Showing papers in "Image and Vision Computing in 1993"

TL;DR: Focus enhancement algorithms are described, closely related to the autofocus system, which may be employed to obtain an entirely sharp image in tissue microscopy.

TL;DR: This scheme provides a route to 3D object recognition through 2D shape description and reduces the problem of perceptual invariance to a series of independent analyses with an associative link established between the outputs.

TL;DR: The skeletonization algorithm includes a beautifying step and a pruning step, which favour the use of the skeleton for shape analysis tasks, and is driven by the Euclidean distance map of the pattern.

TL;DR: This paper proposes a new solution based on variational techniques for optical flow estimation and regularization, which takes into account the discontinuities, and strongly reduces the related problems.

TL;DR: Algorithms for generating shape and colour descriptors from fish images are described, which are unaffected by fish bending and deformations and used to sort 18 species of demersal and five species of pelagic fish.

TL;DR: Two modules are presented: the cavity detector, a method for the segmentation of regions which are not completely surrounded by walls and edgmentation, a modified split-and-merge algorithm for edge preserving image enhancement, segmentation and data reduction.

TL;DR: The performance of an image database system using a specific methodology for the efficient representation, indexing and retrieval of images based on spatial relationships and properties of objects is evaluated.

TL;DR: Extensive experimentation on real images shows that vanishing points can be identified and located even in cluttered images, and that the proposed algorithm is suitable for recovering the heading direction of a robot moving in corridors and offices.

TL;DR: The connective HT with focus of attention reduces the computational complexity of the DCHT and increases the S/N ratio of the peak in its accumulator and may be regarded as a principled method for curve tracing.

TL;DR: The resulting structure is a form of a partial persistent quadtree and can be used to represent sequences of similar binary raster images emerging in various applications such as computer graphics, image processing, geographic information systems, or even satellite pictures.

TL;DR: A technique was developed for region merging which combines both edge and region boundary information and a rule based bone extraction stage then follows which labels the extracted regions as bone or background.

TL;DR: A modified linear form of the Hopfield neural network with a summation layer to significantly decrease the number of interconnections is proposed for the reconstruction of cross-sectional images from projections.

TL;DR: The development of a method for describing the 3D surface shape of individual faces explicitly is described, and some initial results suggest that this method does capture some psychologically relevant dimensions of facial variation.

TL;DR: In this article, the offset of image centre has little effect on the determinations of the position and orientation of a coordinate frame; however, the scale factor has a great influence on the position of the coordinate frame and the accuracy of measurement.

TL;DR: A range of locally adaptive contrast enhancement methods are described, including a number of recently developed structure-sensitie techniques, countering the difficulty that such methods are typically evaluated only subjectively.

TL;DR: It is shown that buses can be successfully used both as topological descriptors and as powerful computational devices in a multiprocessor reconfigurable architecture, allowing fast algorithms to solve a number of fundamental image processing and computer vision tasks.

TL;DR: A depth estimation algorithm based on the assumption that a planar surface should have a constant first derivative on the propagating phase, which indicates that the estimated depth error is kept within 2% when k is greater than or equal to 6.

TL;DR: A method for segmenting retinal images using positions of blood vessels supplying the retina to define a process of merging primary regions into large patches using image properties such as texture and intensity, and semantic interpretations of boundaries and their measured properties.

TL;DR: An automatic, unsupervised procedure for the recognition of a small set of gestures drawn with the aid of a mouse in the Windows environment of a modern computer workstation and some experimental results are presented which indicate the potential for accurate gesture recognition.

TL;DR: A new catalogue of junction labellings is given when smooth edges (discontinuities of surface curvature) are permitted on object surfaces, extending the work of Malik on curved objects with piecewise C3surfaces.

TL;DR: This study shows the different behaviour between 3D operators and their 2D correspondents, and provides useful guidelines in choosing and using appropriate operators in 3D edge detection aplications.

TL;DR: These techniques make it possible to reconstruct the shape of an object in its intrinsic coordinates without having to match new observations to a partially reconstructed description.

TL;DR: A new two-view motion algorithm is presented and then extended to long sequence motion analysis, which automatically finds the proper model that applies to an image sequence and gives the globally optimal solution for the motion and structure parameters under the chosen model.

TL;DR: The representation of shape in machine vision is reviewed with emphasis on the most common types of representation and recent developments, and particular emphasis is placed on the importance of invariant descriptions and on the representation ofshape classes.

TL;DR: This paper traces the evolution of thought and techniques in shape recognition using volumetric primitives, and identifies a number of issues as a basis for assessing these achievements, and proposes some methodological considerations for choosing a way forward.

TL;DR: An automatic thresholding method based on aspects of the human visual system which preserved edge structure in images is proposed, and first determine edge thresholds based on human visual perception, and then use these edge thresholds to find several edge intervals.

TL;DR: A new algorithm is proposed for skeletonizing binary digital images that identifies regions of particular shapes in the image and substitutes apppropriate skeleton patterns for them, generating the final skeleton.

TL;DR: A data structure that models 3D parametric objects is described, and experimentation suggests that by treating the model and camera parameters uniformly, correct pose determination can be obtained from minimal matching information relative to the number of matches required per component individually.

TL;DR: The convergences between James Gibson's ecological approach to vision and recent research in computer vision are explored with reference to the existing research on picture perception and production.

TL;DR: A parallel model which is designed for use as a basis for implementation of edge tracking algorithms on parallel architectures using a tracking algorithm founded on features such as the mid-point, orientation and the length of edge segments is outlined.