Showing papers on "Segmentation-based object categorization published in 1993"

Image Processing: Analysis and Machine Vision

Abstract: List of Algorithms. Preface. Possible Course Outlines. 1. Introduction. 2. The Image, Its Representations and Properties. 3. The Image, Its Mathematical and Physical Background. 4. Data Structures for Image Analysis. 5. Image Pre-Processing. 6. Segmentation I. 7. Segmentation II. 8. Shape Representation and Description. 9. Object Recognition. 10. Image Understanding. 11. 3d Geometry, Correspondence, 3d from Intensities. 12. Reconstruction from 3d. 13. Mathematical Morphology. 14. Image Data Compression. 15. Texture. 16. Motion Analysis. Index.

Review of MR image segmentation techniques using pattern recognition.

Abstract: This paper has reviewed, with somewhat variable coverage, the nine MR image segmentation techniques itemized in Table II. A wide array of approaches have been discussed; each has its merits and drawbacks. We have also given pointers to other approaches not discussed in depth in this review. The methods reviewed fall roughly into four model groups: c-means, maximum likelihood, neural networks, and k-nearest neighbor rules. Both supervised and unsupervised schemes require human intervention to obtain clinically useful results in MR segmentation. Unsupervised techniques require somewhat less interaction on a per patient/image basis. Maximum likelihood techniques have had some success, but are very susceptible to the choice of training region, which may need to be chosen slice by slice for even one patient. Generally, techniques that must assume an underlying statistical distribution of the data (such as LML and UML) do not appear promising, since tissue regions of interest do not usually obey the distributional tendencies of probability density functions. The most promising supervised techniques reviewed seem to be FF/NN methods that allow hidden layers to be configured as examples are presented to the system. An example of a self-configuring network, FF/CC, was also discussed. The relatively simple k-nearest neighbor rule algorithms (hard and fuzzy) have also shown promise in the supervised category. Unsupervised techniques based upon fuzzy c-means clustering algorithms have also shown great promise in MR image segmentation. Several unsupervised connectionist techniques have recently been experimented with on MR images of the brain and have provided promising initial results. A pixel-intensity-based edge detection algorithm has recently been used to provide promising segmentations of the brain. This is also an unsupervised technique, older versions of which have been susceptible to oversegmenting the image because of the lack of clear boundaries between tissue types or finding uninteresting boundaries between slightly different types of the same tissue. To conclude, we offer some remarks about improving MR segmentation techniques. The better unsupervised techniques are too slow. Improving speed via parallelization and optimization will improve their competitiveness with, e.g., the k-nn rule, which is the fastest technique covered in this review. Another area for development is dynamic cluster validity. Unsupervised methods need better ways to specify and adjust c, the number of tissue classes found by the algorithm. Initialization is a third important area of research. Many of the schemes listed in Table II are sensitive to good initialization, both in terms of the parameters of the design, as well as operator selection of training data.(ABSTRACT TRUNCATED AT 400 WORDS)

A review of recent texture segmentation and feature extraction techniques

Abstract: The area of texture segmentation has undergone tremendous growth in recent years. There has been a great deal of activity both in the refinement of previously known approaches and in the development of completely new techniques. Although a wide variety of methodologies have been applied to this problem, there is a particularly strong concentration in the development of feature-based approaches and on the search for appropriate texture features. In this paper, we present a survey of current texture segmentation and feature extraction methods. Our emphasis is on techniques developed since 1980, particularly those with promise for unsupervised applications.

Transform coding of arbitrarily-shaped image segments

Abstract: Envisioned advanced multimedia video services include both rectangular and arbitrarily-shaped image segments. Image segments of the TV weather reporter produced by the chromo-key technique and image segments produced by video segmentation or image editing are typical examples. In this paper, we investigate efficient transform coding techniques of arbitrarily-shaped image segments. We formulate the optimal representation problem in two different domains — the full rectangular domain and the shape-projected domain. In the former, we still use the traditional rectangular transform coding method (e.g. DCT) but try to find optimal pixel values outside the segment boundary in order to make the transform spectrum as compact as possible. A simple but efficient mirror-image extension technique is proposed. In the shape-projected domain, we project the image segment and all basis functions into the subspace spanned over the image region only. Existing coding algorithms, such as orthogonal transform by Gilge [1] and iterative coding by Kaup and Aach [2], can be intuitively interpreted. To demonstrate the flexibility of the proposed formulation, we also derive a new KLT-like algorithm in the shape-projected domain. We analyze tradeoff between compression performance, computational complexity, and codec complexity for different coding schemes. Simulation results show that complicated algorithms (e.g. iterative, adaptive) can improve the quality by about 5-10 dB at some computational or hardware cost. On the other hand, the proposed simple mirror-image extension technique improves the quality by about 3-4 dB without any overheads. The contributions of this paper lie in efficient problem formulation, new transform coding techniques, and numerical tradeoff analyses. Currently, we are implementing a software program for AS image object editing and manipulation .

