Abstract
Image segmentation is still an open problem especially when intensities of the objects of interest are overlapped due to the presence of intensity inhomogeneities. A bias correction embedded level set model is proposed in this paper where inhomogeneities are estimated by orthogonal primary functions. First, an inhomogeneous intensity clustering energy is defined based on global distribution characteristics of the image intensities, and membership functions of the clusters described by the level set function are then introduced to define the data term energy of the proposed model. Second, a regularization term and an arc length term are also included to regularize the level set function and smooth its zero-level set contour, respectively. Third, the proposed model is extended to multichannel and multiphase patterns to segment colorful images and images with multiple objects, respectively. Experimental results and comparison with relevant models demonstrate the advantages of the proposed model in terms of bias correction and segmentation accuracy on widely used synthetic and real images and the BrainWeb and the IBSR image repositories.
Highlights
Image segmentation is a fundamental but one of the most important problems in pattern recognition and computer vision [1, 2]
If the image is corrupted by noises and inhomogeneous intensities, intensity homogeneity of the image will be destroyed due to intensity overlaps between different objects caused by the noises and biases, which certainly brings challenges to classical segmentation methods that are based upon edge detection or thresholding [14–16]
Image segmentation and bias correction are determined by the final level set function Φb and the optimal weighting coefficients Ŵ that are obtained by minimizing the energy functional E ðΦ, c, wÞ defined in Equation (19)
Summary
Image segmentation is a fundamental but one of the most important problems in pattern recognition and computer vision [1, 2]. Edge-based level set methods are good at identifying boundaries from images with strong intensity gradients They inevitably suffer from a weak boundary problem caused by the presence of intensity inhomogeneities and noises [27]. This drawback restricts their applications while in turn promotes the passion of researchers in this field to develop regionbased level set models which take statistical information of the image intensities in general as guide descriptors to identify each region of interest [28]. Another contribution of this work is that the proposed model is further extended to segment multichannel images and images with multiple objects.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
