This paper explores two issues which are relevant in practical halftoning situations on the CNN universal machine: block processing of large images with small CNN arrays, and the use of no larger than 3/spl times/3 te...
详细信息
This paper explores two issues which are relevant in practical halftoning situations on the CNN universal machine: block processing of large images with small CNN arrays, and the use of no larger than 3/spl times/3 templates. It is shown that block processing can be performed without noticeable boundary artifacts by careful selection of boundary cell values. In this example, a standard 3/spl times/3 halftoning template is used; higher quality halftones can be obtained only by using larger templates. A CNNUM algorithm is introduced which uses only a 3/spl times/3 template but emulates a much larger effective template through an iterative procedure. The method is to discretize the CNN transient in time and then implement the spatial correlations at each time step with a CNN transient. An A-B-template pair was designed for a single CNN transient to approximate a very simple linear filter model of the human visual system. The resulting discrete-time system was analyzed. The iterative procedure is demonstrated to produce a visually pleasing halftone.
This paper introduces a neural network architecture called R2MAP, which is based on the representational redescription hypothesis in cognitive science and adaptive resonance theory (ART) neural networks. The R2MAP net...
详细信息
This paper introduces a neural network architecture called R2MAP, which is based on the representational redescription hypothesis in cognitive science and adaptive resonance theory (ART) neural networks. The R2MAP network learns to classify arbitrary sequences of input patterns using a re-iterative process whereby knowledge that gets embedded in the network via ARTMAP-style error-driven learning is redescribed and becomes available to it for further learning. The knowledge redescription phase is triggered when the perceived level of difficulty of the given task exceeds a certain threshold, and is achieved through the dynamic creation of new features that better distinguish between output classes. This way the R2MAP network is capable of learning complex, relational input-output dependencies that cannot be represented efficiently using solely the features extracted through ordinary learning of statistical relationships. A simple proof-of-concept example is presented to illustrate the main ideas. Some related work is also discussed.
Printed circuit board layout inspection methods are mostly based on local geometric information, therefore they are well suited to the cellular neural networks (CNN) paradigm. The wire break, the wire and isolation wi...
详细信息
Printed circuit board layout inspection methods are mostly based on local geometric information, therefore they are well suited to the cellular neural networks (CNN) paradigm. The wire break, the wire and isolation width violation and an "H" type short circuits detector analogic algorithms were tested on a 20*22 CNN Universal Machine (CNNUM) chip working in the CNN Chip Prototyping System (CCPS) and on the CNN Engine Board (CNNEB), and the results were compared to the commercially available inspection systems.
Cellular neural Networks (CNN's) with space-varying interconnections are considered here to implement associative memories, A fast learning method is presented to compute the interconnection weights. The algorithm...
详细信息
Cellular neural Networks (CNN's) with space-varying interconnections are considered here to implement associative memories, A fast learning method is presented to compute the interconnection weights. The algorithm was carefully tested and compared to other methods. Storage capacity, noise immunity, and spurious state avoidance capability of the proposed system are discussed.
Markovian approaches to early vision processes need a huge amount of computing power. These algorithms can usually be implemented on parallel computing structures. with the Cellular neural Networks (CNN), a new image ...
详细信息
Markovian approaches to early vision processes need a huge amount of computing power. These algorithms can usually be implemented on parallel computing structures. with the Cellular neural Networks (CNN), a new image processing toot is coming into consideration, Its VLSI implementation takes place on a single analog chip containing several thousands of cells. Herein se use the CNN UM architecture for statistical image segmentation. The Modified Metropolis Dynamics (MMD) method can be implemented into the raw analog architecture of the CNN, We are able to implement a (pseudo) random held generator using one layer (one memory/cell) of the CNN. We can introduce the whole pseudostochastic segmentation process in the CNN architecture using 8 memories/cell, We use simple arithmetic functions (addition, multiplication), equality-test between neighboring pixels and very simple nonlinear output functions (step, jigsaw, With this architecture, a real VLSI CNN chip can execute a pseudostochastic relaxation algorithm of about 100 iterations in about 1 ms, In the proposed solution the segmentation is unsupervised. We have developed a pixel-level statistical estimation model. The CNN turns the original image into a smooth one. Then we have two gray-level values for every pixel: the original and the smoothed one. These two values are used for estimating the probability distribution of region label at a given pixel. Using the conventional first-order Markov Random Field (MRF) model, some misclassification errors remained at the region boundaries, because of the estimation difficulties in case of low SNR. By using a greater neighborhood, this problem has been avoided. In our CNN experiments, we used a simulation system with a fixed-point integer precision of 16 bits, Our results show that even in the case of the very constrained conditions of value-representations (the interval is (-64, +64), the accuracy is 0.002) can result in an effective and acceptable segmentation.
