Showing papers on "Segmentation-based object categorization published in 1988"
TL;DR: A piecewise-smooth surface model for image data that possesses surface coherence properties is used to develop an algorithm that simultaneously segments a large class of images into regions of arbitrary shape and approximates image data with bivariate functions so that it is possible to compute a complete, noiseless image reconstruction based on the extracted functions and regions.
Abstract: The solution of the segmentation problem requires a mechanism for partitioning the image array into low-level entities based on a model of the underlying image structure. A piecewise-smooth surface model for image data that possesses surface coherence properties is used to develop an algorithm that simultaneously segments a large class of images into regions of arbitrary shape and approximates image data with bivariate functions so that it is possible to compute a complete, noiseless image reconstruction based on the extracted functions and regions. Surface curvature sign labeling provides an initial coarse image segmentation, which is refined by an iterative region-growing method based on variable-order surface fitting. Experimental results show the algorithm's performance on six range images and three intensity images. >
1,151 citations
05 Dec 1988
TL;DR: An algorithm that separates the pixels in the image into clusters based on both their intensity and their clusters is developed, which performs better than the K-means algorithm and its nonadaptive extensions that incorporate spatial constraints by the use of Gibbs random fields.
Abstract: A generalization of the K-means clustering algorithm to include spatial constraints and to account for local intensity variations in the image is proposed. Spatial constraints are included by the use of a Gibbs random field model. Local intensity variations are accounted for in an iterative procedure involving averaging over a sliding window whose size decreases as the algorithm progresses. Results with an eight-neighbor Gibbs random field model applied to pictures of industrial objects and a variety of other images show that the algorithm performs better than the K-means algorithm and its nonadaptive extensions. >
247 citations
TL;DR: In this paper, a hierarchical structured segmentation algorithm is presented, which is based on the hypothesis that an area to be segmented is defined by a set of uniform motion and position parameters denoted as mapping parameters.
210 citations
05 Jun 1988
TL;DR: A method that combines region growing and edge detection for image segmentation with 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.
Abstract: The authors present a method that combines region growing and edge detection for image segmentation. They start with a split-and-merge algorithm where the parameters have been set up so that an oversegmented image results. Then region boundaries are 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). >
186 citations
TL;DR: The authors investigate the use of a priori knowledge about a scene to coordinate and control bilevel image segmentation, interpretation, and shape inspection of different objects in the scene.
Abstract: The authors investigate the use of a priori knowledge about a scene to coordinate and control bilevel image segmentation, interpretation, and shape inspection of different objects in the scene. The approach is composed of two main steps. The first step consists of proper segmentation and labeling of individual regions in the image for subsequent ease in interpretation. General as well as scene-specific knowledge is used to improve the segmentation and interpretation processes. Once every region in the image has been identified, the second step proceeds by testing different regions to ensure they meet the design requirements, which are formalized by a set of rules. Morphological techniques are used to extract certain features from the previously processed image for rule verification purposes. As a specific example, results for detecting defects in printed circuit boards are presented. >
147 citations
29 Mar 1988
TL;DR: In this article, the authors present an approach to color image understanding that can be used to segment and analyze surfaces with color variations due to highlights and shading, and apply this theory in stages to identify the object and highlight colors.
Abstract: In this paper, we present an approach to color image understanding that can be used to segment and analyze surfaces with color variations due to highlights and shading. We begin with a theory that relates the reflected light from dielectric materials, such as plastic, to fundamental physical reflection processes, and describes the color of the reflected light as a linear combination of the color of the light due to surface reflection (highlights) and body reflection (object color). This theory is used in an algorithm that separates a color image into two parts: an image of just the highlights, and the original image with the highlights removed. In the past, we have applied this method to hand-segmented images. The current paper shows how to perform automatic segmentation method by applying this theory in stages to identify the object and highlight colors. The result is a combination of segmentation and reflection analysis that is better than traditional heuristic segmentation methods (such as histogram thresholding), and provides important physical information about the surface geometry and material properties at the same time. We also show the importance of modeling the camera properties for this kind of quantitative analysis of color. This line of research cRn lead to physics-based image segmentation methods that are both more reliable and more useful than traditional segmentation methods.
