基于RGB彩色模型的草莓图像色调分割算法

A new method of segmentation of strawberry image

针对目前草莓采摘机器人草莓图像分割运算量大、耗时多等问题,根据CIE-XYZ颜色模型及其色度图,提出了一种在RGB彩模型中进行草莓图像色调分割的方法。该方法无需彩色模型转换,时间复杂性能较Lab彩色模型下a通道阈值分割算法与BP神经网络分割算法优越。对该算法进一步改进后,只需加减运算,无需乘除运算。试验结果表明:该算法能很好地实现成熟草莓果实与图像背景的分离,并较好的保存草莓轮廓信息,分割效率>85%;进一步对分割后的图像进行形态学处理,如膨胀、腐蚀等,有效消除了孔洞现象。

Segmentation of strawberry image is a very important part of the vision system of strawberry harvesting robot. The method of segmentation in the a channel of the Lab color space and the method of segmentation using BP neural network under the HSV color space both need to change the color space, there for, need more time to compute. Base on CIE-XYZ photo of chroma, a new method that segment with hue in the RGB color space, and without using transformation was introduced. So the segment processes were speeded up. Promoting the method, it only need addi- tion and subtraction operations, thus the multipication and division operations were omitted . The result of experiments showed that expected segmentation result has been achieved by using this new method, the segmentation effect is greater than 85%. After taking further processing, such as dilation and erosion etc, the result could be more better.