The frequency-sensitive competitive learning (FSCL) algorithm and its associated VLSI neuroprocessor have been developed for adaptive vector quantisation (AVQ). Simulation results show that the FSCL algorithm is capab...
详细信息
The frequency-sensitive competitive learning (FSCL) algorithm and its associated VLSI neuroprocessor have been developed for adaptive vector quantisation (AVQ). Simulation results show that the FSCL algorithm is capable of producing a good-quality codebook for AVQ at high compression ratios of more than 20 in real time. This VLSI neural-network-based vector quantization design includes a fully parallel vector quantizer and a pipelined codebook generator to provide an effective data compression scheme. It provides a computing capability as high as 3.33 billion connections per second. Its performance can achieve a speedup of 750 compared with SUN-3/60 and a compression ratio of 33 at a signal-to-noise ratio of 23.81 dB.< >
We address the problem of reconstructing a smooth curve from sparse and noisy information that is invariant to the choice of the coordinate system. Tikhonov regularization is used to form a well-posed mathematical pro...
详细信息
We address the problem of reconstructing a smooth curve from sparse and noisy information that is invariant to the choice of the coordinate system. Tikhonov regularization is used to form a well-posed mathematical problem statement, and conditions for an invariant reconstruction are given. The resulting functional minimization problem is shown to be nonconvex. Approximations to the invariant functional are often used to form a convex problem that can be solved efficiently. Two common approximations, those of cubic and weighted cubic splines, are detailed, and examples are given to show that the approximations are often invalid. To form a valid approximation to the invariant functional we propose a two-step algorithm. The first step forms a piecewise-linear curve, which is invariant to the coordinate system. This piecewise-linear curve is then used to construct a parameterization of the curve for which we can make a valid approximation to the invariant functional. Examples are given to demonstrate the effectiveness of the algorithm, and two example applications for which the invariant property is important are given.
This paper investigates some approaches for designing one-dimensional linear phase finite-duration impulse-responses (FIR) notch filters, which are based on the modification of several established design techniques of...
详细信息
This paper investigates some approaches for designing one-dimensional linear phase finite-duration impulse-responses (FIR) notch filters, which are based on the modification of several established design techniques of linear phase FIR band-selective filters. Based on extensive design examples and theoretical analysis, formulae have been developed for estimating the length of a linear phase FIR notch filter meeting the given specifications. In addition, the design of two-dimensional linear phase FIR notch filters is briefly considered. Illustrative examples are included.
The image reconstruction capacity of the actual foveal cone sampling mosaics of an adult monkey and human is investigated through computer simulations. A retinocortical mapping model in which positions are known and d...
An algorithm that removes impulse noise in images using a modified dynamic programming method is outlined. The restoration process combines two operations. In the first operation, the pixels corrupted by impulse noise...
详细信息
An algorithm that removes impulse noise in images using a modified dynamic programming method is outlined. The restoration process combines two operations. In the first operation, the pixels corrupted by impulse noise are detected by a dynamic programming based searching strategy. In the second operation, the original gray levels of the noisy pixels are estimated by a linear interpolation or extrapolation approach. A separable process is used to detect the noisy pixels of the images. In order to preserve the details of the images and reduce the computational complexity of conventional dynamic programming, two modifications have been developed. One is called neighborhood searching with an adaptive figure of merit and the other is called overlapping block processing. The actual performance of the proposed technique is compared with that of the commonly used median filter by filtering noise-corrupted images. Some computational considerations are also discussed.< >
Developed is a new method of automated border detection using graph searching principles. The method takes into account the added information available in temporal or spatial sequences of images that are commonly avai...
详细信息
Developed is a new method of automated border detection using graph searching principles. The method takes into account the added information available in temporal or spatial sequences of images that are commonly available in biomedical applications. This method is applied to detect endocardial and epicardial borders in short-axis magnetic resonance images of the heart. Preliminary results show that it may be more robust in areas of noise when compared to a method based only on data from a single image. They suggest that this method could potentially improve performance of automated edge detection methods when spatial or temporal sequences of frames are available.< >
A technique is presented that can be used in the visualization and analysis of cardiac wall motion abnormalities by digital two-dimensional echocardiography. This technique is based on the use of a curvature function ...
详细信息
A technique is presented that can be used in the visualization and analysis of cardiac wall motion abnormalities by digital two-dimensional echocardiography. This technique is based on the use of a curvature function extracted from the endocardial boundary locations and can be used for shape or shape change analysis of the heart. The authors call the locations of high absolute curvature landmarks. Identification of landmarks on the endocardial boundary provides a simplified but powerful description of the boundary that allows visualization and analysis of wall motion. The simplification of the heart beating process is an excellent tool for the identification of infarcted areas of the heart.
Many image compression techniques involve segmentation of a gray level image. With such techniques, information is extracted that describes the regions in the segmented image, and this information is then used to form...
详细信息
Digital processing of black and white images has received most attention during the last 25 years, and has led to various algorithms for the enhancement, smoothing, and zooming of images. Due to the decreasing cost an...
详细信息
A new class of morphological filters is proposed for smoothing an image contaminated with noise. A multiple model that includes the combination of linear and nonlinear operations is used in the design of the new filte...
详细信息
A new class of morphological filters is proposed for smoothing an image contaminated with noise. A multiple model that includes the combination of linear and nonlinear operations is used in the design of the new filter. The performance of the averaging version of this new filter is similar to that of the alpha-trimmed mean filter. The structure-preserving properties of this new filter depend on the values assigned to the coefficients in the filter. The idempotent property is obtained when a closing-min and opening-max version of the filter is used. The root structure of the output signal is also investigated.< >
暂无评论