A New Approach for Robust Segmentation of the Noisy or Textured Images

Abstract

Abstract. Segmentation of noisy or textured images remains challenging in both accuracy and computational efficiency. In this paper, we propose a new approach for segmentation of noisy or textured images that exist widely in real life. The proposed approach finds the mean values of different pixel classes more efficiently and accurately than the benchmark expectation maximization (EM) and K-means methods. With these mean values, the segmentation is achieved by clustering the pixels to its nearest mean. When too much noise is left for the presegmentation result or when textured objects are involved, we propose transforming the density distribution of labeled pixels into grayscale distribution by down-sampling the image with a bicubic function. An optimal threshold is automatically selected from the slope difference distribution of the histogram for the final segmentation. The extracted boundary is then refined by an energy minimization function with the detected edges when enough clear edges can be obtained. A large variety of images are used to validate the proposed approach, and the results verify its effectiveness in segmenting both noisy and textured images.

Unsupervised EA-Based Fuzzy Clustering for Image Segmentation

“…When the value of C i is smaller than the one of C in N i , the value of variation I i is within interval [2,3] and increases with the decrease of C i . Moreover, when C i is not smaller than C, variation I i is within interval [1,2] and increases with the increase of C i . In the real world situations, the homogenous regions of images are composed of pixels with small differences.…”

“…Then variation I i can be computed by normalizing the coefficient ξ i and protecting the normalized coefficient over interval [1,3]. When the value of C i is smaller than the one of C in N i , the value of variation I i is within interval [2,3] and increases with the decrease of C i . Moreover, when C i is not smaller than C, variation I i is within interval [1,2] and increases with the increase of C i .…”

“…Step ( 4) If max v (b) − v (b+1) < ε, then stop. Otherwise, set b = b + 1 and go to step (2). When the iteration stops, the label matrix of pixels based on fuzzy membership matrix can be obtained.…”

“…It is an essential step in many image applications. And thus a number of approaches [2], [3] have been proposed to utilize effective techniques to deal with image segmentation. Among the widely applied techniques for image segmentation, clustering [4]- [6] aims to divide a dataset into a finite set of clusters.…”