High connectivity of artificialneural network chip-embodiments combined with currently emerging 3-dimensionally stacked multichip modules for rear-time applications of target classification require a scrutiny for low...
详细信息
ISBN:
(纸本)0819428396
High connectivity of artificialneural network chip-embodiments combined with currently emerging 3-dimensionally stacked multichip modules for rear-time applications of target classification require a scrutiny for low power technology insertion. Conventional CMOS high power consumption limits the allowable density of synapse/neuron elements. However Silicon-On-Insulator (SOI) technology has the potential for successful implementation of-high density neural network because of the following unique features: (a) Operating voltage is reduced 3-fold from 5 to 1.5 volts, reducing power requirements by 9-fold;(b) Reduced substrate offers reduced capacitance and power and an increased speed;and, (c) Latch-up phenomenon is eliminated. Here we describe two practical winner/loser-take-all (W/LTA) circuits fabricated with 0.25 mu m fully depleted SOI technology that are useful for neuralnetworks and as compared to other such circuits offer considerable advantage of speed and performance. SPICE circuit simulations show that up to 9-bit resolution can be obtained between a winner and a loser input and with two cascaded circuits. Final characterization tests prove that constructing circuit elements from SOI technology would allow us to build large size neuralnetworks for practical applications.
This paper introduces 3-D profile noncontact measurement method by using a kind of artificialneural network-Backpropagation network. It makes a detail discussion on optical system, data acquisition, and how to vain t...
详细信息
ISBN:
(纸本)0819430196
This paper introduces 3-D profile noncontact measurement method by using a kind of artificialneural network-Backpropagation network. It makes a detail discussion on optical system, data acquisition, and how to vain the BP network. An object is projected by a laser stripe light, and the line image of the object is captured by a CCD camera The input data of BP network is given from a series of image on CCD by certain imageprocessing technology. Once a mapping relationship of coordinates between CCD camera image plane and related position in object's space is set up by using BP network, the image plane coordinates of any objects are used as the input data of BP network, it will immediately obtain the corresponding coordinates in space according to the output. Thus the profile can be easily established by these coordinates. This method reduces the requirement for system accuracy, and no need to test and align the whole system in advance accurately.
Rare event applications are characterized by the event-of-interest being hidden in a large volume of routine data. The key to success in such situations is the development of a cascade of data elimination strategies, ...
详细信息
ISBN:
(纸本)0819429155
Rare event applications are characterized by the event-of-interest being hidden in a large volume of routine data. The key to success in such situations is the development of a cascade of data elimination strategies, such that each stage enriches the probability of finding the event amidst the data retained for further processing. Automated detection of aberrant cells in cervical smear slides is an example of a rare event problem. Each slide can amount to 2.5 gigabytes of raw data and only 1 in 20 slides are abnormal. In this paper we examine the use of template matching, artificialneuralnetworks, integrated optical density and morphological processing as algorithms for the first data elimination stage. Based on the experience gained, we develop a successful strategy with improves the overall event probability in the retained data from 0.01 initially to 0.87 after the second stage of processing.
This book presents four keynote speeches, eight invited papers and over a hundred papers selected from 180 submissions from more than 25 countries around the world. The contributions investigate applications of comput...
详细信息
ISBN:
(数字)9789814528948
ISBN:
(纸本)9789810233525
This book presents four keynote speeches, eight invited papers and over a hundred papers selected from 180 submissions from more than 25 countries around the world. The contributions investigate applications of computational intelligence and multimedia in various areas, such as artificial intelligence, artificialneuralnetworks, pattern recognition, evolutionary computations, logic synthesis, fuzzy logic, imageprocessing, image retrieval, virtual reality, etc.
This Volume E81-A of the conference proceedings contains 44 papers. Topics discussed include general fundamental and boundaries, digital signal processing, systems and control, VLSI design technology and computer aide...
详细信息
This Volume E81-A of the conference proceedings contains 44 papers. Topics discussed include general fundamental and boundaries, digital signal processing, systems and control, VLSI design technology and computer aided design, modeling and simulation, algorithms and data structures, graphs and networks, spread spectrum technologies and applications, artificial intelligence and knowledge, human communications and ergonomics, neuralnetworks, image theory, universal personal communications, analog signal processing, nonlinear problems and information security.
