The conversion of newspaper pages into digital resources is an important task that greatly contributes to the preservation of and access to newspaper archives. In this paper, an integrated methodology is presented for...
详细信息
The conversion of newspaper pages into digital resources is an important task that greatly contributes to the preservation of and access to newspaper archives. In this paper, an integrated methodology is presented for segmenting newspaper pages and identifying newspaper articles. In the first stage, a succession of imageprocessing and document analysis algorithms is employed for segmenting newspaper page images into various objects (text, images and drawings, titles). A rule based approach is subsequently applied to the objects identified during the page segmentation phase for reconstructing individual articles. Experimental results, obtained from a large testbed of old newspaper issues, are presented which clearly demonstrate the applicability of our integrated approach to successful newspaper page segmentation and identification of newspaper articles.
The paper presents and evaluates the design and the implementation of a self-checking neural system for photon event identification in intensified charge-coupled device detectors. The neural approach reveals more effe...
详细信息
The paper presents and evaluates the design and the implementation of a self-checking neural system for photon event identification in intensified charge-coupled device detectors. The neural approach reveals more effective than classical algorithmic approaches thanks to its learning through example ability. Implementation is accomplished by SRAM-based FPGAs, which have generated increasing interest in the space community. The adoption of suitable on-line fault detection techniques is illustrated taking into account in a specific way SEU induced faults. The techniques are based on AN coding, particularly 3N coding, which constitutes a reasonable trade-off between circuit complexity and computational delay. Estimations of circuit area overhead and fault coverage are reported.
The objective in the scenic beauty estimation (SBE) problem is to develop an automatic classification algorithm that matches human subjective ratings. algorithms such as principal components analysis (PCA) and decisio...
详细信息
The objective in the scenic beauty estimation (SBE) problem is to develop an automatic classification algorithm that matches human subjective ratings. algorithms such as principal components analysis (PCA) and decision trees (DT) have been applied to this problem with limited success, motivating our search for a better classifier. Since this is obviously a nonlinear classification problem, we applied two nonlinear techniques: independent component analysis (ICA) and support vector machines (SvMs). We evaluated these algorithms on a standard, publicly available data set using a variety of combinations of features. The optimally configured ICA and SvM systems achieved misclassification rates of 33.4% and 32.2% respectively. This is a significant improvement over the best results previously reported on this task: 36.6% for PCA and 43% for DT. Since ambiguity in the features space is a significant problem in this application, these results validate the effectiveness of nonlinear classification techniques.
This brief describes an algorithmic development of the optimal low-rank approximation (LRA) of multidimensional (M-D) signals with M greater than or equal to 3. The algorithms developed can he regarded as a dimensiona...
详细信息
This brief describes an algorithmic development of the optimal low-rank approximation (LRA) of multidimensional (M-D) signals with M greater than or equal to 3. The algorithms developed can he regarded as a dimensional generalization of the singular value decomposition (SvD) which is of fundamental importance for analyzing signals that can be represented in a matrix form. In particular, iterative algorithms for optimal and suboptimal LRA of three-dimensional (3-D) arrays are presented in detail. Application of the 3-D LRA to the compression of image sequences is discussed.
Ultrasound is one of the leading medical imaging modalities because it is safe, noninvasive, portable, easy to use, relatively inexpensive and displays images in real-time. Due to its real-time nature, an ultrasound m...
详细信息
Ultrasound is one of the leading medical imaging modalities because it is safe, noninvasive, portable, easy to use, relatively inexpensive and displays images in real-time. Due to its real-time nature, an ultrasound machine must be able to process its incoming data quickly. High computational and throughput requirements in modern ultrasound machines have restricted their internal design to algorithm-specific hardware with limited programmability. Adding new ultrasound imaging applications or improving a machine's internal algorithms can require costly hardware redesigns and replacements of boards or of the entire machine. In an effort to address these problems, we have reviewed each of the essential functions in modern ultrasound machines and analyzed their computational requirements in programmable systems. These functions include dynamic downconversion, tissue signal processing, color flow processing, scan conversion, and tissue/flow decision. Our estimate of the total computing requirement to current ultrasound machines is 54.71 billion operations per second (BOPS) when transcendental functions are implemented in software and 31.26 BOPS when transcendental functions are implemented in lookup tables taking 160.26 Mbytes of memory. To investigate the feasibility of programmable generalized ultrasound systems, we have designed a flexible and parallel programmable ultrasound processing subsystem, called the Programmable Ultrasound image Processor (PUIP). As the need for programmable ultrasound machines increases in the future due to various advantages (e.g., lower system cost, faster clinical use and lower research/development expenses), it will be crucial to develop not only a high-performance, scalable, reconfigurable parallel computer architecture meeting the computing requirement, but also efficient ultrasound algorithms that can be optimally mapped into the parallel architecture. (C) 1998 Published by Elsevier Science B.v. All rights reserved.
