Segmentation of Fiber Image Based on GVF Snake Model with Clustering Method

摘要

In the fiber image analysis system, correctly segmenting fiber from fiber micrograph is critical for fiber feature extraction and further identification. In this paper, the GVF snake model with the initial contour obtained by contour tracking method based on K-means clustering segmentation is proposed for fiber segmentation. Firstly, the K-means clustering method is used to obtain the initial coarse contour of fiber, and then the GVF Snake algorithm is applied to calculate the accurate fiber contour. Due to the noise of fiber image, some fiber contours have burrs, which can be removed by contour tracking method. Experiment result shows that this algorithm can obtain the boundaries of desired object from fiber image effectively and accurately, meanwhile, the new method expands apply area of the snake model to process the complicated image.