83 citations
11 Apr 1988
TL;DR: A model-fitting approach to the cluster validation problem based upon Akaike's information criterion (AIC) is proposed and is shown to correctly identify the known number of clusters in the synthetically generated data and to result in good subjective segmentations in aerial photographs.
Abstract: A model-fitting approach to the cluster validation problem based upon Akaike's information criterion (AIC) is proposed. The explicit evaluation of the AIC for the image segmentation problem is achieved through an approximate maximum-likelihood-estimation algorithm. The efficacy of the proposed approach is demonstrated through experimental results for both synthetic mixture data, where the number of clusters is known, and stochastic model-based image segmentation operating on real-world images, for which the number of clusters is unknown. This approach is shown to correctly identify the known number of clusters in the synthetically generated data and to result in good subjective segmentations in aerial photographs. >
43 citations
25 Oct 1988
TL;DR: This manuscript demonstrates that X-ray CT image can be modeled by a finite normal mixture and the use of simulated and real image data demonstrate the very promising results of this proposed image segmentation approach.
Abstract: This manuscript demonstrates that X-ray CT image can be modeled by a finite normal mixture. The number of image classes in the observed image is detected by the information criteria (AIC or MDL). Parameters of the model are estimated by a modified K-mean algorithm and Bayesian decision criterion is the basis for this image segmentation approach. The use of simulated and real image data demonstrate the very promising results of this proposed technique.
28 citations
TL;DR: A unified approach to feature extraction for segmentation purposes by means of the rank-order filtering of grey values in a neighbourhood of each pixel of a digitized image is outlined.
Abstract: The aim of this paper is to outline a unified approach to feature extraction for segmentation purposes by means of the rank-order filtering of grey values in a neighbourhood of each pixel of a digitized image. In the first section an overview of rank-order filtering for image processing is given, and a fast histogram algorithm is proposed. Section 2 deals with the extraction of a “locally most representative grey value”, defined as the maximum of the local histogram density function. In Section 3 several textural features are described, which can be extracted from the local histogram by means of rank-order filtering, and their properties are discussed. Section 4 formulates some general requirements to be met by the process of image segmentation, and describes a method based upon the features introduced in the former sections. In the last section some experimental results applied to aerial views obtained with the segmentation method of Sect. 4 are reported. These test images have been analyzed within the scope of an investigation centered on terrain recognition for agricultural and ecological purposes.
27 citations
TL;DR: A coarse segmentation algorithm is presented for segmenting textured images which are composed of regions in each of which the data are modeled as one of C Markov random fields (MRFs), a maximum-likelihood (ML) segmentation.
Abstract: A coarse segmentation algorithm is presented for segmenting textured images which are composed of regions in each of which the data are modeled as one of C Markov random fields (MRFs). The segmentation sought is a maximum-likelihood (ML) segmentation. The image is partitioned into relatively small disjoint square windows. Each window is examined to see whether it is homogeneous or is mixed, and the texture region(s) that comprises the window is (are) decided by a multiple hypothesis test. The formulation of the complex ML segmentation problem in terms of this simpler window-based multiple-hypothesis problem provides huge computational savings, as ML segmentation is only performed at the windows that fall on the boundary between two regions and with the full knowledge of the two populations that are present in the window. Although the problems and solutions presented are for textured image segmentation, they are extendable to problems such as system identification, speech recognition, and data fusion. >
10 Oct 1988
TL;DR: An iterative parallel segmentation algorithm, which avoids the problem of dependency on the order in which portions of the image are processed by performing the globally best merges first, is presented.
