基于HSV彩色空间与直方图信息的植物叶脉FFCM算法提取

Plant leaf vein extraction algorithm based on HSV color space and FFCM using histogram information

为满足植物分类和识别对植物叶片叶脉信息的需要,提出了基于HSV彩色空间与直方图信息FFCM聚类算法相结合的植物叶片叶脉提取方法。该算法可以简述为以下四个步骤:a)将植物叶片图像由RGB转换到HSV彩色空间;b)使用FFCM聚类算法实现叶片图像像素点的聚类,通过比较叶脉和叶肉像素值的均值大小对植物叶片进行颜色的分类,通过对自定义偏移量值的判定对图片进行受光度的分类;c)针对不同类别的植物叶片,分别进行去除部分叶肉的处理;d)使用FFCM算法再次聚类,在最终的聚类结果中提取叶脉像素点。实验结果表明,该方法既能有效处理和区分绿色和枯黄的叶片图像,也能很好地处理和区分受光均匀和受光不均匀的叶片图像,可以应用于植物的分类与识别。

In order to meet the needs of leaf vein information in the field of plant classification and identification,this paper proposed a method based on HSV color space and FFCM clustering algorithm using histogram information.The algorithm could be summarized into the following 4 steps.a)It converted the plant leaf image from RGB to HSV color space.b)It clustered the pixels of leaf image by FFCM,and classified the color of the plant leaves through comparing the mean of veins and mesophyll pixel values.Through the determination of custom offset values,leaves are classified by spectrophotometry of pictures.c)For different types of plant leaves,it removed part pixels of the mesophyll respectively.d)It clustered pixels by the FFCM clustering algorithm again,and extracted the vein pixels in clustering result.The experimental results show that this method can effectively deal with the images of leaves which is green or yellow,and a good result will be gained as used for images which the intensity of light falling on are not even.Therefore,this method may be a good choice applied to the classification and identification of plants.