Showing papers on "Range segmentation published in 1988"

Segmentation through variable-order surface fitting

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. >

Integrating region growing and edge detection

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). >

A maximum likelihood approach to segmenting range data

Abstract: The problem of segmenting a range image into homogeneous regions in each of which the range data correspond to a different surface is considered. The segmentation sought is a maximum-likelihood (ML) segmentation. Only planes, cylinders, and spheres are considered as presented in the image. The basic approach to segmentation is to divide the range image into windows, classify each window as a particular surface primitive, and group like windows into surface regions. Mixed windows are detected by testing the hypothesis that a window is homogeneous. Homogeneous windows are classified according to a generalized likelihood ratio test which is computationally simple and incorporates information from adjacent windows. Grouping windows of the same surface types is cast as a weighted ML clustering problem. Finally, mixed windows are segmented using an ML hierarchical segmentation algorithm. A similar approach is taken for segmenting visible-light images of Lambertian objects illuminated by a point source at infinity. >

Textured image segmentation as a multiple hypothesis test

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. >

Range Image Segmentation for 3-D Object Recognition

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.

A Hybrid Approach to Range Image Segmentation

Abstract: One of the most significant problems arising out of the analysis of range images of 3D objects is image segmentation. This paper describes a hybrid approach to the problem, where “hybrzd” refers to a combination of both region- and edge-based considerations. We assume that the range image of an object. which may be constructed of both curved and planar surfaces, is divided into surface prz’mitives which are homogeneous in their intrinsic differential geometric propert.ies and do not contain discontinuities in either depth or surface orientation. The method is based on the computation of the Gaussian and mean curvatures from which a curvature sign map is computed. Two initial edgebased segmentations are also computed from the partial derivatives and depth values. One detects jump edges while the other highlights roof edges. The three image maps are then combined to produce the final range image segmentation. Experimental results were obtained for both synthetic and real range data of polyhedral and curved objects. Acknowledgements

An Expert System For Image Segmentation

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.

Binary morphological coding of grayscale images

Abstract: One approach to image coding is to segment the original gray-scale image and then encode the boundaries and the interiors of the segments to represent the image. The authors propose a novel image coding method based on this approach, with an important difference. With the proposed coding technique, an alternative representation of the image segments is used; instead of segment boundaries, morphological skeletons are used to represent the segments. The skeleton is a thin-lined caricature of the segment that summarizes its shape and conveys information about its size, orientation, and connectivity. A binary image morphological skeletonization procedure is used to create skeletons of the image segments. In this way, the gray-scale image is represented for coding purposes by the skeletons and the intensities of the segments. Preliminary estimates show a data rate of 0.12 bits per pixel can be obtained with this coding technique. >

An Integrated Image Segmentation/Image Analysis System

Abstract: The paper describes an integrated image segmentation/image analysis system. A segmentation algorithm which operates by tuning it's output to a pre-defined mathematical optimum is firstly outlined. Implementation of a rule-based image analysis system which makes use of the data computed during the segmentation process is then discussed. A pyramidal data structure is suggested in which the image data flows from the base upwards with the control data used in analysing the image moving in the reverse direction. It is hoped that by means of this approach the image analysis process will be capable of exerting a degree of control over the segmentation algorithm leading to a more flexible system.

Description of 3D object in range image

Abstract: A method is presented for the segmentation and description of 3-D objects in a range image. The image is segmented in two parallel ways: by region classification based on surface normal analysis and by edge detection using a relaxation process. The two results are combined by a small rule-based system to obtain a more reliable segmentation. Features of the 3D object at different levels are extracted and a surface attributes graph is constructed. This kind of view-independent description of 3D objects is effective and robust for object recognition and location. The objects considered are industrial parts composed of planes and quadric surfaces. The range image is provided by a laser scanner. >

Image segmentation and approximation through surface type labelling and region merging

Abstract: The letter presents a two-stage image segmentation and approximation method for an image coding application. The proposed method is based on the surface fitting of an intensity image by a second-order polynomial function. Experimental results show its much improved performance compared with the method of Kocher-Leonardi.

PIPS: A Procedure For Interactive Pyramid Segmentation

Abstract: The Procedure for Interactive Pyramid Segementation was designed to identify regions of spatially connected and spectrally homogeneous pixels in multispectral image data, and to allow these regions to be interactively manipulated without the use of processing parameters. The objective is to provide the user with the capability to easily extract and identify regions corresponding to target objects of interest. The approach is to segment a multispectral image into a set of regions and to allow the analyst to interactively refine the segmentation by direct manipulation. The user can elect to: interactively display maps of the spatial distribution of regions for any designated image subset; display the statistics of a given region; and label, merge, or split regions.

