Abstract: A method that combines region growing and edge detection for image segmentation is presented. The authors start with a split-and merge algorithm wherein the parameters have been set up so that an over-segmented image results. Region boundaries are then eliminated or modified on the basis of criteria that integrate contrast with boundary smoothness, variation of the image gradient along the boundary, and a criterion that penalizes for the presence of artifacts reflecting the data structure used during segmentation (quadtree in this case). The algorithms were implemented in the C language on a Sun 3/160 workstation running under the Unix operating system. Simple tool images and aerial photographs were used to test the algorithms. The impression of human observers is that the method is very successful on the tool images and less so on the aerial photograph images. It is thought that the success in the tool images is because the objects shown occupy areas of many pixels, making it is easy to select parameters to separate signal information from noise. >

Abstract: We describe a three-dimensional (3D) segmentation method that comprises (a) user interactive identification of tissue classes; (b) calculation of a probability distribution for each tissue; (c) creation of a feature map of the most probable tissues; (d) 3D segmentation of the magnetic resonance (MR) data; (e) smoothing of the segmented data; (f) extraction of surfaces of interest with connectivity; (g) generation of surfaces; and (h) rendering of multiple surfaces to plan surgery. Patients with normal head anatomy and with abnormalities such as multiple sclerosis lesions and brain tumors were scanned with a 1.5 T MR system using a two echo contiguous (interleaved), multislice pulse sequence that provides both proton density and T2-weighted contrast. After the user identified the tissues, the 3D data were automatically segmented into background, facial tissue, brain matter, CSF, and lesions. Surfaces of the face, brain, lateral ventricles, tumors, and multiple sclerosis lesions are displayed using color coding and gradient shading. Color improves the visualization of segmented tissues, while gradient shading enhances the perception of depth. Manipulation of the 3D model on a workstation aids surgical planning. Sulci and gyri stand out, thus aiding functional mapping of the brain surface.

Abstract: A two-stage method of image segmentation based on gray level cooccurrence matrices is described. An analysis of the distributions within a cooccurrence matrix defines an initial pixel classification into both region and interior or boundary designations. Local consistency of pixel classification is then implemented by minimizing the entropy of local information, where region information is expressed via conditional probabilities estimated from the cooccurrence matrices, and boundary information via conditional probabilities which are determined a priori. The method robustly segments an image into homogeneous areas and generates an edge map. The technique extends easily to general edge operators. An example is given for the Canny operator. Applications to synthetic and forward-looking infrared (FLIR) images are given. >

Abstract: 1. Introduction.- 2. Preprocessing.- 3. Segmentation.- 4. Classification.- 5. Data.- 6. Control.- 7. Knowledge.- References.

Abstract: The authors empirically compare three algorithms for segmenting simple, noisy images: simulated annealing (SA), iterated conditional modes (ICM), and maximizer of the posterior marginals (MPM). All use Markov random field (MRF) models to include prior contextual information. The comparison is based on artificial binary images which are degraded by Gaussian noise. Robustness is tested with correlated noise and with object and background textured. The ICM algorithm is evaluated when the degradation and model parameters must be estimated, in both supervised and unsupervised modes and on two real images. The results are assessed by visual inspection and through a numerical criterion. It is concluded that contextual information from MRF models improves segmentation when the number of categories and the degradation model are known and that parameters can be effectively estimated. None of the three algorithms is consistently best, but the ICM algorithm is the most robust. The energy of the a posteriori distribution is not always minimized at the best segmentation. >

Abstract: A rule-based, low-level segmentation system that can automatically identify the space occupied by different structures of the brain by magnetic resonance imaging (MRI) is described. Given three-dimensional image data as a stack of slices, it can extract brain parenchyma, cerebro-spinal fluid, and high-intensity abnormalities. The multiple feature environment of MR imaging is used to comput several low-level features to enhance the separability of voxels of different structures. The population distribution of each feature is considered and a confidence function is computed whose amplitude indicates the likelihood of a voxel, with a given feature value, being a member of a class of voxels. Confidence levels are divided into a set of ranges to define notions such as highly confident, moderately confident, and least confident. The rule-based system consists of a set of sequential stages in which partially segmented binary scenes of one stage guide the next stage. Some important low-level definitions and rules for a clinical imaging protocol are presented. The system is applied to several MR images. >

Abstract: A rule-based system for automatically segmenting a document image into regions of text and nontext is presented. The initial stages of the system perform image enhancement functions such as adaptive thresholding, morphological processing, and skew detection and correction. The image segmentation process consists of smearing the original image via the run length smoothing algorithm, calculating the connected components locations and statistics, and filtering (segmenting) the image based on these statistics. The text regions can be converted (via an optical character reader) to a computer-searchable form, and the nontext regions can be extracted and preserved. The rule-based structure allows easy fine tuning of the algorithmic steps to produce robust rules, to incorporate additional tools (as they become available), and to handle special segmentation needs. >

