Showing papers on "Segmentation-based object categorization published in 1987"
TL;DR: To compute the flow predicted by the segmentation, a recent method for reconstructing the motion and orientation of planar surface facets is used and the search for the globally optimal segmentation is performed using simulated annealing.
Abstract: This paper presents results from computer experiments with an algorithm to perform scene disposition and motion segmentation from visual motion or optic flow. The maximum a posteriori (MAP) criterion is used to formulate what the best segmentation or interpretation of the scene should be, where the scene is assumed to be made up of some fixed number of moving planar surface patches. The Bayesian approach requires, first, specification of prior expectations for the optic flow field, which here is modeled as spatial and temporal Markov random fields; and, secondly, a way of measuring how well the segmentation predicts the measured flow field. The Markov random fields incorporate the physical constraints that objects and their images are probably spatially continuous, and that their images are likely to move quite smoothly across the image plane. To compute the flow predicted by the segmentation, a recent method for reconstructing the motion and orientation of planar surface facets is used. The search for the globally optimal segmentation is performed using simulated annealing.
356 citations
TL;DR: The directional image can be thought of as an image transform, where each pixel of the image represents direction of the local grey level uniformity, and has been found to work better than the segmentation based on grey level statistics and other methods.
191 citations
Book•
01 Jan 1987TL;DR: Presents the first unified theory of image segmentation, written by the winners of the 1985 Pattern Recognition Society medal, which describes a new class of algorithms (based, in part, on quadtrees) and demonstrates their applications, including grey level and texture segmentation.
Abstract: From the Publisher:
Presents the first unified theory of image segmentation, written by the winners of the 1985 Pattern Recognition Society medal Until now, image processing algorithms have always been beset by uncertainties, no one method proving completely satisfactory Wilson and Spann tackle the problem of uncertainty head-on They describe a new class of algorithms (based, in part, on quadtrees) and demonstrate their applications, including grey level and texture segmentation These algorithms produce excellent results in a wide range of synthetic and natural data Provides many examples of applications from medicine to remote sensing
122 citations
TL;DR: A new approach to the segmentation problem is presented by optimizing a criterion which estimates the quality of a segmentation by using a graph-based description of a partition of an image and a merg...
Abstract: We present a new approach to the segmentation problem by optimizing a criterion which estimates the quality of a segmentation. We use a graph-based description of a partition of an image and a merg...
67 citations
01 Oct 1987-Graphical Models \/graphical Models and Image Processing \/computer Vision, Graphics, and Image Processing
TL;DR: A segmentation algorithm based on deterministic relaxation with varying neighborhood structures for the segmentation of noisy images, modeled as a discrete-valued Markov random field, corrupted by additive, independent, Gaussian noise is presented.
Abstract: This paper presents a segmentation algorithm based on deterministic relaxation with varying neighborhood structures for the segmentation of noisy images. The image is modeled as a discrete-valued Markov random field (MRF), or equivalently a Gibbs random field, corrupted by additive, independent, Gaussian noise; although, additivity and Gaussian assumptions are not necessary for the algorithm. The algorithm seeks to determine the maximum a posteriori (MAP) estimate of the noiseless scene. Using varying neighborhoods during relaxation helps pick up certain directional features in the image which are otherwise smoothed out. The parallelism of the algorithm is underscored by providing its mapping to mesh-connected and systolic array processors suitable for VLSI implementation. Segmentation results are given for 2- and 4-level Gibbs distributed and geometric images corrupted by noise of different levels. A comparative study of this segmentation algorithm with other relaxation algorithms and a single-sweep dynamic programming algorithm, all seeking the MAP estimate, is also presented.
44 citations
01 Oct 1987
TL;DR: Two algorithms are described for automatic image segmentation using a homogeneity measure and a contrast measure defined on the co-occurrence matrix of the image using the concept of logarithmic response of the human visual system.
Abstract: Two algorithms are described for automatic image segmentation using a homogeneity measure and a contrast measure defined on the co-occurrence matrix of the image. The measure of contrast involves the concept of logarithmic response (adaptability with background intensity) of the human visual system. Provisions are made in two different ways to remove the undesirable thresholds. The effectiveness of the algorithms is demonstrated for a set of images having different types of histograms. Their performance is compared to that of existing algorithms.
30 citations
01 Mar 1987
TL;DR: A way of extracting the parameters of these types of surfaces using normal analysis is described, which allows efficient and robust use of a matching process for object recognition and pose determination even though there are many undefined or occluded surface regions.
Abstract: This paper addresses the problem of extracting certain types of surfaces for 3-dimensional object recognition in the presence of partial occlusion and noise using range information. We restrict consideration to man made industrial parts. In this case, we need only work with a small set of surface shapes such as planes, cylinders, and spheres since such shapes are common in industrial parts and it is only necessary to recognize and locate a sufficient set of surface regions to uniquely distinguish the part being observed from all others that might be present. We describe a way of extracting the parameters of these types of surfaces using normal analysis. The use of surface parameters then allows efficient and robust use of a matching process for object recognition and pose determination even though there are many undefined or occluded surface regions.