Abstract: An iterative parallel segmentation algorithm, which avoids the problem of dependency on the order in which portions of the image are processed by performing the globally best merges first, is presented. The segmentation approach and two implementations of the approach on a massively parallel processor (MPP) are discussed. Application of the segmentation approach to data compression and image analysis is described, and results of the application are given for a Landsat thematic mapper image. >
Proceedings Article•
01 Jan 1988TL;DR: A parallel network of simple processors is proposed to find color boundaries irrespective of spatial changes in illumination, and to spread uniform colors within marked regions.
Abstract: We propose a parallel network of simple processors to find color boundaries irrespective of spatial changes in illumination, and to spread uniform colors within marked regions.
TL;DR: In this paper, a decision-directed filtering algorithm was developed for model-based segmentation and space-variant restoration of blurred images, where the spacevariant blur can be represented by a collection of L distinct point spread functions, where L is a predetermined integer, such that at each pixel one of the L point spread function will be more or less matched to the observed data.
14 Nov 1988
TL;DR: The technique presented solves the problem of texture segmentation in two steps: a hierarchical clustering algorithm related to a choice of ultrametric distances and a recursive procedure with vertical and horizontal segments of smaller and smaller size, converging towards the correct texture boundaries.
Abstract: The technique presented solves the problem of texture segmentation in two steps. In the first, a textured image is divided into small squares (20*20 in this case) and a hierarchical clustering algorithm related to a choice of ultrametric distances is used to obtain an initial segmentation. In the second step, the texture boundaries are improved using a recursive procedure with vertical and horizontal segments of smaller and smaller size, converging towards the correct texture boundaries. >
01 Jul 1988-Graphical Models \/graphical Models and Image Processing \/computer Vision, Graphics, and Image Processing
TL;DR: A new method for image segmentation and region classification based on the texture content of different regions in an image that is data-directed, computationally efficient, and operationally flexible to accomodate various textural properties and distances is presented.
Abstract: This paper presents a new method for image segmentation and region classification based on the texture content of different regions in an image. It uses a new texture information measure for the initiation of the texturally homogeneous core regions. It then uses the information measure together with a new texture distance measure, known as the event set distance, to direct the growth of various homogeneous regions. Since the texture information measure reflects both the local and global properties of an image, the segmentation process is highly adaptable to various images. The event set distance is defined over a set of grey level and gradient vector histograms derived from the texture content within image blocks. The method is data-directed, computationally efficient, and operationally flexible to accomodate various textural properties and distances. Segmentation of complex textured images has been successful.
TL;DR: In this paper, a knowledge-based system for the segmentation of seismic sections is presented, which can be functionally divided into a texture feature extraction part and a knowledge based segmentation part.
Abstract: A knowledge-based system for the segmentation of seismic sections is presented. The system can be functionally divided into a texture feature extraction part and a knowledge-based segmentation part. An important characteristic of the proposed approach is the iterative quadtree splitting (IQS) scheme used to control the segmentation process. The final output of the system is a segmentation of the input section into regions (segments) of common signal character. Test runs of the system on a real seismic section from the Gulf of Mexico show that the introduction of domain expert geologic knowledge can significantly improve the overall segmentation. The IQS control scheme provides two functions essential to most knowledge-based image processing and interpretation systems: (1) the coordination of all parallel-operated processes over the entire section for an overall balanced result; and (2) the incorporation of various types of knowledge into the different levels of decision-making in those processes. >
14 Nov 1988
TL;DR: Two methods are proposed for characterizing image analysis strategies: one from a software engineering viewpoint and the other from a knowledge representation viewpoint.
Abstract: Expert systems for image processing are classified into four categories, and their objectives, knowledge representation, reasoning methods, and shortcomings are discussed. The categories are: (1) consultation systems for image processing; (2) knowledge-based program composition systems; (3) rule-based design systems for image segmentation algorithms; and (4) goal-directed image segmentation systems. The importance of choosing effective image analysis strategies is discussed. Two methods are proposed for characterizing image analysis strategies: one from a software engineering viewpoint and the other from a knowledge representation viewpoint. Several examples are given to demonstrate the effectiveness of these methods. >
01 Jan 1988
TL;DR: A general purpose scene segmentation system based on the model that the gradient value at region borders exceeds the gradient within regions, which is also internally modular, so that another segmentation algorithm or another region formation algorithm could be included without redesigning the entire system.