In this paper, a general philosophy about feature windows based applications and a windows-based application are presented for analysis, algorithm development, testing and validation studies for Signal, image and Data...
详细信息
ISBN:
(纸本)0819428949
In this paper, a general philosophy about feature windows based applications and a windows-based application are presented for analysis, algorithm development, testing and validation studies for Signal, image and Data processing (SIDP) for Space-Based Surveillance (SBS) Applications. This dedicated facility is called AUG_SIDP. It performs several specialized tasks such as blur estimation, restoration, CFAR detection, clutter modeling, registration, pixel and data level fusion, target tracking and classification. It is still in the development and testing phase.
Parallel implementations of algorithms for medical imageprocessing mostly focus on the use of multiprocessor parallelism. Modern processor architectures however, provide several additional forms of parallelism at the...
详细信息
Parallel implementations of algorithms for medical imageprocessing mostly focus on the use of multiprocessor parallelism. Modern processor architectures however, provide several additional forms of parallelism at the processor level: subword parallelism, speculative execution, superscalar pipelining, very long instruction word, etc. In this article, we show that well-known parallelization techniques for multiprocessor systems can be used to exploit subword parallelism. Loop unrolling, loop fusion and if-hoisting prove to be valuable to achieve this goal. To illustrate this, we transformed the inner loops of a positron emission tomography image reconstruction algorithm. We achieved a speed-up of 45% on Sun's UltraSPARC processor. (C) 1998 Elsevier Science B.v. All rights reserved.
We present a novel content-adaptive multiresolution SAR image formation processing algorithm that incorporates dynamic, on-line detection algorithms into the image formation process. The idea is to vary image resoluti...
详细信息
ISBN:
(纸本)0819428191
We present a novel content-adaptive multiresolution SAR image formation processing algorithm that incorporates dynamic, on-line detection algorithms into the image formation process. The idea is to vary image resolution locally depending on scene content, focusing the SAR imagery to fine resolution only in regions where the scene reflectivity varies rapidly, while forming the rest of the image at coarser resolution or with reduced fidelity. Our "decision-directed" SAR image formation algorithm may have applications in systems where on-board processing or datalink constraints limits the area coverage rates or resolution. We present examples of this multiresolution SAR processing on SAR imagery and show that compression rates on the order of 70:1 or more (i.e., 0.45 bits/pixel starting from 32bits/complex sample, 16 bits/I, 16 bits/Q), can be obtained while still preserving coherent target signatures and with minor degradation in perceptual image quality.
New Class 3 imageprocessingalgorithms are presented. They are direct extensions of previously published one-dimensional algorithms. Class 3 algorithms require almost no apriori information knowledge about the signal...
详细信息
New Class 3 imageprocessingalgorithms are presented. They are direct extensions of previously published one-dimensional algorithms. Class 3 algorithms require almost no apriori information knowledge about the signal and noise that are being processed. Their performance depends upon the kind of smoothing used and on the images being processed by the filter. The previously published Class 3 filter algorithms require that the filter input be stationary, and that the noise spectrum have zero mean and be uncorrelated to the signa. For the new Class 3 imageprocessingalgorithms, the only additional assumption for the noise is that its spectrum be white. Simulations using Lena demonstrate such better performance using the new Class 3 algorithms over the standard Class S algorithms.
An investigation is made concerning implementations of competitive learning algorithms in analog vLSI circuits and systems. Analog and low power digital circuits for competitive learning are currently important for th...
详细信息
An investigation is made concerning implementations of competitive learning algorithms in analog vLSI circuits and systems. Analog and low power digital circuits for competitive learning are currently important for their applications in computationally-efficient speech and image compression by vector quantization, as required for example in portable multi-media terminals. A summary of competitive learning models is presented to indicate the type of vLSI computations required, and the effects of weight quantization are discussed. Analog circuit representations of computational primitives for learning and evaluation of distortion metrics are discussed. The present state of vLSI implementations of hard and soft competitive learning algorithms are described, as well as those for topological feature maps. Tolerance of learning algorithms to observed analog circuit properties is reported. New results are also presented from simulations of frequency-sensitive and soft competitive learning concerning sensitivity of these algorithms to precision in vLSI learning computations. Applications of these learning algorithms to unsupervised feature extraction and to vector quantization of speech and images are also described.
暂无评论