Range Image Segmentation and Classification via Split-and-Merge Based on Surface Curvature

Abstract: Surface curvatures have been widely used for the purpose of segmenting and classifying range images. However, since surface curvatures include the second order derivatives, they become unreliable in the presence of noise, yielding false segmentation and/or classification of the range images. In this paper we investigate the sensitivity of the surface curvatures to the noise, and provide some observations on the characteristics of surface curvatures in the presence of noise. Following these observations, we then propose a scheme for reliable range image segmentation and classification. This scheme first differentiate planar region from curved region using planarity test. Surface curvatures are then computed only from the points belonging to the curved region, and mean curvature sign image (MCSI) and Gaussian curvature sign image (GCSI) are generated. Segmentation is done using split-and-merge operations on a quadtree representation of the MCSI. The sign of the Gaussian curvature is incorporated with the segmented MCSI to give a classification of the region. In the preliminary classification stage, the distribution of the Gaussian curvature signs are considered for each segmented region of the segmented MCSI. In the secondary classification stage, unclassified regions from the preliminary classification are classified using split-and-merge operations based on predominant Gaussian curvature signs.

A hybrid to range image detection

Abstract: An approach to the problem of range image segmentation is described. It is a hybrid of region- and edge-based approaches. It is assumed that the range image of an object, which may be constructed of both curved and planar surfaces, is divided into surface primitives which are homogeneous in their intrinsic differential geometric properties and do not contain discontinuities in either depth or surface orientation. The method is based on the computation of the Gaussian and mean curvatures from which a curvature sign map is computed. Two initial edge-based segmentations are also computed from the partial derivatives and depth values. One detects jump edges while the other highlights roof edges. The three image maps are then combined to produce the final range image segmentation. Experimental results for both synthetic and real range data of polyhedral and curved objects have proved the usefulness of the approach. Results for the real data are given. >

Geometry guided radiograph segmentation

Abstract: The author's overall goal is to develop an image understanding system for automatically interpreting dental radiographs. A description is given of the module that integrates the intrinsic image data to form the region adjacency graph that represents the image. Classical segmentation algorithms will not always yield correct results, since blurred edges can cause adjacent anatomical regions to be labeled as one region. The authors' solution is to guide the segmentation by intrinsic properties of the constituent objects, using a connected-components-like algorithm. Their experiments show that for dental radiographs a segmentation using gray-level data in conjunction with edges of the surfaces of teeth gives a robust and reliable segmentation. >

Segmentation Via Multiple Looks Of Image Sequences Using The A Priori Probability Mask

Abstract: A multiple-look segmentation technique for processing image sequences which uses a probability mask is presented. By designing the segmentation process in two stages; the probability mask formation, and the refinement by the Maximum A Posteriori(MAP) decision rule, a more reliable silhouette of a moving object is obtained for classification. Experimental results are presented showing the a priori probability mask formation and a comparison of the results of the Maximum Likelihood(ML) decision rule with those of the MAP decision rule. Here we assume that geometry(or shape) does not change within the "window" of looks required to perform the segmentation.

Image segmentation technique used in estimation of the size distribution of rock fragments in mining

Abstract: In this paper we describe a method for segmenting the image of fragmented ore into regions and estimating a size distrubution of the ore fragments. As the image grey level is inadequate for determining the rock boundary, we propose a two-step method for segmentation. First step, apply a region growing algorithm to construct a primary region division from the image grey level. Second step, use a set of rules to examine the region shapes and correct the division errors. Features of the regions, which are relative to the rock volume, are then measured. The method is simple and economic, while the precision is acceptable. within the rock extent. Segmentation method based on neighborhood grey level will not be able to outline each rock. Some progress has already been achieved in analyzing the ore fragment image. There are systems and methods proposed in U.S.A., Democratic Germany, Sweden et al. The University of Lulea (in Sweden) developed a system for automatic processing of fragment image

The Unification Of Region And Edge Information For Image Segmentation

Abstract: An initial pixel classification into both region class and interior or boundary designation is made based on an analysis of the distributions within the grey level cooccurrence matrices of an image. Local consistency of classification is then implemented by minimising the local entropy of region and boundary information. This is a robust way of simultaneously segmenting an image into texturally homogeneous areas with a thresholded edge map. Examples are shown using forward looking infrared (FLIR) images. The technique has been extended to cover any digital edge operator and an example is shown for the Spacek operator.

Object Surface Characterization From Range Images

Abstract: Recently we have proposed a range image segmentation method based on the equidistance contour map extracted from a range image. An equidistance contour of a range image is formed by pixels on the range image whose corresponding scene points are all at a same specified distance from the sensor. We have observed that in different ways the contours reflect the existence of object surface edges, the geometries of object surfaces, and the orientations of these surfaces in the 3-D space. In this paper we present a method which uses the equidistance contours to further characterize the regions obtained from segmentation. This is an integral part of an object recognition process. The meaning of characterization is to find the geometries of the corresponding 3-D object surfaces of the regions from the contours. If a surface has nice analytic form such as planar, spherical, cylindrical, or conical, we determine not only its type but also the values of the parameters which describe its geometry.

3-d object in range i mage

Abstract: This paper presents a new method for the segmentation and description of 3-D object in range image. It segments the image in two parallel ways:region classification based on surface normal analysis and edge detection by relaxation process. The two results are comblned together by a small rule-based system to obtain a more reliable segmentation. Features of 3-D object at dlfferent levels are extracted and a surface attributes graph(SAG) is constructed.This kind of view independent description of 3-D object is effective and robust for object recognition and location.The objects considered are industrial parts composed of planes and quadric surfaces. The range image is provided by a laser scanner.