基于K均值聚类的成熟草莓图像分割算法

The Ripe Strawberry Image Segmentation Algorithm Based on K-means Clustering

成熟草莓图像分割是草莓收获机器人识别和定位系统的关键技术。考虑到成熟草莓和其所处环境的颜色特性,在Lab彩色模式下将K均值聚类用于成熟草莓图像的分割。首先把输入的草莓图像从RGB空间转换到Lab空间下,然后初始化三个聚类中心进行K均值聚类的迭代算法,最后为消除成熟草莓花托表面细小瘦果对分割产生的影响,利用数学形态学的闭运算对分割的图像进行了修正加工。研究表明,K均值聚类分割算法在Lab模式下能够较好地分割出成熟草莓图像,并且Lab模式比其他彩色模式更适用于K均值聚类的图像分割算法。

Segmentation algorithm of ripe strawberry image is the key technology of recognition and positioning system of strawberry harvesting robot.Considering the color characteristics of ripe strawberry and its environment,K means clustering is used to image segmentation of strawberry in Lab color mode.Firstly,the input of strawberry image is transformed from RGB space to Lab space.Then init three cluster centers for K mean clustering algorithm.Finally,in order to eliminate the adverse effect of achene on the ripe strawberry surface,modify the segmented image by using closing operation of mathematical morphology.Studies show that,in Lab mode,strawberry image can be good segmented by K means clustering segmentation algorithm.At the same time,compared with other color mode,Lab mode is more suitable for K mean clustering image segmentation algorithm.