Abstract: This paper introduces a general purpose scene segmentation system based on the model that the gradient value at region borders exceeds the gradient within regions. All internal and external parameters are identified and discussed, and the methods of selecting their values are specified. User-provided external parameters are based on segmentation scale: the approximate number of regions (within 50%) and typical perimeter:area ratio of objects of interest. Internal variables are assigned values adaptively, based on image data and the external parameters. The algorithm for region formation combines detected edges and a classical region growing procedure which is shown to perform better than either method alone. A confidence measure in the result is provided automatically, based on the match of the actual segmentation to the original model. Using this measure, there is confirmation whether or not the model and the external parameters are appropriate to the image data. This system is tested on many domains, including aerial photographs, small objects on plain and textured backgrounds, CT scans, stained brain tissue sections, white noise only and laser range images. The system is intended to be applied as one module in a larger vision system. The confidence measure provides a means to integrate the result of this segmentation and segmentations based on other modules. This system is also internally modular, so that another segmentation algorithm or another region formation algorithm could be included without redesigning the entire system. Comments University of Pennsylvania Department of Computer and Information Science Technical Report No. MSCIS-88-26. This technical report is available at ScholarlyCommons: https://repository.upenn.edu/cis_reports/609 Adaptive Image Segmentation Helen L. Andersonl Ruzena Bajcsy 1 Max Mintz l Computer Science Department University of Pennsylvania Philadelphia, PA 19104
29 Mar 1988
TL;DR: This paper deals with determining the most prominent features in only one sense; namely those features that probably represent man-made objects in outdoor non-urban scenes, and provides more detail on geometric guidance.
Abstract: Many techniques for segmenting images in the absence of domain specific knowledge have been described, all with marginal success. Such an approach has been shown to be intractable. In this paper, we examine a concept bridging the gap between segmentation limitations and interpretation capabilities. In incremental segmentation, no attempt is made to obtain a complete, albeit error prone, segmentation. Instead, various heuristics are used to obtain a segmentation for the most prominent features in the image. This incomplete segmentation is forwarded to the interpretation system for initial hypothesis generation. Based on the hypotheses thus generated, the interpretation system requests the generation of additional segmentation activity to verify each hypothesis. This paper deals with determining the most prominent features in only one sense; namely those features that probably represent man-made objects in outdoor non-urban scenes. Here we provide more detail on geometric guidance. KEYWORDS: segmentation, incremental segmentation, geometric structure, line structure, 2-D description.
TL;DR: A prototype rule-based system which integrates segmentation and recognition processes to analyze and classify objects in an image, quite different from the traditional image analysis paradigm which treats segmentation as a prerequisite for recognition and interpretation.
01 Jan 1988
TL;DR: Range image interpretation with a view of obtaining low-level features to guide mid-level and high-level segmentation and recognition processes is described, and various applications of surface curvatures in mid and high level recognition processes are discussed.
Abstract: Three dimensional scene analysis in an unconstrained and uncontrolled environment is the ultimate goal of computer vision. Explicit depth information about the scene is of tremendous help in segmentation and recognition of objects. Range image interpretation with a view of obtaining low-level features to guide mid-level and high-level segmentation and recognition processes is described. No assumptions about the scene are made and algorithms are applicable to any general single viewpoint range image. Low-level features like step edges and surface characteristics are extracted from the images and segmentation is performed based on individual features as well as combination of features. A high level recognition process based on superquadric fitting is described to demonstrate the usefulness of initial segmentation based on edges. A classification algorithm based on surface curvatures is used to obtain initial segmentation of the scene. Objects segmented using edge information are then classified using surface curvatures. Various applications of surface curvatures in mid and high level recognition processes are discussed. These include surface reconstruction, segmentation into convex patches and detection of smooth edges. Algorithms are run on real range images and results are discussed in detail.