Adaptive split-and-merge segmentation based on piecewise least-square approximation

Abstract: The performance of the classic split-and-merge segmentation algorithm is severely hampered by its rigid split-and-merge processes, which are insensitive to the image semantics. The author proposes efficient algorithms and data structures to optimize the split-and-merge processes by piecewise least-square approximation of image intensity functions. This optimization aims at the unification of segment finding and edge detection. The optimized split-and-merge algorithm is shown to be adaptive to the image semantics and, hence, improves the segmentation validity of the previous algorithms. This algorithm also appears to work well on noisy sources. Since the optimization is done within the split-and-merge framework, the better segmentation performance is achieved at the same order of time complexity as the previous algorithms. >

Multiresolution texture segmentation with application to diagnostic ultrasound images

Abstract: A multiresolution texture segmentation (MTS) approach to image segmentation that addresses the issues of texture characterization, image resolution, and time to complete the segmentation is presented. The approach generalizes the conventional simulated annealing method to a multiresolution framework and minimizes an energy function that is dependent on the resolution of the size of the texture blocks in an image. A rigorous experimental procedure is also proposed to demonstrate the advantages of the proposed MTS approach on the accuracy of the segmentation, the efficiency of the algorithm, and the use of varying features at different resolution. Semireal images, created by sampling a series of diagnostic ultrasound images of an ovary in vitro, were tested to produce statistical measures on the performance of the approach. The ultrasound images themselves were then segmented to determine if the approach can achieve accurate results for the intended ultrasound application. Experimental results suggest that the MTS approach converges faster and produces better segmentation results than the single-level approach. >

Some Extensions of the K-Means Algorithm for Image Segmentation and Pattern Classification

Abstract: We present some extensions to the k-means algorithm for vector quantization that permit its efficient use in image segmentation and pattern classification tasks. We show that by introducing a certain set of state variables it is possible to find the representative centers of the lower dimensional manifolds that define the boundaries between classes; this permits one, for example, to find class boundaries directly from sparse data or to efficiently place centers for pattern classification. The same state variables can be used to determine adaptively the optimal number of centers for clouds of data with space-varying density. Some examples of the application of these extensions are also given.

Volumetric segmentation of range images of 3D objects using superquadric models

Abstract: The problem of part definition, description, and decomposition is central to the shape recognition systems. We present a geometric model-driven framework for segmenting dense range data of complex 3D objects into their constituent parts in terms of surface (biquadrics) and volumetric (superquadrics) primitives, without a priori domain knowledge or stored models. Surface segmentation uses a novel local-to-global iterative regression approach of searching for the best piecewise biquadric description of the data. The region adjacency information, surface discontinuities, and global shape properties are extracted from biquadrics to guide the volumetric segmentation. Superquadric models are recovered by a global-to-local residual-driven procedure, which recursively segments the scene to derive the part-structure. A set of acceptance criteria provide the objective evaluation of intermediate descriptions and decide whether to terminate the procedure, or selectively refine the segmentation. The control module generates hypotheses about superquadric models at clusters of underestimated data and performs controlled extrapolation of part-models by shrinking the global model. Results are presented for real range images of varying complexity, including objects with occluding parts, and scenes where surface segmentation is not sufficient to guide the volumetric segmentation.

Tissue classification and segmentation of MR images

Abstract: Previously reported classification or segmentation methods are reviewed, and some statistical approaches that may be capable of automatically classifying tissues and segmenting magnetic resonance (MR) images are discussed The image segmentation methods reviewed are edge detection methods and region detection methods The key feature of statistical approaches toward automatically classifying tissues and segmenting MR images is the determination of the number of image classes and the model parameters of these classes from the image data directly by a computer Any free parameter requiring extensive user interactions should be avoided Further research on the Gaussian Markov random field (GMRF) model and the MRF penalty term will push the statistical approaches further along the automatic track As these approaches become more practical they will become more valuable >

Fuzzy random fields and unsupervised image segmentation

Abstract: Statistical unsupervised image segmentation using fuzzy random fields is treated. A fuzzy model containing a hard component, which describes pure pixels, and a fuzzy component which describes mixed pixels, is introduced. A procedure for simulating, a fuzzy field based on a Gibbs sampler step followed by a second step involving white or correlated Gaussian noises is given. Then the different steps of unsupervised image segmentation are studied. Four different blind segmentation methods are performed: the conditional expectation, two variants of the maximum likelihood, and the least squares approach. The parameters required are estimated by the stochastic estimation maximization (SEM) algorithm, a stochastic variant of the expectation maximization (EM) algorithm. These fuzzy segmentation methods are compared with a classical hard segmentation method, without taking the fuzzy class into account. The study shows that the fuzzy SEM algorithm provides reliables estimators. Furthermore, fuzzy segmentation always improves upon the hard segmentation results. >

