基于“淹没法”的双孢菇检测及直径测量方法

随着双孢菇种植面积和产量的增长,智能化采摘机器人的需求已迫在眉睫。为了给采摘机器人提供作业目标和参数信息,通过对双孢菇(白菇)深度图像进行处理和分析,实现了目标检测和直径测量。首先,根据双孢菇三维纵深结构特点,开发了“淹没法”双孢菇-菇床基质分割方法,实现了双孢菇的准确检测;然后,通过坐标变换,得到双孢菇菇盖边缘点的世界坐标,利用Hough圆检测实现双孢菇的直径测量。试验结果表明:此方法双孢菇检出率89%以上,漏检率低于11%,错检率低于2.26%,直径测量误差为2.15%~3.15%,单个蘑菇运算耗时约0.26~0.37s,可为双孢菇机械化采收提供在线、实时的信息决策支持,提高了双孢菇采收质量和效率,并有助于提高双孢菇采收机器人的智能化水平。

Detection and Diameter Measurement Method of Agaricus Bisporus Based on"Submerged Method"

With the growing area and production of Agaricus bisporus,the demand for intelligent picking robots is imminent.In order to provide the working target and parameter information to the picking robot,this paper has realized the target detection and diameter measurement by processing and analyzing the depth image of Agaricus bisporus.Firstly,according to the three-dimensional structure characteristics of the Agaricus bisporus,a novel method called'Submerging Method'was proposed to segment mushroom from the compost it grows on.Then,through coordinate transformation,the world coordinates of the edge points of the mushroom cover are obtained,and the diameter measurement of the mushroom is achieved by using Hough circle detection.The test results show that the positive detection rate of Agaricus bisporus in this method is more than 89%,the missed detection rate is less than 11%,the false detection rate is less than 2.26%,the diameter measurement error is 2.15%-3.15%,and the process time of a single mushroom is about 0.26-0.37s.The method proposed in this paper can provide online and real-time information decision support for the automatic harvesting of Agaricus bisporus,which can improve the quality and efficiency of Agaricus bisporus harvesting,and help to improve the intelligence level of Agaricus bisporus harvesting robot.