Active contour model with local prefitting bias estimation for fast image segmentation

摘要

Due to the advancement of digital image processing, image segmentation, as one of the most fundamental techniques in image processing,1-9 has become a key procedure in image identification and computer vision. Theoretically, image segmentation is a simple process that targets of interest are extracted from the background in the image by certain methods. However, due to the intensity inhomogeneity, an array of algorithms has poor performance on image segmentation. To seek better performance on image segmentation, many experts and scholars have proposed plenty of theories and methods. As an effective and representative method, the active contour model (ACM)10 has been preferred for decades. The early application dates back to 1980s when Kass et al.11 first proposed the conception of ACM, whose purpose is to convert image segmentation theory into minimization of energy functional. Afterward, Osher et al.12 utilized the level set function to represent the evolution of curves on the plane, handling theIntensity inhomogeneity, which is also called bias field, is ubiquitous in digital images. The causes of intensity inhomogeneity are complex and include uneven illumination and defects of imaging equipment. For images with local intensity inhomogeneity, an array of existing segmentation algorithms has poor performance on efficiency, accuracy, or initial robustness. To tackle this problem, an active contour model based on local prefitting bias estimation is proposed. The bias field is approximated through a new function based on a mean filtering algorithm, which can credibly represent the distribution of bias field of an input image. Then, the bias field is incorporated into the optimized energy functional based on the level set method to implement the segmentation process. Specifically, the bias field is computed before iterations and the mean filtering algorithm is much faster than traditional clustering algorithm, so the efficiency is greatly raised. Moreover, a new regularization function is formulated to improve the robustness of the initial contour and noise. Comparing with some traditional models, the proposed model achieves better results on some challenging images. (c) 2021 SPIE and IS&T [DOI: 10.1117/1.JEI.30.2.023025]