple picking machinery designed in this paper can largely improve apple recognition accuracy,accelerate apple picking efficiency,and maximize economic benefits.By analyzing and researching the growth environment and growth characteristics of apple trees,and selecting YOLOv5-s algorithm for apple target detection based on the deep learning principle,we develop a suitable mechanical device for apple picking.The picking mechanical equipment is mainly composed of identification device,telescoping device,sorting device,transfer device,picking channel,main steel structure,picking manipulator,power source device,etc.It is verified through Python simulation experiments that the apple picking machinery can recognize apples with an accuracy of more than 75%after YOLOv5-s target detection,and the detection accuracy is improved by more than 10%compared with image detection models such as RCNN and SSD,and the memory consumption is small and the recognition speed is fast.