CNN Universal Machine Research Articles

An Improved Cellular Nonlinear Network Architecture for Binary and Grayscale Image Processing

Cellular nonlinear networks (CNNs) constitute a very powerful paradigm for single instruction/multiple data computers with fine granularity. Analog and mixed-signal implementations have proven to be suitable for applications in high-speed image processing, robot control, medical signal processing, and many more. Especially digital emulations on field-programmable gate arrays (FPGAs) allow the development of general-purpose computers based on the CNN universal machine with an inherently parallel structure, a high degree of flexibility and a superior computational precision. However, these emulations turn out to be inefficient for the execution of binary operations, which account for more than two-thirds of all processing steps in a typical CNN algorithm. In this contribution, we present an architecture for the emulation of CNNs that supports both a fast and efficient processing of binary images, and a high computational accuracy when needed. With the FPGA implementation of this architecture, a speed-up factor of up to 5 is achieved for binary-data operations.

New Signal Processing Methods for the Development of Seizure Warning Devices in Epilepsy

The seizure prediction problem has been addressed by many researchers from very different fields for more than three decades. The vision of an implantable seizure prediction device may become reality soon: the first clinical study of such a device has been realized very recently and other realizations are not far behind. Cellular Nonlinear Networks (CNN) were firstly introduced by Chua and Yang in 1988 and later extended to an inherently parallel processing framework called the CNN Universal Machine (CNN-UM). This framework combines high computational power with low power consumption and miniaturized design—making it a very promising basis for the realization of a seizure warning device. In this contribution, we compare the seizure prediction performance of an eigenvalue based PCA-preprocessing followed by a nonlinear CNN signal prediction to the performance of a linear signal prediction approach followed by a level-crossing behavior analysis as well as to the performance of a combination of the two methods.

NEROvideo: a general-purpose CNN-UM video processing system

Emulations of cellular nonlinear networks (CNN) on digital reconfigurable hardware have proved to be adequate for highly-efficient computation of massive data, exceeding the accuracy and flexibility of full-custom designs. Based on a recently-proposed architecture for the emulation of a large-scale CNN universal machine, a new real-time video processing system has been developed. Due to its free programmability and massively-parallel architecture the system is very suitable for high-speed computation of complex algorithms that follow the idea of spatio-temporal computing. Implemented on a state-of-the-art Xilinx Zynq system-on-chip, the proposed setup is capable of processing a $$640\times 480$$640×480p video stream with up to 1,700 fps, depending on the respective algorithm.

Initial condition for efficient mapping of level set algorithms on many-core architectures

In this paper, we investigated the effect of adding more small curves to the initial condition which determines the required number of iterations of a fast level set (LS) evolution. As a result, we discovered two new theorems and developed a proof on the worst case of the required number of iterations. Furthermore, we found that these kinds of initial conditions fit well to many-core architectures. To show this, we have included two case studies which are presented on different platforms. One runs on a graphical processing unit (GPU) and the other is executed on a cellular nonlinear network universal machine (CNN-UM). With the new initial conditions, the steady-state solutions of the LS are reached in less than eight iterations depending on the granularity of the initial condition. These dense iterations can be calculated very quickly on many-core platforms according to the two case studies. In the case of the proposed dense initial condition on GPU, there is a significant speedup compared to the sparse initial condition in all cases since our dense initial condition together with the algorithm utilizes the properties of the underlying architecture. Therefore, greater performance gain can be achieved (up to 18 times speedup compared to the sparse initial condition on GPU). Additionally, we have validated our concept against numerically approximated LS evolution of standard flows (mean curvature, Chan-Vese, geodesic active regions). The dice indexes between the fast LS evolutions and the evolutions of the numerically approximated partial differential equations are in the range of 0.99±0.003.

The Re-Wiring of History

The Re-Wiring of History

Preprocessor embedding in cellular neural network

Cellular neural network (CNN) is a highly parallel analog circuit suitable for real‐time image processing. However, the needs of preprocessors in real‐world applications usually destroy the parallel processing nature of CNN. In this paper, we propose methods to embed the function of a preprocessor into a single CNN universal machine (CNN‐UM) such that the novel CNN contains the functions of both the preprocessor and the original CNN. The embedding processes are divided into three categories according to the structure of the B template in CNN. In the general case of nonzero B template (Case 1), an unbiased method and three approximation methods are proposed. The unbiased method is equivalent to its sequential counterpart, although templates of twice the neighborhood size should be used. Among the three approximation methods, AM III shows the least deviation from the unbiased method. In some experiments, AM II and AM III even outperform the unbiased method in noise‐elimination. In the two degenerated cases of Case 2 and Case 3 CNN, the embedding process can be fulfilled with templates of the original size. An exact solution can be derived for Case 2 while an approximate solution is developed for Case 3. The deviation due to the approximation in Case 3 is minor and the assumption made can even assist in noise elimination.

Stochastic bitstream-based CNN and its implementation on FPGA

In this paper, we present a new type of cellular nonlinear network (CNN) model and FPGA implementation of cellular nonlinear network universal machine. Our approach uses stochastic bitstreams as da...

