Results 1  10
of
26
SUSAN  A New Approach to Low Level Image Processing
 International Journal of Computer Vision
, 1995
"... This paper describes a new approach to low level image processing; in particular, edge and corner detection and structure preserving noise reduction. ..."
Abstract

Cited by 224 (3 self)
 Add to MetaCart
This paper describes a new approach to low level image processing; in particular, edge and corner detection and structure preserving noise reduction.
An edge detection technique using genetic algorithmbased optimization
 PATTERN RECOGNIT
, 1994
"... In this paper we present a genetic algorithmbased optimization technique for edge detection. The problem of edge detection is formulated as one of choosing a minimum cost edge configuration. The edge configurations are viewed as twodimensional chromosomes with fitness values inversely proportional ..."
Abstract

Cited by 31 (1 self)
 Add to MetaCart
In this paper we present a genetic algorithmbased optimization technique for edge detection. The problem of edge detection is formulated as one of choosing a minimum cost edge configuration. The edge configurations are viewed as twodimensional chromosomes with fitness values inversely proportional to their costs. The design of the crossover and the mutation operators in the context of the twodimensional chromosomal representation is described. The knowledgeaugmented mutation operator which exploits knowledge of the local edge structure is shown to result in rapid convergence. The incorporation of metalevel operators and strategies such as the elitism strategy, the engineered conditioning operator and adaptation of mutation and crossover rates in the context of edge detection are discussed and are shown to improve the convergence rate. The genetic algorithm with various combinations of metalevel operators is tested on synthetic and natural images. The performance of the genetic algorithmbased cost minimization technique is compared both qualitatively and quantitatively with local searchbased and simulated annealingbased cost minimization approaches. The genetic algorithmbased technique is shown to perform very well in terms of robustness to noise, rate of convergence and quality of the final edge image.
Deformable Contours: Modeling, Extraction, Detection And Classification
, 1994
"... This thesis presents an integrated approach in modeling, extracting, detecting and classifying deformable contours directly from noisy images. We begin by conducting a case study on regularization, formulation and initialization of the active contour models (snakes). Using minimax principle, we deri ..."
Abstract

Cited by 17 (0 self)
 Add to MetaCart
This thesis presents an integrated approach in modeling, extracting, detecting and classifying deformable contours directly from noisy images. We begin by conducting a case study on regularization, formulation and initialization of the active contour models (snakes). Using minimax principle, we derive a regularization criterion whereby the values can be automatically and implicitly determined along the contour. Furthermore, we formulate a set of energy functionals which yield snakes that contain Hough transform as a special case. Subsequently, we consider the problem of modeling and extracting arbitrary deformable contours from noisy images. We combine a stable, invariant and unique contour model with Markov random field to yield prior distribution that exerts influence over an arbitrary global model while allowing for deformation. Under the Bayesian framework, contour extraction turns into posterior estimation, which is in turn equivalent to energy minimization in a generalized active...
Computer Vision Algorithms on Reconfigurable Logic Arrays
 IEEE TRANS. ON PARALLEL AND DISTRIBUTED SYSTEMS
, 1999
"... Computer vision algorithms are natural candidates for high performance computing due to their inherent parallelism and intense computational demands. For example, a simple 3 x 3 convolution on a 512 x 512 gray scale image at 30 frames per second requires 67.5 million multiplications and 60 million a ..."
Abstract

Cited by 16 (1 self)
 Add to MetaCart
Computer vision algorithms are natural candidates for high performance computing due to their inherent parallelism and intense computational demands. For example, a simple 3 x 3 convolution on a 512 x 512 gray scale image at 30 frames per second requires 67.5 million multiplications and 60 million additions to be performed in one second. Computer vision tasks can be classified into three categories based on their computational complexity andcommunication complexity: lowlevel, intermediatelevel and highlevel. Specialpurpose hardware provides better performance compared to a generalpurpose hardware for all the three levels of vision tasks. With recent advances in very large scale integration (VLSI) technology, an application specific integrated circuit (ASIC) can provide the best performance in terms of total execution time. However, long design cycle time, high development cost and inflexibility of a dedicated hardware deter design of ASICs. In contrast, field programmable gate arrays (FPGAs) support lower design verification time and easier design adaptability atalower cost. Hence, FPGAs with an array of reconfigurable logic blocks canbevery useful compute elements. FPGAbased custom computing machines are
On GDM's: Geometrically Deformed Models for the Extraction of Closed Shapes from Volume Data
 Masters thesis, Rensselaer Polytechnic Institute
, 1990
"... The advent of nondestructive sensing equipment (CT, MRI) created an entirely new field of research for image engineers. The equipment generates a point sampling of a true threedimensional object. Typically, this point sampling is presented as a series of slices through the 3D object. It has become ..."
Abstract

Cited by 14 (0 self)
 Add to MetaCart
The advent of nondestructive sensing equipment (CT, MRI) created an entirely new field of research for image engineers. The equipment generates a point sampling of a true threedimensional object. Typically, this point sampling is presented as a series of slices through the 3D object. It has become apparent, however, that displaying individual slices does not lend itself to conveying the true threedimensional structure of the scanned object. Research, therefore, has focused on alternative methods of presenting volume data. This thesis proposes an approach that will generate a topologically closed simple geometric model of an object within a scalar field. A Geometrically Deformed Model, GDM, is created by placing an initial simple model in the data set which is then deformed by minimizing a set of constraints. The constraint functions evaluated at each vertex in the model control the local deformation, the interaction between the model and the data set, and maintain the shape and topol...
An Edge Detection Technique Using Local Smoothing and Statistical Hypothesis Testing
, 1996
"... An edge detection technique based on local smoothing and statistical hypothesis testing for the detection and localization of step edges and roof edges is proposed. Smoothing and statistical hypothesis testing procedures for detection and localization of step edges and roof edges are formulated. Exp ..."
Abstract