11 Apr 1988
TL;DR: Two of the most important problems in scene analysis of image sequences are the segmentation of image frames into moving and nonmoving components and the 2-D motion estimation of the moving parts of the scene.
Abstract: Two of the most important problems in scene analysis of image sequences are the segmentation of image frames into moving and nonmoving components and the 2-D motion estimation of the moving parts of the scene. The algorithm has two parts: segmentation and 2-D motion estimation. These two parts are interactively connected in a mutually beneficial way. The authors concentrate on the 2-D motion estimation problem assuming that the segmentation has been performed. Two algorithms for 2-D motion estimation are presented. The first is based on the method of differentials, while the second is a boundary matching method. >
14 Nov 1988
TL;DR: Several texture analysis methods are studied, leading for each image region to a condensed textural information vector determined by stepwise discriminant analysis.
Abstract: Several texture analysis methods are studied, leading for each image region to a condensed textural information vector determined by stepwise discriminant analysis. Image segmentation is performed by split and merge, using a classification based on the Mahalanobis distance. More than 90% of the regions are well classified. >
24 Apr 1988
TL;DR: A postprocessor of the split-and-merge procedure is presented to solve two problems encountered in image segmentation: the proper setting of parameters is difficult and the result may be undergrown or overgrown, and the outlines obtained do not lie at edges.
Abstract: A postprocessor of the split-and-merge procedure is presented to solve two problems encountered in image segmentation: (1) the proper setting of parameters is difficult and the result may be undergrown (too many regions) or overgrown (too few regions); and (2) the outlines obtained do not lie at edges. The postprocessor carries out two operations, namely, boundary elimination and edge editing, to solve these two problems, respectively. The processing time of these two procedures grows about linearly with the image complexity. >
18 Jan 1988
TL;DR: The design of a new image segmentation system which uses an expert-system approach for solving this problem is described and results are illustrated by applying the system to the segmentation of actual high-resolution aerial imagery.
Abstract: Probably the most critical problem in image understanding is the segmenting images into disjoint regions of uniform gray-tone or texture. This paper describes the design of a new image segmentation system which uses an expert-system approach for solving this problem. The system is composed of two parts: (1) an image processing tool set which processes a given image by localized brightness compensation and heuristic edge extraction, and (2) a knowledge-base controller which is composed of an inference engine, a rule base, and a hypothesis base. The localized brightness compensation preprocesses the input image under control of the knowledge-base controller to obtain improved image quality. The heuristic edge extraction obtains spatially-accurate, one-pixel-wide boundaries of uniform-property subregions. These boundaries are, then, encoded using specially designed codes which facilitate the application of rules for final region formation. The combined use of the image processing tools and the knowledge-base controller makes the segmented region boundaries one pixel wide, spatially accurate, without edge gaps, and without noisy micro-edges inside segmented regions. The paper is illustrated with results by applying the system to the segmentation of actual high-resolution aerial imagery.
01 Jan 1988
TL;DR: Expert-system-guided scene segmentation of histopathologic imagery applies human understanding of segmentation problems, with suitable remedial procedures, and knowledge of the structure of the tissues to the segmentation.
Abstract: The use of model-based reasoning, combined with locally adaptive selection of segmentation procedures, has already been found productive in expert-system-guided scene segmentation of histopathologic imagery. It applies human understanding of segmentation problems, with suitable remedial procedures, and knowledge of the structure of the tissues to the segmentation. Expert-system-guided scene segmentation thus implements certain aspects of image understanding to attain robustness. For diagnostic expert systems, though, image understanding in a much broader sense is required. A pathologist's verbal description of histopathologic patterns must be related to specific information extraction and analytic processes, which are to be executed by the automated system. >