Showing papers in "Pattern Recognition in 1973"

TL;DR: In this experiment, the difference between the desired position and the actual position of the object is recognized visually and corrected by moving the manipulator.

TL;DR: A cursive script recognition program which has correctly identified 79 per cent of a test sample of 84 words, which compares favorably in performance level with previously reported programs appropriately “normalized”, while not requiring input pertaining to stroke sequence and stroke segmentation that is essential to these other programs.

TL;DR: A new K-L technique is described that overcomes some of the limitations of the earlier procedures and is suggested that it is particularly useful for pattern recognition when combined with classification procedures based upon discriminant functions obtained by recursive least squares analysis.

TL;DR: The various classes of functions which can be implemented in the cellular array are discussed and sample programs explained in detail.

TL;DR: Flexible templates (rubber masks) have been devised which, when fitted to natural data, can be used for measurement, data reduction, data smoothing, and classification of highly irregular waveforms and image shapes.

TL;DR: This paper briefly summarizes much of the work in pattern recognition to date, and relates the rubber mask technique to previous work, and a scheme for incorporating flexible-mask methods into a proposed pattern recognition and memory system is presented.

TL;DR: The new design philosophy of a three-stage structure is believed to offer at least a suboptimal search strategy for recognizing printed Chinese characters with a dictionary of 7000–8000 characters.

TL;DR: Fuzzy set theory is summarized, and finite fuzzy automata and languages are described, and some results in fuzzy languages based on the “max(min)” rule are discussed.

TL;DR: An algorithm which generates a digital image representing an estimate of the difference between two individually distorted digital images is described, which approximates the nonlinear transformation that is necessary to obtain spatial registration of the two images under comparison.

TL;DR: This paper presents a formalization of the concept of cluster analysis, and begins with an intuitive description of clustering, and discusses the separation of the measurement problem from the clustering problem.

TL;DR: A method for quantitatively measuring properties of soil cracking patterns is described, using a computer program to measure properties of the cracking patterns on scanned images of the photographs.

TL;DR: The purpose of the glossary is to state in the simplest possible way the general meaning or word usage for many of the terms in image pattern recognition.

TL;DR: This paper proposes a digital optical computer technology using holographic programming and a simple laboratory demonstration is given where the Identity Function and the Exclusive-OR are generated using this technology.

TL;DR: A descriptive, generative model of the pen direction sequences measured during the hand drawing of characters is shown to be useful in recognition.

TL;DR: Some recent work on filters synthesis will be summarized, and the implications in optical computation, image processing (enhancement), image restoration, image restoration and image detection examined.

TL;DR: Some specific aspects of language structure are examined in detail in the context of pattern analysis, and the characteristics of the transformational grammars of linguists show up even in this simple example.

TL;DR: The learning algorithm for obtaining the mean pattern vector and the first principal component is described, and also some experimental results on random patterns are presented.

TL;DR: A new device is described that transfers high resolution, two-state images from a noncoherent beam of light to a fully coherentbeam of light in quasi real-time (less than 1 sec) and discusses its application to coherent optical data processing.

TL;DR: It is shown that for a Bayes discriminant analysis certain alternative densities lead to piecewise linear discriminant functions even when the covariance matrices are unequal and provide a theoretical solution for evaluating probabilities of correct classification and of misclassification.

TL;DR: Several algorithms are described that transform the figure into another one of the least possible dimensions, but retaining a set of basic topological and quasi-topological characteristics of the original picture, and are particularly suited for hardware implementation.

TL;DR: The results of computer experiments with artificially generated data and with handprinted alphanumeric characters are given to show that the approach proposed is quite useful for recognition of Markovian patterns.

TL;DR: The results show that, provided an appropriate input device can be constructed, it is possible to build a reading machine capable of reading one printed character at a time with an accuracy sufficient for most reading purposes.

TL;DR: The results suggest that predictive ability may not change significantly over wide ranges of D providing the N/D ratio is constant and greater than two or three.

TL;DR: Parallel, analog image processing techniques are described which enable images to be decomposed into subimages that contain objects that are spatially and spectrally equivalent and hold promise for on-board, computer-controlled processing of remotely obtained images.

TL;DR: A new definition of stereoscopic pattern signal is presented, and is used to devise a special series of random texture stereograms given the class of statistically homogenous textures, stereoscopic patterns will emerge when the elemental disparities perceptually cluster over the field.

TL;DR: A compact parallel image processing system concept has been developed that uses off-axis paraboloidal mirror segments as collimating, Fourier transforming and image reconstructing elements, and the use of a GaAs laser diode as the coherent radiation source.

TL;DR: Cross-correlating an image with a “transfer function image” and thresholding the resultant cross-correlation image will perform the equivalent operations in a single step.

TL;DR: A data compression scheme is developed on the basis that prior to exact repetition, approximate repetitions take place during the iterations, and the resulting picture is almost the same as the original.