农业机器人采摘目标识别技术研究——基于FCM模糊聚类算法  

Research on Picking Target Recognition Technology of Agricultural Robot Based on FCM Fuzzy Clustering Algorithm

作　　者：冯高峰[1] Feng Gaofeng(Jiyuan Vocational and Technical College,Jiyuan 459000,China)

机构地区：[1]济源职业技术学院,河南济源459000

出　　处：《农机化研究》2024年第3期30-33,41,共5页Journal of Agricultural Mechanization Research

基　　金：河南省高等教育教学改革研究与实践重点项目(2021SJGLX675);济源职业技术学院培育课题(JYZY-2022-88)。

摘　　要：介绍了FCM(Fuzzy C-Means)模糊聚类算法的原理,采用权重分配的方法对该算法进行了改进,通过建立模糊的相似矩阵,对目标对象的特征聚类图进行分析,并引入隶属度矩阵对FCM算法进行优化,以加快算法的迭代速度。实验结果表明:农业机器人采用该方法对农作物轮廓分割识别度较高,算法计算效率较快,验证了其可靠性,该方法可用于目标农作物的分割和目标识别。It first introduces the principle of FCM(fuzzy c-means)fuzzy clustering algorithm,and improves the algorithm by using the method of weight distribution.By establishing a fuzzy similarity matrix,it analyzed the characteristic clustering diagram of the target object,introduced the membership matrix to optimize the FCM algorithm to accelerate the iterative speed of the algorithm.The experimental results show that the recognition rate of agricultural robot using the method for crop contour segmentation is high,and the algorithm calculation efficiency is fast.The reliability of this algorithm is verified,which shows that this method is suitable for target crop segmentation and target recognition.

关 键 词：农业机器人 FCM 模糊聚类 隶属度矩阵 目标识别 

分 类 号：S225[农业科学—农业机械化工程] TP242[农业科学—农业工程]