Abstract: Several texture segmentation algorithms based on deterministic and stochastic relaxation principles, and their implementation on parallel networks, are described. The segmentation process is posed as an optimization problem and two different optimality criteria are considered. The first criterion involves maximizing the posterior distribution of the intensity field given the label field (maximum a posteriori estimate). The posterior distribution of the texture labels is derived by modeling the textures as Gauss Markov random fields (GMRFs) and characterizing the distribution of different texture labels by a discrete multilevel Markov model. A stochastic learning algorithm is proposed. This iterated hill-climbing algorithm combines fast convergence of deterministic relaxation with the sustained exploration of the stochastic algorithms, but is guaranteed to find only a local minimum. The second optimality criterion requires minimizing the expected percentage of misclassification per pixel by maximizing the posterior marginal distribution, and the maximum posterior marginal algorithm is used to obtain the corresponding solution. All these methods implemented on parallel networks can be easily extended for hierarchical segmentation; results of the various schemes in classifying some real textured images are presented. >

Abstract: A clustering scheme is used for model parameter estimation. Most of the existing clustering procedures require prior knowledge of the number of classes, which is often, as in unsupervised image segmentation, unavailable and must be estimated. This problem is known as the cluster validation problem. For unsupervised image segmentation the solution of this problem directly affects the quality of the segmentation. A model-fitting approach to the cluster validation problem based on Akaike's information criterion is proposed, and its efficacy and robustness are demonstrated through experimental results for synthetic mixture data and image data. >

Abstract: A new segmentation algorithm that can be used for robot applications is presented. The input images are dense range data of industrial parts. The image is segmented into a number of surfaces. The segmentation algorithm uses residual analysis to detect edges, then a region-growing technique is used to obtain the final segmented image. The use of the segmentation output for determining the best holdsite position and orientation of objects is studied. As compared to techniques based on intensity images, the use of range images simplifies the holdsite determination. This information can then be used to instruct the robot to grip the object and move it to the required position. The performance of the algorithm on a number of range images is presented. >

Abstract: The authors formulate the segmentation task as a search for a set of descriptions which minimally encodes a scene. A novel framework for cooperative robust estimation is used to estimate descriptions that locally provide the most savings in encoding an image. A modified Hopfield-Tank networks finds the subset of these descriptions which best describes an entire scene, accounting for occlusion and transparent overlap among individual descriptions. Using a part-based 3-D shape model the authors have implemented a system that is able to successfully segment images into their constituent structure. >

Abstract: Toboggan contrast enhancement is a noniterative single-parameter linear execution time method for selectively augmenting the contrast of multispectral images of arbitrary dimensionality. Toboggan enhancement followed by contrast segmentation is compared with adaptive smoothing, all iterative, multiple parameter, parallel approach that achieves similar results. The segmentation produced by toboggan enhancement followed by contrast segmentation appears equal in quality to that of very complex optimal regional growing segmentation methods. It is concluded that toboggan enhancement is easy to understand and manipulate and is applicable to any image (multispectral, multidimensional) for which one can define a function of local discontinuity at a pixel. >

Abstract: An adaptive segmentation system that utilizes a genetic algorithm in image segmentation. The system incorporates a closed-loop feedback mechanism in the segmentation/learning cycle. The system can adapt to changes appearing in the images being segmented, caused by variations of such factors as time and weather. Adaptation is achieved with a measure based on differences of analyzed past imagery and current imagery and on the criteria for segmentation quality. The invention is not dependent on any particular segmentation algorithm or specific sensor type.

Abstract: An iterative method is described for segmenting image sequences into independently moving regions while computing the motion parameters of each region. In each iteration, image points are classified into regions based on their consistency with the different motion estimates, and motion estimates are then updated using the obtained regions. The motion estimates and the segmentation improve with every iteration, and the iteration stops when a stable segmentation is obtained. Accurate motion parameters are recovered for each segment. The process is performed directly on gray-level images and does not require detection of special feature points and the computation of point correspondence. It is also faster and more robust than optical-flow-based segmentation methods. >

Abstract: A method of representing black and white images through the description of the boundaries of the objects that define such images is proposed. In order to obtain such a representation, this method uses several algorithms which perform boundary extraction, contour following, segmentation, pattern classification, and curve fitting. One of the advantages of this method is that the image can be reconstructed at any size. It can also be rotated or translated without losing any quality. In addition to achieving a good data-compression rate, the coding-decoding process is computationally very efficient. Also shown is the application of these algorithms to characters in order to obtain fonts that may be downloaded for modern laser printers. >

Abstract: Given a 3D range image of a scene containing multiple arbitrarily shaped objects, the authors segment the scene into homogeneous surface patches. A novel modular framework for the segmentation task is proposed. In the first module, over-segmentation is achieved using zeroth and first order local surface properties. The segmentation is then refined in the second module using high order surface representations dictated by the high level vision tasks. The procedure has been applied successfully to many range images, five of which are presented. >