28 citations
Journal Article•
TL;DR: An image segmentation method which may be applied to various tasks such as natural segmentation of monochromatic or color, three dimensional seismic or scanner images, based on the region growing principle is presented.
Abstract: We present an image segmentation method which may be applied to various tasks such as natural segmentation of monochromatic
or color, three dimensional seismic or scanner images . Our algorithme is based on the region growing principle. Its originality
lies on optimising the use of a sequence of criteria . We separate the common strategy of using segmentation criteria from the
task specific definition of those criteria . This separation between algorithm and mathematical aspects of our method provides
for its generality . Experiments results are shown .
20 citations
01 May 1987
TL;DR: A novel view of the segmentation/ description process is presented and an effective algorithm based on the model is described.
Abstract: Edge-based binocular correspondence produces a sparse disparity map, available information being distributed along space curves which project to matched image edges. To become useful, these contours must be parsed into describable sections. A novel view of the segmentation/ description process is presented and an effective algorithm based on the model is described.
20 citations
01 Apr 1987
TL;DR: A new segmentation method based on the properties of the human visual system that is part of a new second generation image coder and one of the properties is that it is not necessary to transmit (or store) the visual residual for use in reconstructing the received signal.
Abstract: A new segmentation method based on the properties of the human visual system is described in this paper. The segmentation method is part of a new second generation image coder. In addition, one of the properties is that it is not necessary to transmit (or store) the visual residual for use in reconstructing the received signal. It is assumed that the characteristics of this visual residual are known at the receiver and can be used in the reconstruction process.
13 citations
14 Oct 1987
TL;DR: An attempt is made to extend a hybrid DPCM/transform coding configuration with object based features by exploiting the knowlegde of typical interpersonal videocommunication scenes such as head and shoulders to create a pyramidal image data approach.
Abstract: The goal of image coding is to reduce the number of bits required to represent an image or a sequence of images. For the new generation of video coders the coding should be tailored to image understanding and should more and more replace the purely statistical or waveform coding techniques. At present the more conventional algorithms are prefered in the light of a direct hardware realization. Most of these methods are block based and do not consider the global image features. In this paper an attempt is made to extend a hybrid DPCM/transform coding configuration with object based features. For videoconferencing and videophone applications we intend to use a simple method exploiting the knowlegde of typical interpersonal videocommunication scenes such as head and shoulders. The first step is the segmentation of the image into participants and background. For this segmentation information available in the coder is combined with knowledge obtained from the actual analysis. To make the segmentation more robust to various input signals we accept the fact that objects often are more easily recognized in images that have a very low spatial sampling rate. At this low level details are not taken into account which yields a more homogenous segmentation. This leads to a pyramidal image data approach. By applying a coarse to fine structure the procedure starts at a low resolution level and is refined at ever-increasing resolutions. A comparison is given of a coder with and without this a priori knowlegde.
13 Oct 1987
TL;DR: An improved segmenter has been developed which partitions a monochrome image into homogeneous regions using local neighborhood operations and a technique is introduced which segments the remainder of the image to reveal details that were previously lost.
Abstract: An improved segmenter has been developed which partitions a monochrome image into homogeneous regions using local neighborhood operations. Perkin's well-known edge-based segmenting algorithm [1] is used to partition those portions of an image with little detail (low edge density) into regions of uniform intensity. A technique is introduced which segments the remainder of the image to reveal details that were previously lost. Region merging is then performed by removing selected boundary pixels that separate sufficiently similar (e.g., in average intensity) regions subject to the constraint that the boundary pixel quality (e.g. edge strength) is below a selected threshold. Region merging is repeated using less and less restrictive merging criteria until the desired degree of segmentation (e.g. number of regions) is obtained.
11 May 1987
TL;DR: Specific methods used for model selection and model guided segmentation are shown along with a method for segmentation evaluation.
Abstract: A technique for segmenting objects in an image guided by a model is shown. Specific methods used for model selection and model guided segmentation are shown along with a method for segmentation evaluation. Examples of segmenting ground targets in infrared imagery are shown.
01 Dec 1987
TL;DR: In this paper, the suitability of algorithms derived from the study of fractal geometry to the specific problem of image segmentation was investigated and the use of two widely used methods and an original hybrid technique for calculating fractal dimension was demonstrated.
Abstract: : The purpose of this study was to investigate the suitability of algorithms derived from the study of fractal geometry to the specific problem of image segmentation The use of two widely used methods and an original hybrid technique for calculating fractal dimension is demonstrated All three techniques appeared to be effective to some extent in segmenting images, but shared a common problem with the establishment of suitable thresholds for robust segmentation
01 Apr 1987
TL;DR: The image classes are shown in this paper as visual aids in judging the classification procedure and are encoded using matrix quantizers with separate codebook for each class of the image.