FPGA-Based Real Time, Multichannel Emulated-Digital Retina Model Implementation

The function of the low-level image processing that takes place in the biological retina is to compress only the relevant visual information to a manageable size. The behavior of the layers and different channels of the neuromorphic retina has been successfully modeled by cellular neural/nonlinear networks (CNNs). In this paper, we present an extended, application-specific emulated-digital CNN-universal machine (UM) architecture to compute the complex dynamic of this mammalian retina in video real time. The proposed emulated-digital implementation of multichannel retina model is compared to the previously developed models from three key aspects, which are processing speed, number of physical cells, and accuracy. Our primary aim was to build up a simple, real-time test environment with camera input and display output in order to mimic the behavior of retina model implementation on emulated digital CNN by using low-cost, moderate-sized field-programmable gate array (FPGA) architectures.

Stochastic bitstream‐based CNN and its implementation on FPGA

AbstractIn this paper, we present a new type of cellular nonlinear network (CNN) model and FPGA implementation of cellular nonlinear network universal machine. Our approach uses stochastic bitstreams as data carriers. With the help of stochastic data streams more complex nonlinear cell interactions can be realized than conventional CNN hardware implementations have. The accuracy as indexed by bit depth resolution can be improved at the expense of computation time without influencing hardware complexity. Our simulation results prove the universality and utility of the model. Our experimental results show that the proposed model can be implemented on an FPGA hardware. Copyright © 2008 John Wiley & Sons, Ltd.

Implementation of embedded emulated‐digital CNN‐UM global analogic programming unit on FPGA and its application

AbstractThe paper addresses the issue of implementing an embedded global analogic programming unit (GAPU) on the reconfigurable emulated‐digital cellular neural/nonlinear networks universal machine (CNN‐UM) architecture that has been extended by a flexible Xilinx MicroBlaze soft processor core to take full advantage of the joint computing power of high‐speed distributed arithmetics and programmability. The implemented GAPU provides a stand‐alone operation, which is capable of controlling complex sophisticated CNN analogic algorithms similar to various visual microprocessors, such as the ACE4k, ACE16k, and Bi‐i vision systems. The quality of the embedded GAPU implementation is demonstrated by an analogic algorithm, in which sequences of template operations are required. Based on the experiments, several important issues relating to the acceleration efficiency, accuracy, cell size, and area consumption are discussed and compared with different CNN‐UM implementations. Copyright © 2008 John Wiley & Sons, Ltd.

Statistical physics on cellular neural network computers

The computational paradigm represented by Cellular Neural/nonlinear Networks (CNN) and the CNN Universal Machine (CNN-UM) as a Cellular Wave Computer, gives new perspectives also for computational statistical physics. Thousands of locally interconnected cells working in parallel, analog signals giving the possibility of generating truly random numbers, continuity in time and the optical sensors included on the chip are just a few important advantages of such computers. Although CNN computers are mainly used and designed for image processing, here we argue that they are also suitable for solving complex problems in computational statistical physics. This study presents two examples of stochastic simulations on CNN: the site-percolation problem and the two-dimensional Ising model. Promising results are obtained using an ACE16K chip with 128×128 cells. In the second part of the work we discuss the possibility of using the CNN architecture in studying problems related to spin-glasses. A CNN with locally variant parameters is used for developing an optimization algorithm on spin-glass type models. Speed of the algorithms and further trends in developing the CNN chips are discussed.

FROM DYNAMICAL EMERGING PATTERNS TO PATTERNS IN VISUAL ART

In this communication the relationship between emerging patterns generated by using CNN Universal Machine and patterns appearing in visual art masterpieces is discussed. The template based programming tools are outlined and a gallery of impressive examples is presented. The paper remarks the impressive role of Cellular Nonlinear Networks in covering the gap among art, science and information technology.

CNN-Based Hybrid-Order Texture Segregation as Early Vision Processing and Its Implementation on CNN-UM

In this paper, a biologically inspired, CNN-based, multi-channel, texture boundary detection technique is presented. The proposed approach is similar to human vision system. The algorithm is simple and straightforward such that it can be implemented on the cellular neural networks (CNNs). CNN contains several important advantages, such as efficient real-time processing capability and feasible very large-scale integration (VLSI) implementation. The proposed algorithm also had been widely tested on synthetic texture images. Those texture images are randomly selected from the Brodatz textures database (1966). According to our simulation results, the boundaries of uniform textures can be detected quite successfully. For the nonuniform or nonregular textures, the results also indicate meaningful properties, and the properties also are consistent to the human visual sensation. The proposed algorithm also has been implemented on the CNN universal machine (CNN-UM), and yields similar results as the simulation on the PC. Based on the efficient performance of CNN-UM, the algorithm becomes very fast.