A summary of image segmentation techniques

Abstract: Machine vision systems are often considered to be composed of two subsystems: low-level vision and high-level vision. Low level vision consists primarily of image processing operations performed on the input image to produce another image with more favorable characteristics. These operations may yield images with reduced noise or cause certain features of the image to be emphasized (such as edges). High-level vision includes object recognition and, at the highest level, scene interpretation. The bridge between these two subsystems is the segmentation system. Through segmentation, the enhanced input image is mapped into a description involving regions with common features which can be used by the higher level vision tasks. There is no theory on image segmentation. Instead, image segmentation techniques are basically ad hoc and differ mostly in the way they emphasize one or more of the desired properties of an ideal segmenter and in the way they balance and compromise one desired property against another. These techniques can be categorized in a number of different groups including local vs. global, parallel vs. sequential, contextual vs. noncontextual, interactive vs. automatic. In this paper, we categorize the schemes into three main groups: pixel-based, edge-based, and region-based. Pixel-based segmentation schemes classify pixels based solely on their gray levels. Edge-based schemes first detect local discontinuities (edges) and then use that information to separate the image into regions. Finally, region-based schemes start with a seed pixel (or group of pixels) and then grow or split the seed until the original image is composed of only homogeneous regions. Because there are a number of survey papers available, we will not discuss all segmentation schemes. Rather than a survey, we take the approach of a detailed overview. We focus only on the more common approaches in order to give the reader a flavor for the variety of techniques available yet present enough details to facilitate implementation and experimentation.

Segmentation of range images: an orthogonal moment-based integrated approach

Abstract: A new approach to range image segmentation is presented. The proposed approach involves two phases in which the region and edge information detected using a set of orthogonal Zernike moment-based operators are combined to provide robust segmentation of range images. In the first phase, each range image point is characterized by the surface normal vector and the depth value at that point. A surface feature-based clustering of range image points yields its initial region-based segmentation. This initial segmentation phase often produces oversegmented images. In the second phase of the proposed technique, the oversegmented image is resegmented by appropriately merging adjacent regions using the edge information to produce final segmentation. One attractive characteristic of the proposed technique is that the same set of three moment-based operators is used to extract both surface and edge features. Thus only three convolution operations are needed at an image point to compute all the desired surface and edge features associated with that point. The performances of the proposed Zernike moment-based operators in surface and edge feature detection are theoretically analyzed. >

Segmentation of thermal images using the fuzzy C-means algorithm

Abstract: A segmentation methodology based on the fuzzy clustering algorithm is developed. The algorithm is utilized to segment a thermal image of occupants in a room taken by a thermoviewer. The purpose of segmentation is to identify the number and the positions of the occupants. Some useful applications can be realized, such as control of air-conditioning systems, security systems, and so on. The approach consists of two stages. The first stage is to distinguish occupants from a background in an image using the fuzzy C-means (FCM) algorithm. The authors have selected a suitable measure for determining the number of clusters and modified it for FCM. The purpose of the second stage is to distinguish each occupant by locating local temperature peaks in the image. A region-growing algorithm is introduced for more accurate segmentation based on the membership value determined by FCM and the number of located peaks. Some experimental results are included that relate to thermal images obtained in a meeting room. >

Document image segmentation and text area ordering

Abstract: A system for document image segmentation and ordering text areas is described and applied to both Japanese and English complex printed page layouts. There is no need to make any assumption about the shape of blocks, hence the segmentation technique can handle not only skewed images without skew-correction but also documents where column are not rectangular. In this technique, on the bottom-up strategy, the connected components are extracted from the reduced image, and classified according to their local information. The connected components are merged into lines, and lines are merged into areas. Extracted text areas are classified as body, caption, header, and footer. A tree graph of the layout of body texts is made, and we get the order of texts by preorder traversal on the graph. The authors introduce the influence range of each node, a procedure for the title part, and extraction of the white horizontal separator. Making it possible to get good results on various documents. The total system is fast and compact. >

3d morphological segmentation for image sequence processing

Abstract: The aim of this paper is to present a method for segmenting image sequences. This method is based on a three dimensional (3D) morphological segmentation. As morphological tools are very efficient in order to obtain a segmentation based on the real objects of the scene, the proposed scheme extends to three dimensions a morphological segmentation method for still images. This extension arises different problems due to the fact that we are not dealing with three dimensional objects but with a sequence of images taken at discrete instants on the time dimension. Two different solutions are proposed to deal with the 3D morphological segmentation: interpolation of the lattice of the image sequence and a modification of the basic morphological tool for segmentation: the watershed algorithm.

