基于改进FCM的作物病害图像分割方法研究

Research on crop disease image segmentation based on improved FCM

针对复杂背景下作物病害叶片分割问题,提出一种改进模糊C均值聚类(Fuzzy C Means,FCM)的作物病害图像分割方法。该方法综合考虑图像的像素点的局部空间信息和灰度信息,计算出更为准确的局部空间信息,减少噪声的同时更好地保留了图像细节,从而使图像分割效果更为精确。选取黄瓜病害叶片图像进行本文算法的验证,并与其它分割方法进行比较实验。实验结果表明,本文提出的方法分割效果更好,其分割正确率为97.81%。

Focused on the problem of the lowsegmentation accuracy,a disease segmentation method is proposed based on improved FCM（ Fuzzy C Means）. The local spatial information and gray level information of the pixels of the image are considered,and more accurate local spatial information is calculated,which could reduce noise,better preserve the details of the image,thereby make the image segmentation effect more accurate. By Choosing the cucumber disease leaves to verify the algorithm of this paper,and compared with other segmentation methods,the experimental results showthat the proposed method is effective and its segmentation accuracy rate is 97.81%.