Showing papers in "Pattern Recognition in 1987"

TL;DR: It is shown how the boundaries of an arbitrary non-analytic shape can be used to construct a mapping between image space and Hough transform space, which makes the generalized Houghtransform a kind of universal transform which can beused to find arbitrarily complex shapes.

TL;DR: The modified Hubert index, proposed here for the first time, is shown to perform better than the Davies-Bouldin index under all experimental conditions and demonstrates the difficulty inherent in estimating the number of clusters.

TL;DR: A model for determining the directional relationship in 2-D space between two simply-connected polygons of arbitrary shape, size and distance from each other is developed and found to have a computational complexity of O ( n ), where n is the total number of vertices or cells used to represent the two polygons.

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.

TL;DR: The paper concludes that without cheap highly parallel processing power, the common iterative methods are of no use in a cost effective OCR system.

TL;DR: It is shown that moment functions derived from contour sequences are invariant to shape translation, rotation and scale transformations.

TL;DR: A method to estimate the number of clusters in a data set E, using the bootstrap technique, which involves the generation of several “fake” data sets by sampling patterns with replacement in E (bootstrapping).

TL;DR: This paper presents fast invariant moment computations using the “Delta Method”, as a means of scene representation, which computes moments of contiguous images by using the contribution of each line rather than individual pixel.

TL;DR: A corner detection algorithm which performs close to humans expectations is described, based on previous work done by Freeman and Davis, which uses an approach of averaging multiple s values to avoid the selection of spurious corners.

TL;DR: New forms of shape invariants are derived, using matrix trace operators, for pattern recognition, which form a non-redundant set, have more straightforward and understandable geometric meanings, and require fewer computations.

TL;DR: It is shown that the well known two-pass sequential local transformation algorithm for computing a distance transformation in rectangular domains may fail in some convex integer domains, but that a four-pass algorithm is sufficient in all two-dimensional convex domains.

TL;DR: The basic steps in a general image analysis process are outlined, the state of the art with respect to each step is summarized, limitations of present methods are pointed out and potential directions for future work are indicated.

TL;DR: A new simple and computationally efficient approach to image segmentation via recursive region splitting and merging is presented, based on a generalization of a two-class gradient relaxation method.

TL;DR: A technique based on the Clough-Tocher subtriangulation is described which determines piecewise cubic mapping functions for image registration which is a piecewise combination of local mapping functions.

TL;DR: 3-D multiview object representations are presented as an alternative approach to traditional 3-D volumetric object representations that store features in a viewer-centered representation and thus can be immediately used to match features derived from 2-D images.

TL;DR: Experimental results show that the technique can eliminate a lot of stationary regions and thus can reduce the amount of processing required in the interpretation process.

TL;DR: A new technique using the straight-line Hough transform is advanced for locating curved objects or target tracks in a 2-D image by thresholding the Hough space and performing transformations and an inverse Houghtransform.

TL;DR: This paper presents a design of a post-processor which corrects the Devanagari symbol string based on this observation and its accumulated penalty value for a word gives a measure of its confidence level.

TL;DR: An overview of the current methodologies on image analysis and pattern recognition is presented, with a distinction made between those two streams and later each subject is treated from a methodological point of view.

TL;DR: Results indicate that this approach is promising in computing optical flow with the multiconstraint approach, which relies on the gradient equation, and the use of various image functions.

TL;DR: This paper applies clustering methods to a new problem domain and presents a new method based on a cluster-structure approach for the recognition of 2-D partially occluded objects, which allows the identification of all parts of the model which match in the Occluded scene.

TL;DR: Techniques developed for the second phase of the analysis of human body motion analysis are discussed, using the notion of antiparallel lines to abstract the regions into a higher level primitive which enables one to define and develop operations such as concatenation and deletion.

TL;DR: In this paper, a simple, but effective, algorithm to analyse the rotational symmetry of a given closed-curve shape S is proposed, and several problems caused by the use of discrete image coordinates are analysed and solved.

TL;DR: A modification of this procedure is suggested whereby the estimation of the parameters is performed using statistics which allow for efficient use of cellular or neighborhood image processors rather than those computed using the usual arithmetic operations.

TL;DR: A VLSI architecture based on the space-time domain expansion approach which can compute the string distance and also give the matching index-pairs which correspond to the edit sequence is proposed and can obtain high throughput by using extensive pipelining and parallelism.

TL;DR: Utilization of the final point condition makes the algorithm faster and the flexible deletion of end points produces a line pattern of good quality even for noisy figures.

TL;DR: The proposed algorithm has been applied to satellite images with their contours sectioned by clouds and leads to the guarantee of very good registration, in spite of image contours badly hampered by noise and/or clouds.

TL;DR: This proposed method blends some of the results developed in image enhancement and edge detection literature to minimize the effect of noise.

TL;DR: This paper describes an alternative approach, which segments the envelope into regions of consistent print style and chooses a region most likely to be the address on grounds of position, size, print style, etc. without attempting to read the characters.

TL;DR: An adaptive algorithm obtained by incorporating shape adaptivity into this sequential process is given, and the distortions in the skeleton at the right-angle and acute-angle corners are minimized in the adaptive algorithm.