A new approach to texture analysis is presented based on the spatial distribution of local features in unsegmented textures. The textures are described using features derived from and clusters detected in generalized ...
详细信息
A new approach to texture analysis is presented based on the spatial distribution of local features in unsegmented textures. The textures are described using features derived from and clusters detected in generalized cooccurrence matrices (GCM). A GCM is determined by a spatial constraint predicate and a set of local features. Features derived from GCM's are discussed and several examples are presented.
Pulse-echo waveform patternrecognition provides one useful approach to characterization of the tissue from its ultrasonic signature. The effectiveness of this method was demonstrated by the analysis of the ultrasonic...
详细信息
Pulse-echo waveform patternrecognition provides one useful approach to characterization of the tissue from its ultrasonic signature. The effectiveness of this method was demonstrated by the analysis of the ultrasonic backscatter from the normal and pyelonephritic rabbit kidney in vivo. Average classification rate of 98% was achieved on a per animal basis using the best 4 of 65 measures and the sample size of 256 waveforms per animal. Using 1728 waveforms from a group of animals, 82% success rate is obtained.
A syntactic model for texture analysis has been previously described. A texture pattern is divided into fixed-size windows. Windowed patterns are described by using tree representations and characterized by tree gramm...
详细信息
A syntactic model for texture analysis has been previously described. A texture pattern is divided into fixed-size windows. Windowed patterns are described by using tree representations and characterized by tree grammars. This paper introduces a texture grammar interference procedure whcih employs a clustering algorithm and a stochastic regular grammar interference procedure.
Fast computation algorithms are developed for two-dimensional convolution. It is shown that the ″overlap-and-save″ method enjoys a clear advantage over the ″overlap-and-add″ method with respect to speed and storag...
详细信息
Fast computation algorithms are developed for two-dimensional convolution. It is shown that the ″overlap-and-save″ method enjoys a clear advantage over the ″overlap-and-add″ method with respect to speed and storage in two-dimensional application. computational burden is explicitly specified when such methods are used in conjunction with radix-2 Fast Fourier Transform algorithms.
A new measure of scene content based on the concept of structural entropy is presented. The measure utilizes unary and binary relationships extracted from a relational representation of a scene. Bounds on the entropy ...
详细信息
A new measure of scene content based on the concept of structural entropy is presented. The measure utilizes unary and binary relationships extracted from a relational representation of a scene. Bounds on the entropy measure are dependent only on the structure of the data base. Experimental results illustrating the concepts developed in the paper are presented.
A synthesis procedure is presented for a class of two-dimensional finite impulse response filters that are easily implemented by sequential convolution of small convolution operators, 3 multiplied by 3 for example. Th...
详细信息
A synthesis procedure is presented for a class of two-dimensional finite impulse response filters that are easily implemented by sequential convolution of small convolution operators, 3 multiplied by 3 for example. The design method is based upon minimization of the mean square error between the synthesized filter and a design prototype.
The classification of finite duration sample functions from zero-mean vector stationary Gaussian distributed time series is considered. A minimum probability of error classification procedure is achieved by a minimum ...
详细信息
The classification of finite duration sample functions from zero-mean vector stationary Gaussian distributed time series is considered. A minimum probability of error classification procedure is achieved by a minimum Kullback Leibler number rule. Time and frequency domain formulas for the Kullback Leibler number between stationary Gaussian time series are developed. Practical computational formulas for the estimation of KL numbers from sample time series are also developed.
A fast algorithm is presented for doing two-dimensional median filtering. It is based on storing and updating the gray level histogram of the picture elements in the window. The algorithm is much faster than conventio...
详细信息
A fast algorithm is presented for doing two-dimensional median filtering. It is based on storing and updating the gray level histogram of the picture elements in the window. The algorithm is much faster than conventional sorting methods. For a window size of m multiplied by n, the computer time required is O(n).
An adaptive object ″growing″ algorithm is presented for detection of individual small rounded opacities in coal workers' chest x-rays at an early stage of pneumoconiosis. An object of suspected lesion is grown f...
详细信息
An adaptive object ″growing″ algorithm is presented for detection of individual small rounded opacities in coal workers' chest x-rays at an early stage of pneumoconiosis. An object of suspected lesion is grown from a ″seed″ pixel so that a figure of merit called the isolation contrast integral is maximized, and is then classified according to several parameters. This is an improvement over the algorithm which was presented previously.
A feasibility method of a real time image bandwidth compression system has been constructed using high speed pipeline FFT processing techniques developed for radar signal processors. The effective image sample rate is...
详细信息
A feasibility method of a real time image bandwidth compression system has been constructed using high speed pipeline FFT processing techniques developed for radar signal processors. The effective image sample rate is 10 MHz and the bandwidth compression model provides a 4:1 compression, with excellent image fidelity. The model operates at a clock rate of 10 Mhz in computing the required two dimensional cosine transform and image compression and expansion algorithms.
暂无评论