Segmentation-based image compression

Abstract: A segmentation-based image coding technique is described. Both uniform and textured region extraction algorithms are used for segmentation. Textured regions are reconstructed using 2-D noncausal Gaussian-Markov random field models. Uniform regions are reconstructed using polynomial expansions. An arithmetic coder is used for coding the boundaries of regions. Reasonable quality images are obtained at compression factors of 85:1.

Motion-field segmentation using an adaptive MAP criterion

Abstract: The authors propose a general formulation for adaptive, maximum a posteriori probability (MAP) segmentation of image sequences on the basis of interframe displacement and gray level information. The segmentation classifies pixel sites to independently moving objects in the scene. In this formulation, two methods for characterizing the conditional probability distribution of the data given the segmentation process are proposed. The a priori probability distribution is characterized on the basis of a Gibbsian model of the segmentation process, where a novel motion-compensated spatiotemporal neighborhood system is defined. The proposed formulation adapts to the displacement field accuracy by appropriately adjusting the relative emphasis on the estimated displacement field, gray level information, and prior knowledge implied by the Gibbsian model. Experiments have been performed with a five-frame simulated sequence containing translation and rotation. >

Unsupervised Bayesian segmentation of SAR images using the Pearson system distributions

Abstract: The authors' work deals with the unsupervised segmentation of radar images Usually the marginal distribution of each class for SAR image segmentation is supposed Gaussian or Gamma The authors propose the use of different marginal distributions in order to improve the fitness of the statistic model with the data The distributions grouped in the Pearson system provide an approximation to a wide variety of observed distributions like in radar image of the sea, ice, etc The mixture of distributions which characterizes the statistic of the image is estimated by the SEM algorithm and the segmentation is Bayesian The algorithm obtained is tested on a synthetic image and also applied to the segmentation of real SEASAT scene >

Recovering and tracking pose of curved 3D objects from 2D images

Abstract: A method of locating and tracking rigid moving objects with arbitrary curved surfaces is presented. Motion of the moving objects in a sequence of images is used to perform image segmentation and boundary extraction. The silhouette of the object model is derived by the curvature method of Basri and Ullman. The derived silhouette is then fitted to the observed silhouette to determine the object pose. Correspondence is guided by template matching, where the similarity measure is based on the minimization of the overall Euclidean distance between the derived silhouette and the observed silhouette. Bench tests and simulations confirm the viability of the approach, even when the observed silhouette is imperfect due to partial occlusion of the object or imperfect boundary extraction. >

Bayesian segmentation of MR images using 3D Gibbsian priors

Abstract: A Bayesian approach for segmentation of three-dimensional (3-D) magnetic resonance imaging (MRI) data of the human brain is presented. Connectivity and smoothness constraints are imposed on the segmentation in 3 dimensions. The resulting segmentation is suitable for 3-D display and for volumetric analysis of structures. The algorithm is based on the maximum a posteriori probability (MAP) criterion, where a 3-D Gibbs random field (GRF) is used to model the a priori probability distribution of the segmentation. The proposed method can be applied to a spatial sequence of 2-D images (cross-sections through a volume), as well as 3-D sampled data. We discuss the optimization methods for obtaining the MAP estimate. Experimental results obtained using clinical data are included.

Multi-Resolution Stochastic 3D Shape Models for Image Segmentation

Abstract: 3D Shape modeling has been a very prominent part of Computer Vision over the past decade. Several shape modeling techniques have been proposed in literature, some are local (distributed parameter) while others are global (lumped parameter) in terms of the parameters required to describe the shape. Hybrid models that combine both ends of this parameter spectrum have been in vogue only recently. However, they do not allow a smooth transition between the two extremes of this parameter spectrum.

A multi-layer Kohonen's self-organizing feature map for range image segmentation

Abstract: A self-organizing neural network for range image segmentation is proposed and described. The multi-layer Kohonen's self-organizing feature map (MLKSFM), which is an extension of the traditional single-layer Kohonen's self-organizing feature map (KSFM), is seen to alleviate the shortcomings of the latter in the context of range image segmentation. The problem of range image segmentation is formulated as one of vector quantization and is mapped onto the MLKSFM. The MLKSFM is currently implemented on the Connection Machine CM-2, which is a fine-grained single instruction multiple data (SIMD) computer. Experimental results using both synthetic and real range images are presented. >