Abstract
In this article, a new level set model is proposed for the segmentation of biomedical images. The image energy of the proposed model is derived from a robust image gradient feature which gives the active contour a global representation of the geometric configuration, making it more robust in dealing with image noise, weak edges, and initial configurations. Statistical shape information is incorporated using nonparametric shape density distribution, which allows the shape model to handle relatively large shape variations. The segmentation of various shapes from both synthetic and real images depict the robustness and efficiency of the proposed method. © 2013 The Authors. International Journal for Numerical Methods in Biomedical Engineering published by John Wiley & Sons, Ltd.
Highlights
Image segmentation involves the partitioning of an image such that objects of interest can be extracted from the image background
It is useful to design a robust algorithm for the automatic delineation of anatomic structures from images acquired from different imaging modalities such as magnetic resonance imaging, computed tomography imaging, and ultrasound imaging
The image-based energy is derived from the global interaction of image intensity gradient vectors. This gradient vector interaction field is known as the geometric potential field, and we have shown in [21] that its vector form can increase the robustness and efficiency of the active contours in handling image noise, challenging initialization, weak edge, and even broken object boundaries
Summary
Image segmentation involves the partitioning of an image such that objects of interest can be extracted from the image background. Geometric reconstruction from biomedical image volumes by manual labeling can be very tedious due to the sheer size of the image datasets, and the complexity and variability of the anatomic object shapes. Some of the main challenges include the extraction of object boundaries or regions from images with noise and intensity inhomogeneity, which often exist in biomedical images due to factors such as sampling artifacts and bias field. Other factors such as weak object edges, low resolution, and spatial aliasing can affect the accuracy and efficiency of the shape extraction process
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have