Cited by 12 (5 self)
 Add to MetaCart
An edge detection technique based on local smoothing and statistical hypothesis testing for the detection and localization of step edges and roof edges is proposed. Smoothing and statistical hypothesis testing procedures for detection and localization of step edges and roof edges are formulated. Experimental results on grayscale images are presented. The merits, limitations and factors critical to the performance of the proposed technique are discussed. Possible improvements and future research directions are outlined.
Modeling Image Analysis Problems Using Markov Random Fields
, 2000
"... this article are addressed mainly from the computational viewpoint. The primary concerns are how to dene an objective function for the optimal solution for an image analysis problem and how to nd the optimal solution. The reason for dening the solution in an optimization sense is due to various unce ..."
Abstract

Cited by 9 (1 self)
 Add to MetaCart
this article are addressed mainly from the computational viewpoint. The primary concerns are how to dene an objective function for the optimal solution for an image analysis problem and how to nd the optimal solution. The reason for dening the solution in an optimization sense is due to various uncertainties in imaging processes. It may be dicult to nd the perfect solution, so we usually look for an optimal one in the sense that an objective, into which constraints are encoded, is optimized
Simulated Annealing and Genetic Algorithms for Shape Detection
, 1996
"... this paper we consider the problem of recognizing simple geometric shapes in a picture corrupted by noise. The algorithmic techniques we use for its solution are simulated annealing, genetic algorithms and a constructive method based on noise filtering. Simulated annealing is a powerful stochastic t ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
this paper we consider the problem of recognizing simple geometric shapes in a picture corrupted by noise. The algorithmic techniques we use for its solution are simulated annealing, genetic algorithms and a constructive method based on noise filtering. Simulated annealing is a powerful stochastic technique for solving combinatorial optimization problems. One of the main drawbacks of simulated annealing is its high computational requirements. Because of this, a number of parallel implementations have been proposed [1, 5, 8, 10, 17, 23, 30]. In particular, in [10] some problem independent parallel implementations of simulated annealing have been described. Simulated annealing has been proposed to solve image recognition problems [6, 7, 28]. In particular, in [6] a parallel implementation of simulated annealing for the shape detection problem has been proposed. In this paper we present the results obtained using the farming implementation of simulated annealing as it was proposed in [10] for other applications. In Section 2 of this paper, the shape detection problem is formally defined and its representation in terms of a combinatorial optimization problem is described. In Section 3 the general simulated annealing algorithm is described together with some of the parallel implementations proposed for it. In Section 4 we describe a genetic algorithm for the shape detection problem. This algorithm is inherently parallel. In Section 5 we present a constructive heuristic for the shape detection problem which is based on a noise filter. Performance measurements presented in Section 6 for the different algorithms finish the paper. 2 The Shape Detection Problem
A Markovian Model For Contour Grouping.
 In Proc. International Conference on Pattern Recognition, volume A
, 1994
"... : In order to interpret and analyse a scene, determining the contours is a fundamental step. Classical methods of contour extraction do not always allow the detection of all the contours. We notice, for example, that the contours obtained by a CannyDeriche filter have some gaps, especially at corner ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
: In order to interpret and analyse a scene, determining the contours is a fundamental step. Classical methods of contour extraction do not always allow the detection of all the contours. We notice, for example, that the contours obtained by a CannyDeriche filter have some gaps, especially at corners or at Tjunctions. In short, the boundaries which are detected are not always closed. In this report, we present an algorithm that restores incomplete contours. We model the image by Markov Random Fields and we define the Gibbs Distribution associated with it. In order to complete the contours, several criteria are defined and introduced in an energy function, which has to be optimized. The deterministic ICM ("Iterated Conditional Mode") relaxation algorithm is implemented to minimize this energy function. The result is a contour image consisting of closed contours. This method has been tested on different images which present different types of difficulties (indoors, outdoors, satellite (...
Modelbased polyhedral object recognition using edgetriple features
, 1997
"... While significant progress has been made in the computer vision field over the past decade, and machines capable of performing specialised visual inspection tasks are now being used in many industrial applications, the problem of recognising threedimensional objects from twodimensional imagery rem ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
While significant progress has been made in the computer vision field over the past decade, and machines capable of performing specialised visual inspection tasks are now being used in many industrial applications, the problem of recognising threedimensional objects from twodimensional imagery remains an area of ongoing research. Vision is undoubtedly our most important sense, and solutions to the problem of general threedimensional machine vision must be found if the long term goal of autonomous robotic agents interacting naturally with humans in the real world is to be realised. In this work the problem of recognising polyhedra from twodimensional images is investigated. The use of perceptual grouping and intermediatelevel geometric features is considered, in particular the &quot;edgetriple &quot; feature. The edgetriple feature consists of three connected straight edges of an object, projecting to a triple of connected lines in the image, and can be used as a key feature, or indexing primitive, in modelbased object recognition. The geometric constraints provided by matching such a configuration of image lines to an edgetriple are sufficient to uniquely determine the pose of the object. A