A novel technique for automatic segmentation of a brain region in single channel MR images for visualization and analysis of a human brain is presented. The method generates a volume of brain masks by automatic thresholding using a dual curve fitting technique and by 3D morphological operations. The dual curve fitting can reduce an error in curve fitting to the histogram of MR images. The 3D morphological operations, including erosion, labeling of connected-components, max-feature operation, and dilation, are applied to the cubic volume of masks reconstructed from the thresholded brain masks. This method can automatically segment a brain region in any displayed type of sequences, including extreme slices, of SPGR, T1-, T2-, and PD-weighted MR image data sets which are not required to contain the entire brain. In the experiments, the algorithm was applied to 20 sets of MR images and showed over 0.97 of similarity index in comparison with manual drawing.

We present an efficient algorithm using a region-based texture feature for the extraction of texture regions. The key idea of this algorithm is to use the variations of local correlation coefficients (LCCs) according to different orientations to classify texture and shade regions. Experimental results show that the proposed feature suitably extracts the regions that appear visually as texture regions.

This paper first presents robust algorithms to extract invariants of the linear Lie algebra model from 3D objects. In particular, an extended 3D Hough transform is presented to extract accurate estimates of the normal vectors. The Least square fitting is used to find normal vectors and representation matrices. Then an algorithm of segmentation for 3D objects is shown using the invariants of the linear Lie algebra. Distributions of invariants, both in the invariant space and on the object surface, are used for clustering and edge detection.

We consider the services of broadcasting short messages via forward signaling channels in wireless cellular networks. In the provision of such services, the negative effect of short messages on the delivery of delay-sensitive control messages must be restricted. On the other hand, it is desirable to accommodate the users' demands for service enhancements involving timeliness and informativeness. As a way to resolve such conflicting arguments, we present a generic scheme in which a short message may be rejected or removed according to the buffer occupancy at the base station and is split into a number of segments for the transmission across a forward signaling channel. However, the rejection, removal and segmentation exhibit a trade-off among several facets of service enhancements. Thus, for a quantitative evaluation of the scheme and efficient optimization of design parameters, we develop an analytical method to calculate the moments of delay times experienced by control and short messages at a base station. Using the analytical method, we investigate the delay and loss performance of control and short messages with respect to the message load and short message length.

This paper proposed an automated segmentation algorithm for MR brain images through the complementary use of T1-weighted, T2-weighted, and PD images. The proposed segmentation algorithm is composed of 3 steps. The first step involves the extraction of cerebrum images by placing a cerebrum mask over the three input images. In the second step, outstanding clusters that represent the inner tissues of the cerebrum are chosen from among the 3-dimensional (3D) clusters. The 3D clusters are determined by intersecting densely distributed parts of a 2D histogram in 3D space formed using three optimal scale images. The optimal scale image results from applying scale-space filtering to each 2D histogram and a searching graph structure. As a result, the optimal scale image can accurately describe the shape of the densely distributed pixel parts in the 2D histogram. In the final step, the cerebrum images are segmented by the FCM (Fuzzy c-means) algorithm using the outstanding cluster center value as the initial center value. The ability of the proposed segmentation algorithm to calculate the cluster center value accurately then compensates for the current limitation of the FCM algorithm, which is unduly restricted by the initial center value used. In addition, the proposed algorithm, which includes a multi spectral analysis, can achieve better segmentation results than a single spectral analysis.

Nowadays, the World Wide Web is continuing to expand at an amazing rate as a medium for conducting business in addition to disseminating information, and Web users are remarkably increasing. Human activities in virtual space as the Web are producing large volumes of data, and Web data mining to extract information from Web data has become an important research area. In this paper, we examine the features of Web log data and propose a method for transaction identification. We also introduce a new problem of user segmentation and present a method for solving this problem.

This paper proposes a new method for image segmentation and extraction using nonlinear cellular networks. Flexible segmentation of complicated natural scene images is achieved by using resistive-fuse networks, and each segmented regions is extracted by nonlinear oscillator networks. We also propose a nonlinear cellular network circuit implementing both resistive-fuse and oscillator dynamics by using pulse-modulation techniques. The basic operation of the nonlinear network circuit is confirmed by SPICE simulation. Moreover, the 1010-pixel image segmentation and extraction are demonstrated by high-speed circuit simulation.

Several methods have been developed for solving blind deconvolution problem. Recursive inverse filtering method is proposed recently and shown to have good convergence properties. This method requires accurate estimate of the region of support. In this paper, we propose to modify the original method by incorporating split, merge and grouping algorithm to find the region of support automatically.

