The general goal of this work is modeling and reconstruction of a three-dimensional (3D) rat from a series of two-dimensional (2D) images captured from an inexpensive digital camera. We proposed a hybrid segmentation method that incorporates symmetric region grow (symRG) and active contour modeling (ACM) to robustly extract regions of interest (ROIs), such as organs, spines, and vessels. symRG is employed to enhance the segmentation performance while the edge information passed from the ACM can help prevent over-segmentation. We built a component-based software platform that includes the symRG and ACM components as well as the other image enhancement, post-segmentation processing, surface rendering components allowing the user to dynamically compose a streamlined 3D rat reconstruction procedure or script. The example dataset in this paper include 284 slices of 2D rat whole-body images. Separate scripts were used to model and visualize the body, heart, lung, stomach, and head. Few user-imposed parameters were required and the whole processing , from loading series of 2D images towards 3D rendition to demonstrate the results, is within two minutes.

Fuzzy Connectedness segmentation emerged in recent years as an alternative to traditional "hard" image-segmentation approaches. It employs scale-based affinity, which incorporates both fuzziness and degree of hanging-togetherness of a region, to extract regions of interest from, especially, medical images. Computation complexity has been, however, one of its arguable issues that needs further theoretical investigation and improvement. Furthermore, the homogeneity parameter needs to be specified on per image fashion. In this paper we propose an improved fuzzy connectedness segmentation method by utilizing a sequential grow-and-merge scheme that we called symmetric convolution and an adaptive thresholding technique that incorporates an entropy-guided process to determine the homogeneity parameter. The proposed approach with symmetric convolution is proven valid and efficient. We employ a simulated on-line Brain database-BrainWeb to generate the testbed to evaluate the accuracy and robustness of the proposed algorithm.

Of the many proposed image segmentation methods, region growing has been one of the most popular. Research on region growing, however, has focused primarily on the design of feature measures and on growing and merging criteria. Most of these methods have an inherent dependence on the order in which the points and regions are examined. This weakness implies that a desired segmented result is sensitive to the selection of the initial growing points. We define a set of theoretical criteria for a subclass of region-growing algorithms that are insensitive to the selection of the initial growing points. This class of algorithms, referred to as symmetric region growing algorithms, leads to a single-pass region-growing algorithm applicable to any dimensionality of images. Furthermore, they lead to region-growing algorithms that are both memory- and computation-efficient. Results illustrate the method's efficiency and its application to 3D medical image segmentation.

A new compression scheme called the hybrid compression model (HCM) is proposed for compressing sets of similar images. The HCM employs the region growing technique to partition the median image of a set of similar images; and furthermore, it uses the centroid method to characterize the original image data. The differences between the predicted and the original image data are stored and encoded for later use. The efficacy of its application on progressive transmission of similar images over the networks is also studied. The experimental results on various images show that our method provides significant improvement in compression efficiency, ranging from 5.6% to 134.9% in comparison with traditional centroid methods.

This paper discusses thinning on 3D binary images with the 4-subfield approach. Although a thinning algorithm concerns binary images, the algorithm itself can be represented as a set of three-color reduction templates. A thinning algorithm is topology preserving if the set of all three-color templates is topology preserving. Sufficient and necessary conditions of time complexity O(n) were proposed for verifying the topological soundness of a 3D 4-subfield thinning algorithm of n three-color templates. Theories and techniques for computerizing such conditions were discussed. Two 4-subfield thinning algorithms on 3D images, one for generating medial curves, and the other one for generating medial surfaces, are proposed and proved to preserve topology by our sufficient and necessary conditions.

The objective of image segmentation is to define disjoint regions of interest from a digital image. The region-growing approach among many segmentation methods employs connectivity, local homogeneity, and other image-dependent characteristics as features for segmentation. A three-dimensional CT (computed tomography) image can be formed by imaging a contrast-injected subject. The spreading scenario is similar to region growing. The intensity degradation along the contrast-spreading paths requires local-homogeneity information for better segmentation. We present the properties of a symmetric region-growing (SymRG) approach that is suitable for processing medical CT images. We review the concept and definitions of SymRG, describe its seed-invariant property and computational separability. These significant factors govern the region-growing behavior (unidirectional region growing and inter-slice merging) and computational and memory-usage efficiency. We also propose a general SymRG algorithm for any dimensional images and demonstrate experimental results.

The goal of image segmentation is to partition a digital image into disjoint regions of interest. Of the many proposed image segmentation methods, region growing has been one of the most popular. Research on region growing, however, has focused primarily on the design of feature measures and on growing and merging criteria. Most of these methods have an inherent dependence on the order in which the points and regions are examined. This weakness implies that a desired segmented result is sensitive to the selection of the initial growing points. We define a set of theoretical criteria for a subclass of region-growing algorithms that are insensitive to the selection of the initial growing points. This class of algorithms, referred to as symmetric region growing (SymRG), leads to a single-pass region-growing approach applicable to any dimensionality of images. Furthermore, they lead to region-growing algorithms that are both memory- and computation-efficient. Finally, by-products of this general paradigm are algorithms for fast connected-component labeling and cavity deletion. The paper gives theoretical results and 3-D image examples.

High-resolution micro-computed tomography (CT) scanners now exist for imaging small animals. In particular, such a scanner can generate very large three-dimensional (3-D) digital images of the rat's hepatic vasculature. These images provide data on the overall structure and function of such complex vascular trees. Unfortunately, human operators have extreme difficulty in extracting the extensive vasculature contained in the images. Also, no suitable tree representation exists that permits straight-forward structural analysis and information retrieval. This work proposes an automatic procedure for extracting and representing such a vascular tree. The procedure is both computation and memory efficient and runs on current PCs. As the results demonstrate, the procedure faithfully follows human-defined measurements and provides far more information than can be defined interactively.