Cellular Watersheds: A Parallel Implementation of the Watershed Transform on the CNN Universal Machine

The watershed transform has been used as a powerful morphological segmentation tool in a variety of image processing applications. This is because it gives a good segmentation result if a topographical relief and markers are suitably chosen for different type of images. This paper proposes a parallel implementation of the watershed transform on the cellular neural network (CNN) universal machine, called cellular watersheds. Owing to its fine grain architecture, the watershed transform can be parallelized using local information. Our parallel implementation is based on a simulated immersion process. To evaluate our implementation, we have experimented on the CNN universal chip, ACE16k, for synthetic and real images.

OPTIMAL CNN TEMPLATES FOR LINEARLY-SEPARABLE ONE-DIMENSIONAL CELLULAR AUTOMATA

In this tutorial, we present optimal Cellular Nonlinear Network (CNN) templates for implementing linearly-separable one-dimensional (1-D) Cellular Automata (CA). From the gallery of CNN templates presented in this paper, one can calculate any of the 256 1-D CA Rules studied by Wolfram using a CNN Universal Machine chip that is several orders of magnitude faster than conventional programming on a digital computer.

A Massively Parallel Face Recognition System

We present methods for processing the LBPs (local binary patterns) with a massively parallel hardware, especially with CNN-UM (cellular nonlinear network-universal machine). In particular, we present a framework for implementing a massively parallel face recognition system, including a dedicated highly accurate algorithm suitable for various types of platforms (e.g., CNN-UM and digital FPGA). We study in detail a dedicated mixed-mode implementation of the algorithm and estimate its implementation cost in the view of its performance and accuracy restrictions.

3D tactile sensor array processed by CNN‐UM: a fast method for detecting and identifying slippage and twisting motion

AbstractIn this paper, we present a fast and efficient technique for detecting and identifying the slippage and twisting motion of touching objects. This kind of action cannot be detected with tactile sensors sensing only the normal (perpendicular) component of the forces acting between surfaces. Our approach utilizes an integrated sensing–processing–actuating system comprising: (1) A 2 × 2 taxel (tactile pixel) array mounted on a two‐fingered robot hand, (2) a 64 × 64 CNN‐UM (Cellular Neural Network‐Universal machine), and (3) a closed‐loop controller. This arrangement, along with the proper analogic algorithm, allows detection and the control of the tactile event. It is essential to know and comprehend the forces between contact surfaces and the related 3D pressure fields is essential in many robotic applications discussed in the paper. Copyright © 2006 John Wiley & Sons, Ltd.

Stochastic simulations on the cellular wave computers

The computational paradigm represented by Cellular Neural/nonlinear Networks (CNN) and the CNN Universal Machine (CNN-UM) as a Cellular Wave Computer, gives new perspectives for computational physics. Many numerical problems and simulations can be elegantly addressed on this fully parallelized and analogic architecture. Here we study the possibility of performing stochastic simulations on this chip. First a realistic random number generator is implemented on the CNN-UM, and then as an example the two-dimensional Ising model is studied by Monte Carlo type simulations. The results obtained on an experimental version of the CNN-UM with 128 * 128 cells are in good agreement with the results obtained on digital computers. Computational time measurements suggests that the developing trend of the CNN-UM chips - increasing the lattice size and the number of local logic memories - will assure an important advantage for the CNN-UM in the near future.

PERSPECTIVES FOR MONTE CARLO SIMULATIONS ON THE CNN UNIVERSAL MACHINE

Possibilities for performing stochastic simulations on the analog and fully parallelized Cellular Neural Network UniversalMachine (CNN-UM) are investigated. By using a chaotic cellular automaton perturbed with the natural noise of the CNN-UM chip, a realistic binary random number generator is built. As a specific example for Monte Carlo type simulations, we use this random number generator and a CNN template to study the classical site-percolation problem on the ACE16K chip. The study reveals that the analog and parallel architecture of the CNN-UM is very appropriate for stochastic simulations on lattice models. The natural trend for increasing the number of cells and local memories on the CNN-UM chip will definitely favor in the near future the CNN-UM architecture for such problems.

A cellular nonlinear network: real-time technology for the analysis of microfluidicphenomena in blood vessels

A new approach to the observation and analysis of dynamic structural and functionalparameters in the microcirculation is described. The new non-invasive optical system isbased on cellular nonlinear networks (CNNs), highly integrated analogue processor arrayswhose processing elements, the cells, interact directly within a finite local neighbourhood.CNNs, thanks to their parallel processing feature and spatially distributed structure, arewidely used to solve high-speed image processing and recognition problems and in thedescription and modelling of biological dynamics through the solution of timecontinuous partial differential equations (PDEs). They are therefore consideredextremely suitable for spatial–temporal dynamic characterization of fluidic phenomenaat micrometric to nanometric scales, such as blood flow in microvessels and itsinteraction with the cells of the vessel wall. A CNN universal machine (CNN-UM)structure was used to implement, via simulation and hardware (ACE16k), thealgorithms to determine the functional capillarity density (FCD) and red blood cellvelocity (RBCV) in capillaries obtained by intravital microscopy during in vivoexperiments on hamsters. The system exploits the moving particles to distinguishthe functional capillaries from the stationary background. This information isused to reconstruct a map and to calculate the velocity of the moving objects.