Abstract: The use of gray-scale intensities together with the edge information present in a forward-looking infrared (FLIR) image to obtain a precise and accurate segmentation of a target is presented. A model of FLIR images based on gray-scale and edge information is incorporated in a gradient relaxation technique which explicitly maximizes a criterion function based on the inconsistency and ambiguity of classification of pixels with respect to their neighbors. Four variations of the basic technique which provide automatic selection of thresholds to segment FLIR images are considered. These methods are compared, and several examples of segmentation of ship images are given. >

Abstract: A texture analysis approach superior to previous ones in such aspects as classification/segmentation performance and applicability is presented. It is based on a widely adopted human visual model which hypothesizes that the human visual system (HVS) processes input pictorial signals through a set of parallel and quasi-independent mechanisms or channels. This model is referred to as the multichannel spatial filtering model (MSFM). The core of the MSFM presently applied is the recently formulated cortical channel model (CCM), which attempts to model the process of texture feature extraction in each individual channel in the MSFM. With these models, successful algorithms for both texture classification and segmentation (texture edge detection) have been developed. The algorithm for texture feature extraction and classification is compared with the conventional benchmark, i.e., the gray-level cooccurrence matrix approach, and proves to be superior in many aspects. The algorithm for texture edge detection is tested under a variety of textured images, and good segmentation results are obtained. >

Abstract: The authors introduce a general-purpose procedure for image segmentation which combines iterative image enhancement by a symmetric neighborhood filter (SNF) with an iterative, hierarchical connected component (HCC) analysis. Color vector versions of both SNF and HCC are developed, using two common kinds of metrics applied to color vector intensity differences-a simple Euclidean-type metric and a coordinate maximum-type metric. The segmentation procedures are illustrated with three-band color images of indoor and outdoor scenes. >

Abstract: In a speaker verification system, a detecting part detects a speech section of an input speech signal by using a time-series acoustic parameters thereof. A segmentation part calculates individuality information for segmentation by using the time-series acoustic parameters within the speech section, and segments the input speech section into a plurality of blocks based on the individuality information. A feature extracting part extracts features of an unknown speaker for every segmented block by using the time-series acoustic parameters. A distance calculating part calculates a distance between the features of the speaker extracted by the feature extracting part and reference features stored in a memory. A decision part makes a decision as to whether or not the unknown speaker is a real speaker by comparing the calculated distance with a predetermined threshold value. Segmentation is made by calculating a primary moment of the spectrum, over a block, and finding successive values which satisfy a predetermined criterion.

Abstract: An automated procedure that refines the nuclear contour of a previously segmented nucleus is described. The algorithm makes use of intensity information, edge magnitude information and both object and edge connectivity information. This automated procedure generates a closed contour precisely along the edge of the nucleus. The procedure was tested on a database of 3,680 red-green-blue images of thionin-SO2 and orange II-stained cervical cells obtained from normal and dysplastic samples. When used in conjunction with a simple threshold selection algorithm and an artifact removal routine, this edge relocation algorithm resulted in the correct segmentation of over 98% of the nuclei. Only 63 (1.7%) of all nuclei were incorrectly segmented.

Abstract: Consideration is given to the application of Markov random field (MRF) models to the problem of edge labeling in range images. The authors propose a segmentation algorithm which handles both jump and crease edges. The jump and crease edge likelihoods at each edge site are computed using special local operators. These likelihoods are then combined in a Bayesian framework with a MRF prior distribution on the edge labels to derive the a posterior distribution of labels. An approximation to the maximum a posteriori estimate is used to obtain the edge labelings. The edge-based segmentation has been integrated with a region-based segmentation scheme resulting in a robust surface segmentation method. >

Abstract: An algorithm is presented that integrates segmentation maps using both region and edge segmentation maps as input. The result of integration is a region map in which each region is large and compact. The operation is efficient and independent of image sources as well as segmentation techniques. The proposed algorithm allows multiple input maps and applies user-selected weights on various information sources. The scope of integration is parametrically controlled for the desired spatial resolution. A maximum likelihood estimator provides initial solutions of edge positions and strengths from multiple inputs. An iterative procedure is then used to smooth the resultant edge patterns. The edge map is converted to a region map using closed edge contours if desired. Finally, regions are merged to ensure that every region has the required properties. Experimental results are demonstrated using various segmentation techniques and real data from laser radar and thermal sensors. >

Abstract: A segmentation method developed for a system which analyzes magnetic resonance brain images is described. Consideration is also given to how the results of the segmentation method are used to successively identify the brain, suspected lesions, and the interhemispherical fissure, as well as to how these landmarks are used to determine automatically the orientation of a patient brain in 3-D. It is explained how all of the above are used in conjunction with a deformable 3-D model of the relevant brain anatomy to locate lesions in those images. The methods have been tested on more than 1000 images from 17 patients with multiple sclerosis: however, they can also be used in other radiologic tasks. >