Automatic Target Recognition (ATR) is a challenging problem in each case the need exists to observe targets, and to automatically recognize, characterize, and identify them. Moreover, the recognition, characterization...
详细信息
Automatic Target Recognition (ATR) is a challenging problem in each case the need exists to observe targets, and to automatically recognize, characterize, and identify them. Moreover, the recognition, characterization, and identification process must occur in a challenging observation environment. ATR typically involves background effects such as ground clutter or deliberate attempts to camouflage or obscure the presence of a target, or even in the presence of effects such as electronic counter measures (ECM) and deception. Thus, with limited historical data and in a challenging observation environment, the need exists for human-like vision and reasoning in near real time speed. Clearly, improvements to ATR is vital to military mission, affecting both targeting and survivability. In this paper, we present an image classification system for ATR applications using artificialneuralnetworks. The real challenge in this kind of problem is the huge feature vectors representing images to be introduced to neural network algorithms. In order to overcome this problem, increase the number of targets to be classified, and speed up the neural network training and processing, only image features necessary for the classification process need to be computed and used as neural network inputs. In this paper we present two different methods for image feature extraction using cosine and wavelet transform. Results from both of the two developed systems are presented and compared.
Perception is assisted by sensed impressions of the outside world but not determined by them. The primary organ of perception is the brain and, in particular, the cortex. With that in mind, we have sought to see how a...
详细信息
ISBN:
(纸本)0819428256
Perception is assisted by sensed impressions of the outside world but not determined by them. The primary organ of perception is the brain and, in particular, the cortex. With that in mind, we have sought to see how a computer-modeled cortex - the PCNN or Pulse Coupled neural Network - performs as a sensor fusing element. In essence, the PCNN is comprised of an array of integrate-and-fire neurons with one neuron for each input pixel. In such a system, the neurons corresponding to bright pixels reach firing threshold faster than the neurons corresponding to duller pixels. Thus, firing rate is proportional to brightness. In PCNNs, when a neuron fires it sends some of the resulting signal to its neighbors. This linking can cause a near-threshold neuron to fire earlier than it would have otherwise. This leads to synchronization of the pulses across large regions of the image. We can simplify the three-dimensional PCNN output by integrating out the time dimension. Over a long enough time interval, the resulting two-dimensional (x,y) pattern IS the input image. The PCNN has taken it apart and put it back together again. The shorter-term time integrals are interesting in themselves and will be commented upon in the paper. The main thrust of this paper is the use of multiple PCNNs mutually coupled in various ways to assemble a single two-dimensional pattern or fused image. Results of experiments on PCNN image fusion and an evaluation of its advantages are our primary objectives.
In the context of mobile robot, we have developed an artificialneural network permitting a pre-processing of foveal vision: the Retina model. This model is adaptative and its multi-resolution allows to detect a large...
详细信息
artificialneuralnetworks were used to classify blood cells. Compared with existing methods, neuralnetworks are more accurate, efficient, adaptable and information-rich. The implementation of the system in a PC/Wind...
详细信息
artificialneuralnetworks were used to classify blood cells. Compared with existing methods, neuralnetworks are more accurate, efficient, adaptable and information-rich. The implementation of the system in a PC/Windows NT environment using imageprocessing technology and database management allows for a variety of features to be extracted and a variety of training algorithms to be used. In this preliminary study, blood cell images are segmented to individual cells. Features for individual cells, including size, color content and shape related moments, are extracted and used as inputs to a multilayer neural network. Backpropagation and ALOPEX training algorithms were used to train the neural network. After less than 2000 training iterations using 95 training sets, the system recognized three kinds of blood cell in a correctness percentage of 100%. This module provides a platform to build a more sophisticated computational intelligent system for cell classification for clinical use.
The proceedings contain 17 papers. The topics discussed include: entropy-constrained learning vector quantization algorithms and their application in image compression;predictive vector quantization using neural netwo...
The proceedings contain 17 papers. The topics discussed include: entropy-constrained learning vector quantization algorithms and their application in image compression;predictive vector quantization using neuralnetworks;design of an adaptive genetic learning neural network system for image compression;binary image compression using identity mapping backpropagation neural network;block-predictive image coder of neural network in multiresolution domain;and comparison of ML parameter estimation and neural network classifier for texture classification.
暂无评论