Abstract: This paper describes a composite source model for images. An image is segmented into uniform and homogeneous regions using centroid linkage region growing algorithms. The region homogenity is determined by the Student T-statistics. Excessive regions resulting from region growing are merged according to region merging rules. The initially segmented image is then clustered into classes with the help of the K-means algorithm. The image classes are shown in this paper as visual aids in judging the classification procedure. The class numbers and the corresponding pixel counts are also included. Finally, as an application of composite source models, an image is encoded using matrix quantizers with separate codebook for each class of the image.
01 Apr 1987
TL;DR: The problem of image segmentation of noisy images is examined and an adaptive Bayesian parameter estimation/image detection algorithm is developed that allows us to estimate the unknown image and its underlying parameters in an optimal manner.
Abstract: In this paper we examine the problem of image segmentation of noisy images. We consider a doubly stochastic image model. The image is assumed to be the sum of the realizations of two independent random fields: the uncorrupted image and the noise field, consisting of independent, identically distributed, Gaussian random variables. The image segmentation technique employed here is a technique in which the image is represented by a semi-Markov random field corrupted by additive white noise. An adaptive Bayesian parameter estimation/image detection algorithm is developed. This algorithm allows us to estimate the unknown image and its underlying parameters in an optimal manner. We demonstrate the potential of the proposed algorithm in the case of the smoothing/segmentation of two 4-gray level real images.
06 Jun 1987
TL;DR: This project examines some parallel architectures designed for image processing, and then addresses their applicability to the problem of image segmentation by texture analysis, and proposes an architecture for textural segmentation based on local differences in texture elements (texels).
Abstract: This project examines some parallel architectures designed for image processing, and then addresses their applicability to the problem of image segmentation by texture analysis. Using this information, and research into the structure of the human visual system, an architecture for textural segmentation is proposed. The underlying premise is that textural segmentation can be achieved by recognizing local differences in texture elements (texels). This approach differs from most of the previous work where the differences in global, second-order statistics of the image points are used as the basis for segmentation. A realistic implementation of this approach requires a parallel computing architecture which consists of a hierarchy of functionally different nodes. First, simple features are extracted from the image. Second, these simple features are linked together to form more complex texels. Finally, local and more global differences in texels or their organization are enhanced and linked into boundaries.
TL;DR: A new learning model for real-time, grey-level image segmentation is presented and gives excellent results for images with different shapes.
Abstract: A new learning model for real-time, grey-level image segmentation is presented The model gives excellent results for images with different shapes
01 Jan 1987
27 Mar 1987
TL;DR: In this paper a procedure for optimum threshold selection is described and results of this technique have potential significance in many areas of computer vision and robotics.
Abstract: One of the main problems in image segmentation is the proper selection of thresholds. In this paper a procedure for optimum threshold selection is described. Threshold levels are chosen in such a way that the expected value of the overall segmentation error is minimized. Segmentation error is defined as some function of the error between the input image and the segmentation output. A set of equations is derived for the parameters of the segmentation and a solution for them is indicated. Experimental results are presented. Results of this technique have potential significance in many areas of computer vision and robotics.© (1987) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
11 May 1987
TL;DR: This work presents a new approach to the segmentation problem by optimizing a criterion which estimates the quality of a segmentation by using a graph-based description of a partition of an image and a merging strategy based on the optimal use of a sequence of criteria.
Abstract: We present a new approach to the segmentation problem by optimizing a criterion which estimates the quality of a segmentation. Our method offers a general framework for solving a large class of segmentation problem. We use a graph-based description of a partition of an image and a merging strategy based on the optimal use of a sequence of criteria. An efficient data structure enables our implementation to have a low algorithmic complexity. We show how we adapt this method to segment 2-d natural images including color images and how we use results for solving the stereo matching problem.
27 Mar 1987
TL;DR: In this article, two approaches that explicitly incorporate knowledge about the class of imagess to be processed and the tasks to be performed, plays an important role and explicitly incorporate such knowledge are advanced for images containing polygonal shapes.
Abstract: Image segmentation is a highly scene dependent and problem dependent decision making or pattern recognition process Knowledge about the class of imagess to be processed and the tasks to be performed, plays an important role Two approaches that explicitly incorporate such knowledge are advanced for the class of images containing polygonal shapes They can be generalized to other shapes by change of pre-processing steps Inference is both data driven and goal driven It is guided by meta rules that are fired by the outputs of preprocessing Effective suppression of noise is achieved The methods illustrate the potential of AI techniques and tools for low-level image understanding tasks
01 Jan 1987
TL;DR: The current implementation of a system for automatic segmentation of multi-tempora l remotely-sensed images which exploits prior knowledge to isolate the regions of interest is described.
Abstract: In order to cope with the large volume of remotely-sensed data available now and expected in the future, efficient automatic processing techniques are required. A particular problem in automatic interpretation of this data is the identification of relevant connected regions in the image, i.e. segmentation. This can generally only be achieved to a required degree of accuracy if performed manually. This paper describes the current implementation of a system for automatic segmentation of multi-tempora l remotely-sensed images which exploits prior knowledge to isolate the regions of interest. The system is directed principally towards the applications of crop and environmental monitoring.