Demonstrated and motivated on human stereo vision analogic CNN algorithms are proposed to extract 3D spatial information from computer-generated random-dot stereograms as well as real scene random-dot like ones produc...
Demonstrated and motivated on human stereo vision analogic CNN algorithms are proposed to extract 3D spatial information from computer-generated random-dot stereograms as well as real scene random-dot like ones produced with simple optical devices, projector and camera. Several aspects of making real scene stereograms are considered to minimize perspective distortion and enable local CNN processing.
Template parameters of cellular neural networks (CNNs) should be robust enough to random variability of VLSI tolerances and noise. Using the CNN for image processing, one of the main problems is the robustness of a gi...
详细信息
Template parameters of cellular neural networks (CNNs) should be robust enough to random variability of VLSI tolerances and noise. Using the CNN for image processing, one of the main problems is the robustness of a given task in a real VLSI chip. It will be shown that very different tasks such as 2D or 3D deconvolution and texture segmentation can be solved in a real VLSI CNN environment without significant loss of efficiency and accuracy under low precision (about 6-8 bits) and random variability of the VLSI parameters. The CNN turns out to be very robust against template noise, image noise, imperfect estimation of templates and parameter accuracy. The parameters of a template are tuned using genetic learning. These optimized parameters depend on the precision of the architecture. It was found that about 6-8 bits of precision is enough for a complicated multilayer deconvolution, while only 4 bits of precision is enough for difficult texture segmentation in the presence of noise and parameter variances, The tolerance sensitivity of template parameters is considered for VLSI implementation. Theory and examples are demonstrated by many results using real-life microscopic images and natural textures.
In this article, a new analogic CNN algorithm to extract features of postage stamps in fray-scale images is introduced. The Gradient Controlled Diffusion method plays an important role in the approach. In our algorith...
详细信息
In this article, a new analogic CNN algorithm to extract features of postage stamps in fray-scale images is introduced. The Gradient Controlled Diffusion method plays an important role in the approach. In our algorithm, it is used for smoothing and separating Arabic figures drawn with a color which is similar to the background color. We extract Arabic figures in postage stamps by combining Gradient Controlled Diffusion with nearest neighbor linear CNN template and logic operations. Applying the feature extraction algorithm to different test images it has been verified that it is also effective in complete segmentation problems.
In this paper a cellular neural network (CNN) implementation of a reaction-diffusion system is described, which produces distance preserving periodic Turing patterns. The CNN with complex-valued templates are introduc...
详细信息
In this paper a cellular neural network (CNN) implementation of a reaction-diffusion system is described, which produces distance preserving periodic Turing patterns. The CNN with complex-valued templates are introduced, presenting an application for pattern generation. Finally a method for black-and-white pattern detection is described.
Adaptive histogram equalization (AHE), a method of contrast enhancement which is sensitive to local spatial information in image, has demonstrated its effectiveness in many applications. However, this technique is com...
详细信息
Adaptive histogram equalization (AHE), a method of contrast enhancement which is sensitive to local spatial information in image, has demonstrated its effectiveness in many applications. However, this technique is computationally intensive. In this paper we present two computational methods designed to fit well onto the locally interconnected array computer architecture of cellular neural networks (CNNs). CNNs are well known for their image processing capabilities, specially for grey-scale medical images and images of a natural scene. In many applications it would be very useful if the operation of a template or a complex analogic algorithm were highly illumination independent. Our results suggest that we can achieve this goal by using the AHE method in a pre-processing step.
暂无评论