Neighborhood weighted fuzzy clustering image segmentation method

The invention discloses a neighborhood weighted fuzzy clustering image segmentation method. The comprises the steps: calculating the neighborhood pixel weight of each pixel in the image; fusing neighborhood pixel weights into a target function of a fuzzy C-means clustering algorithm; obtaining an objective function JNWFCM, optimizing the objective function JNWFCM to obtain a fuzzy membership degree value of each pixel enabling the objective function JNWFCM to be a local minimum value, and segmenting the image by utilizing the fuzzy membership degree value of each pixel. The method has the advantages that the sensitivity to noise when the FCM algorithm is used for image segmentation can be effectively improved, and the segmentation accuracy is obviously improved. 本发明公开一种邻域加权模糊聚类图像分割方法，计算图像中每一个像素的邻域像素权重，并将邻域像素权重融入到模糊C均值聚类算法的目标函数中，得到目标函数J，对目标函数J进行优化得到使目标函数J为局部最小值的每一个像素的模糊隶属度值，利用每一个像素的模糊隶属度值对图像进行分割。本发明的优点在于，能够有效改善FCM算法用于图像分割时对噪声的敏感性，明显提高分割的准确率。