Analog CMOS implementation of a CNN-based locomotion controller with floating-gate devices

This paper proposes an analog CMOS circuit that implements a central pattern generator (CPG) for locomotion control in a quadruped walking robot. Our circuit is based on an affine transformation of a reaction-diffusion cellular neural network (CNN), and uses differential pairs with multiple-input floating-gate (MIFG) MOS transistors to implement both the nonlinearity and summation of CNN cells. As a result, the circuit operates in voltage mode, and thus it is expected to reduce power consumption. Due to good matching accuracy of devices, the circuit generates stable rhythmic patterns for robot locomotion control. From experimental results on fabricated chip using a standard CMOS 1.5-/spl mu/m process, we show that the chip yields the desired results; i.e., stable rhythmic pattern generation and low power consumption.     

A CNN-based chip for robot locomotion control

In this paper, the paradigm of emergent computation is applied to locomotion control in legged robots: the locomotion gait is the result of self-organization of a network of locally coupled nonlinear oscillators. This means to adopt the biological paradigm of central pattern generator (CPG), implemented by using cellular neural networks (CNNs). The whole control strategy is hybrid in the sense that the gait generation is accomplished by a fully analog CNN, while a simple logic unit modulates the behavior of the CNN-based CPG, so that the strategy is suitable to eventually include sensory feedback. The design of a VLSI chip implementing the CNN-based CPG and some experimental results on the chip are presented. The chip is designed using a switched-capacitor technique, fundamental to obtain in a simple and direct way some key features of the hybrid control discussed. The experimental results confirm the suitability of the approach.     

A recurrent fuzzy cellular neural network system with automatic structure and template learning

It is widely accepted that using a set of cellular neural networks (CNNs) in parallel can achieve higher level information processing and reasoning functions either from application or biologics points of views. Such an integrated CNN system can solve more complex intelligent problems. In this paper, we propose a novel framework for automatically constructing a multiple-CNN integrated neural system in the form of a recurrent fuzzy neural network. This system, called recurrent fuzzy CNN (RFCNN), can automatically learn its proper network structure and parameters simultaneously. The structure learning includes the fuzzy division of the problem domain and the creation of fuzzy rules and CNNs. The parameter learning includes the tuning of fuzzy membership functions and CNN templates. In the RFCNN, each learned fuzzy rule corresponds to a CNN. Hence, each CNN takes care of a fuzzily separated problem region, and the functions of all CNNs are integrated through the fuzzy inference mechanism. A new online adaptive independent component analysis mixture-model technique is proposed for the structure learning of RFCNN, and the ordered-derivative calculus is applied to derive the recurrent learning rules of CNN templates in the parameter-learning phase. The proposed RFCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. The capability of the proposed RFCNN is demonstrated on the real-world defect inspection problems. Experimental results show that the proposed scheme is effective and promising.     

A Recurrent Fuzzy Coupled Cellular Neural Network System With Automatic Structure and Template Learning

The cellular neural network (CNN) is a powerful technique to mimic the local function of biological neural circuits, especially the human visual pathway system, for real-time image and video processing. Recently, many studies show that an integrated CNN system can solve more complex high-level intelligent problems. In this brief, we extend our previously proposed multi-CNN integrated system, called recurrent fuzzy CNN (RFCNN) which considers uncoupled CNNs only, to automatically learn the proper network structure and parameters simultaneously of coupled CNNs, which is called recurrent fuzzy coupled CNN (RFCCNN). The proposed RFCCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. For comparison, the capability of the proposed RFCCNN is demonstrated on the same defect inspection problems. Simulation results show that the proposed RFCCNN outperforms the RFCNN.     

A Review of FPGA-Based Custom Computing Architecture for Convolutional Neural Network Inference

Convolutional neural network (CNN) has been widely adopted in many tasks. Its inference process is usually applied on edge devices where the computing resources and power consumption are limited. At present, the performance of general processors cannot meet the requirement for CNN models with high computation complexity and large number of pa-rameters. Field-programmable gate array (FPGA)-based custom computing architecture is a promising solution to further enhance the CNN inference performance. The software/hardware co-design can effectively reduce the computing overhead, and improve the inference performance while ensuring accuracy. In this paper, the mainstream methods of CNN structure design, hardware-oriented model compression and FPGA-based custom architecture design are summarized, and the improvement of CNN inference performance is demonstrated through an example. Challenges and possible research directions in the future are concluded to foster research efforts in this domain.     