Complex-valued region-based-coupling image clustering (continuous soft segmentation) neural networks are proposed for interferometric radar image processing. They deal with the amplitude and phase information of radar data as a combined complex-amplitude image. Thereby, not only the reflectance but also the distance (optical length) are consistently taken into account for the clustering process. A continuous complex-valued label is employed whose structure is the same as that of input raw data and estimation image. Experiments demonstrate successfully the clustering operations for interferometric synthetic aperture radar (InSAR) images. The method is applicable also to future radar systems for image acquisition in, e.g., invisible fire smoke places and intelligent transportation systems by generating a processed image more recognizable by human and automatic recognition machine.

This paper presents a novel approach to water pollution detection from remotely sensed low-platform mounted visible band camera images. We examine the feasibility of unsupervised segmentation for slick (oily spills on the water surface) region labelling. Adaptive and non adaptive filtering is combined with density modeling of the obtained textural features. A particular effort is concentrated on the textural feature extraction from raw intensity images using filter banks and adaptive feature extraction from the obtained output coefficients. Segmentation in the extracted feature space is achieved using Gaussian mixture models (GMM).

A high-speed context-free marker controlled and minima imposition-free watershed transform is proposed for efficient multi-object detection and segmentation from a complex background. The context-free markers are extracted from a complex backgrounded multi-object image using a noise tolerant attention operator. These make high speed marker-controlled watershed possible without over-segmentation and region merging. The proposed method presents a marker-constrained labeling that can speed up the segmentation of the marker-controlled watershed transform by eliminating the necessity of the minima imposition. Simulation results show that the proposed method can efficiently detect and segment multiple objects from a complex background while reducing the over-segmentation and computation time.

A framework is proposed for segmenting image textures by using Gabor filters to detect boundaries between adjacent textured regions. By performing a multi-channel filtering of the input image with a small set of adaptively selected Gabor filters, tuned to underlying textures, feature images are obtained. To reduce the variance of the filter output for better texture boundary detection, a Gaussian post-filter is applied to the Gabor filter response over each channel. Significant local variations in each channel response are detected using a gradient operator, and combined through channel grouping to produce the texture gradient. A subsequent post-processing produces expected texture boundaries. The effectiveness of the proposed technique is demonstrated through experiments on synthetic and natural textures.

Selective attention mechanism, plays an important role in human visual perception, can be investigated by developing an approach to perceiving the multi-meaningful-dotted-pattern in a color blindness plate (CBP). In this Letter, a perception model driven by a simple active vision mechanism is presented for the image segmentation and understanding of a CBP. Experiments show that to understand one meaningful pattern in an image containing multi-meaningful patterns, the active visual search (i.e., pattern attention) is a very useful function.

The first important step in pre-processing data for 3D virtual colonoscopy requires careful segmentation of a complicated shaped colon. We describe an automatic colon segmentation method with a new patient-friendly bowel preparation scheme. This new bowel preparation makes the segmentation more appropriate for digitally removing undesirable remains in the colon. With the aim of segmenting the colon accurately, we propose two techniques which can solve the partial-volume-effect (PVE) problem on the boundaries between low and high intensity regions. Based on the features of the adverse PVE voxels on the gas and fluid boundary inside the colon, our vertical filter eliminates these PVE voxels. By seriously considering the PVE on the colon boundary, our gradient-magnitude-based region growing algorithm improves the accuracy of the boundary. The result of the automatic colon segmentation method is illustrated with both extracted 2D images from the experimental volumetric abdominal CT datasets and a reconstructed 3D colon model.

A real-time adaptive segmentation method based on new distance features is proposed for the binary centroid tracker. These novel features are distances between the predicted center pixel of a target object by a tracking filter and each pixel in extraction of a moving target. The proposed method restricts clutters with target-like intensity from entering a tracking window and has low computational complexity for real-time applications compared with other complex feature-based methods. Comparative experiments show that the proposed method is superior to other segmentation methods based on the intensity feature only in target detection and tracking.

