Showing papers by "Azriel Rosenfeld published in 1987"
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.
109 citations
TL;DR: Physically realizable, optimal generating kernels are presented and show a better command of details than the ones generated by a simple 4 × 4 averaging, or a computationally equivalent kernel.
Abstract: Construction of image pyramids is described as a two-di-mensional decimation process Frequently employed generating kernels are compared to the optimal kernel that assures minimal information loss after the resolution reduction, ie, the one corresponding to an ideal low pass filter Physically realizable, optimal generating kernels are presented The amount of computation required for generation of the image pyramid can be reduced significantly by employing half-band filters as components of the optimal kernel Image pyramids generated by the optimal kernel show a better command of details than the ones generated by a simple 4 × 4 averaging, or a computationally equivalent kernel
76 citations
01 Jul 1987-Graphical Models \/graphical Models and Image Processing \/computer Vision, Graphics, and Image Processing
TL;DR: Methods of detecting and extracting (“delineating,” i.e., segmenting) compact objects from an image are described, designed for implementation on an exponentially tapering “pyramid” of processors, and require only O (log n ) time for an n by n image.
Abstract: This paper describes methods of detecting and extracting (“delineating,” i.e., segmenting) compact objects from an image. The methods are designed for implementation on an exponentially tapering “pyramid” of processors, and require only O (log n ) time for an n by n image. Objects are detected using an “interest measure” derived from local comparisons between fathers and sons in the pyramid. They are extracted by a top-down tree growing process in which the leaves of the tree are the pixels belonging to the detected object.
51 citations
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.
46 citations
TL;DR: Computational techniques that could serve as a basis for this type of object recognition if implemented on appropriate parallel hardware are suggested.
Abstract: Humans can recognize familiar, but unexpected objects, belonging to highly variable object classes, in a fraction of a second — a few hundred cycles of the neural visual "hardware". This position paper suggests computational techniques that could serve as a basis for this type of object recognition if implemented on appropriate parallel hardware.
45 citations
TL;DR: This note presents a modified method of curve partitioning based on selecting points at which the distance of the curve from its chord(s) is a local maximum that seems to yield generally reasonable results.
29 citations
TL;DR: The status of an ongoing research effort to develop a geographic information system based on a variant of the linear quadtree, which uses quadtree encodings for storing area, point and line features, is presented.
26 citations
TL;DR: A theorem is presented which shows that statistics of cross ratios of groups of four collinear points are a projective invariant and the distance between normalized vectors is used as the similarity measure for texture discrimination.
23 citations
01 May 1987-Graphical Models \/graphical Models and Image Processing \/computer Vision, Graphics, and Image Processing
TL;DR: A bibliography of about 1450 references related to the computer processing of pictorial information, arranged by subject matter is presented, to provide a convenient compendium of references.
Abstract: This paper presents a bibliography of about 1450 references related to the computer processing of pictorial information, arranged by subject matter. Coverage is restricted, for the most part, to a selected set of U.S. journals and proceedings of specialized meetings. The topics covered include digitization, approximation, and compression; transforms, filtering, enhancement, restoration, and reconstruction; architectures, systems, software, and techniques; pictorial pattern recognition; feature detection, segmentation, and image analysis; matching and time-varying imagery; shape and pattern; geometry; texture; and three-dimensional scene analysis. No attempt is made to evaluate or summarize the items cited; the purpose is simply to provide a convenient compendium of references.
17 citations
TL;DR: Simple pyramid algorithms are described that, in O(k) steps, shrink or expand the 1's in a binary image by 2k − 1 (k = 1,2, …) and resample the results at intervals of 2k.
8 citations
TL;DR: If the edge pixels are characterized by gradient magnitudes and directions, the fitting process can be refined to give weight to the directions of the edge pixel as well as to their magnitudes.
TL;DR: It is proved that for any finite connected digital set S and integer-valued metric defined on S, there exists a nonnegative integer d such that, for any positive integer n and any points P1,…,Pn of S,there exists a point P in S whose average distance from the Pi's is exactly d.
01 Jan 1987
TL;DR: A probabilistic analysis of the frequency of matches and their expected robustness for specific masks and classes of masks in white noise images is given to help indicate whether or not a given image region should be considered interesting, as regards frequency of occurrence of line-like masks.
Abstract: : Mask matching is a well known procedure (1) in which the detection of specific features in an image is carried out by matching a set of templates or masks with the neighborhood of each pixel in the image. This paper describes a set of 3 X 3 binary masks that can be used to extract thin features from an image. The masks are used to assign various labels to each pixel (each label corresponding t a particular mask), and to associate with each label a set of confidence measures based on the homogeneity of the foreground and background in the mask, and the difference between them. The idea of this approach is to record a large set of useful information at the pixel level, in order to efficiently make use of it at the later stages of the linear feature detection process. Given is a probabilistic analysis of the frequency of matches and their expected robustness for specific masks and classes of masks in white noise images. These results may help indicate whether or not a given image region should be considered interesting, as regards frequency of occurrence of line-like masks, for example.
30 Apr 1987
TL;DR: This paper considers an important class of operations, involving image segmentation by detection and extraction of global regions and features, and describes algorithms for carrying out these operations on parallel "pyramid" hardware.
Abstract: Many types of image processing operations can be performed sequentially at frame rates, but many of the global operations needed in computer vision systems cannot be performed in real time unless suitable parallel hardware is available. This paper considers an important class of such operations, involving image segmentation by detection and extraction of global regions and features, and describes algorithms for carrying out these operations on parallel "pyramid" hardware.