An Off-line mammography screening system is used in pre-screening mammograms to separate high-risk mammograms from most normal cases. Off-line system can run before radiologist's review and is particularly useful in the national breast cancer screening program which usually consists of high percentage of normal cases. Until now, the shortcomings of on-line detection of clustered microcalcifications from a mammogram remain in the necessity of manual selection of regions of interest. The developed technique focuses on detection of microcalcifications within a region of interest indicated by the radiologist. Therefore, this kind of system is not efficient enough to process hundreds of mammograms in a short time without a large number of radiologists. In this paper, based on a "hierarchically-coarse-to-fine" approach, an off-line mammography screening system for the detection and segmentation of clustered microcalcifications is presented. A serial off-line procedures without any human intervention should consider the complexity of organization of mammograms. In practice, it is impossible to use one technique to obtain clustered microcalcifications without consideration of background text and noises from image acquisition, the position of breast area and regions of interest. "Hierarchically-coarse-to-fine" approach is a serial procedures without any manual operations to reduce the potential areas of clustered microcalcifications from a mammogram until clustered microcalcifications are found. The reduction of potential areas starts with a mammogram, through identification of the breast area, identification of the suspicious areas of clustered microcalcifications, and finally segmentation of clustered microcalcifications. It is achieved hierarchically from coarse level to fine level. In detail, the proposed system includes breast area separation, enhancement, detection and localization of suspicious areas, segmentation of microcalcifications, and target selection of microcalcifications. The system separates its functions into hierarchical steps and follows the rule of thumb "coarse detection followed by fine segmentation" in performing each step of processing. The decomposed hierarchical steps are as follows: The system first extracts the breast region from which suspicious areas are detected. Then precise clustered microcalcification regions are segmented from the suspicious areas. For each step of operation, techniques for rough detection are first applied followed by a fine segmentation to accurately detect the boundaries of the target regions. With this "hierarchically-coarse-to-fine" approach, a complicated work such as the detection of clustered microcalcifications can be divided and conquered. The effectiveness of the system is evaluated by three experienced radiologists using two mammogram databases from the Nijmegen University Hospital and the Taichung Veterans General Hospital. Results indicate that the system can precisely extract the clustered microcalcifications without human intervention, and its performance is competitive with that of experienced radiologists, showing the system as a promising asset to radiologists.

This paper proposes a scheme which combines the conventional technique with a multi-level structure of Thai sentences for detection and segmentation for touching Thai printed characters. The proposed scheme solves problems of both horizontally and vertically touching characters. The complexity of a multi-level structure is employed to classify characters into three zones. The edge detection technique is applied to separate overlapping characters. Then, the horizontal touching characters are determined by using a statistical width of characters. The segmentation point of horizontal touching characters is determined using vertical projection combined with a statistical width of characters. The vertical touching characters are determined by considering the overlapping area of character boundary between zones. The height of line is used to separate the segment of vertical touching characters. Ambiguities are handle by using distinctive features of Thai characters. The effectiveness of the proposed scheme is tested with data from both newspapers and printed documents. The accuracy of 97 and 98 percents are obtained for newspaper and printed documents respectively.

Given N real weights w1, w2, . . . , wN stored in one-dimensional array, we consider the problem for finding an optimal interval I [1, N] under certain criteria. We shall review efficient algorithms developed for solving such problems under several optimality criteria. This problem can be naturally extended to two-dimensional case. Namely, given a NN two-dimensional array of N2 reals, the problem seeks to find a subregion of the array (e. g. , rectangular subarray R) that optimizes a certain objective function. We shall also review several algorithms for such problems. We shall also mention applications of these problems to region segmentation in image processing and to data mining.

This paper proposes a nonlinear oscillator network model for gray-level image segmentation suitable for massively parallel VLSI implementation. The model performs image segmentation in parallel using nonlinear analog dynamics. Because of the limited calculation precision in VLSI implementation, it is important to estimate the calculation precision required for proper operation. By numerical simulation, the necessary precision is estimated to be 5 bits. We propose a nonlinear oscillator network circuit using the pulse modulation approach suitable for an analog-digital merged circuit architecture. The basic operations of the nonlinear oscillator circuit and the connection weight circuit are confirmed by SPICE circuit simulation. The circuit simulation results also demonstrate that image segmentation can be performed within the order of 100 µs.

This paper proposes a Markov random field (MRF) model-based method for unsupervised segmentation of multispectral images consisting of multiple textures. To model such textured images, a hierarchical MRF is used with two layers, the first layer representing an unobservable region image and the second layer representing multiple textures which cover each region. This method uses the Expectation and Maximization (EM) method for model parameter estimation, where in order to overcome the well-noticed computational problem in the expectation step, we approximate the Baum function using mean-field-based decomposition of a posteriori probability. Given provisionally estimated parameters at each iteration in the EM method, a provisional segmentation is carried out using local a posteriori probability (LAP) of each pixel's region label, which is derived by mean-field-based decomposition of a posteriori probability of the whole region image. Experiments show that the use of LAPs is essential to perform a good image segmentation.