基于快速滤波算法的卷积神经网络加速器设计

为减少卷积神经网络(CNN)的计算量,该文将2维快速滤波算法引入到卷积神经网络,并提出一种在FPGA上实现CNN逐层加速的硬件架构。首先,采用循环变换方法设计行缓存循环控制单元,用于有效地管理不同卷积窗口以及不同层之间的输入特征图数据,并通过标志信号启动卷积计算加速单元来实现逐层加速;其次,设计了基于4并行快速滤波算法的卷积计算加速单元,该单元采用若干小滤波器组成的复杂度较低的并行滤波结构来实现。利用手写数字集MNIST对所设计的CNN加速器电路进行测试,结果表明:在xilinx kintex7平台上,输入时钟为100 MHz时,电路的计算性能达到了20.49 GOPS,识别率为98.68%。可见通过减少CNN的计算量,能够提高电路的计算性能。     

New horizon for CNN: with fuzzy paradigms for multimedia

The cellular neural network (CNN) is a powerful technique to mimic the local function of biological neural circuits for real-time image and video processing. Recently, it is widely accepted that using a set of CNNs in parallel can achieve higher-level information processing and reasoning functions either from application or biology points of views. The authors introduce a novel framework for constructing a multiple-CNN integrated neural system called recurrent fuzzy CNN (RFCNN). This system can automatically learn its proper network structure and parameters simultaneously. In the RFCNN, each learned fuzzy rule corresponds to a CNN. Hence, each CNN takes care of a fuzzily separated problem region, and the functions of all CNNs are integrated through the fuzzy inference mechanism. Some online clustering algorithms are introduced for the structure learning, and the ordered-derivative calculus is applied to derive the recurrent learning rules of CNN templates in the parameter-learning phase. RFCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. The capability of the RFCNN is demonstrated on the real-world vision-based defect inspection and image descreening problems proving that the RFCNN scheme is effective and promising.     

异号权重模板CNN的似细化特性及其应用

本文发现并以定理的形式证明了具有异号权重模板的细胞神经网络系统在非均匀增益分段性输函数下的细 化稳态性性。     

基于全卷积神经网络的SAR图像目标分类

近年来,卷积神经网络(Convolutional Neural Network,CNN)在合成孔径雷达(Synthetic Aperture Radar,SAR)图像目标分类中取得了较好的分类结果。CNN结构中,前面若干层由交替的卷积层、池化层堆叠而成,后面若干层为全连接层。全卷积神经网络（All Convolutional Neural Network, A-CNN）是对CNN结构的一种改进,其中池化层和全连接层都用卷积层代替,该结构已在计算机视觉领域被应用。针对公布的MSTAR数据集,提出了基于A-CNN的SAR图像目标分类方法,并与基于CNN的SAR图像分类方法进行对比。实验结果表明,基于A-CNN的SAR图像目标分类正确率要高于基于CNN的分类正确率。     

音频标记一致性约束CRNN声音事件检测

级联卷积神经网络(CNN)结构和循环神经网络(RNN)结构的卷积循环神经网络(CRNN)及其改进是当前主流的声音事件检测模型.然而,以端到端方式训练的CRNN声音事件检测模型无法从功能上约束CNN和RNN结构的作用.针对这一问题,该文提出了音频标记一致性约束CRNN声音事件检测方法(ATCC-CRNN).该方法在CRN...     

一种新的细胞神经网络阴影检测器的研究

该文深入研究了一种新的二维细胞自动机(CA),找到了几种新的算法规则可以用来实现字符的粗化处理和阴影检测,并且用这些规则设计了几种新的细胞神经网络,文中详细介绍了这些算法规则的布尔代数表达式和细胞神经网络学习算法。仿真结果证明了这种新的细胞神经网络是简单而有效的,同时也证明了可以用 CA规则来设计新的细胞神经网络,为细胞神经网络的设计找到了一种新颖有效的方法。     

Realization of Boolean Functions via CNN: Mathematical Theory, LSBF and Template Design

As a paradigm for nonlinear spatial-temporal processing, cellular nonlinear networks (CNN) are biologically inspired systems where computation emerges from a collection of simple locally coupled nonlinear cells. Our investigation is an exploration of an important and difficult aspect of implementing arbitrary Boolean functions by using CNN. A typical class of basic key Boolean functions is the class of linearly separable ones. In this paper, we focus on establishing a complete set of mathematical theories for the linearly separable Boolean functions (LSBF) that are identical to a class of uncoupled CNN. First, we obtain an essential relationship between the template and the offset levels as well as the basis of the binary input vector set in the uncoupled CNN. More precisely, we construct a neat binary input–output truth table and some interesting properties of the offset levels of the uncoupled CNN, and develop a practical design formula for the class of CNN template. Especially, we found a criterion for LSBF, which depends only on symbolic relations between a Boolean function's outputs. Furthermore, we develop a method for representing any linearly nonseparable Boolean function into a logic operation of a sequence of linearly separable ones for a small number of inputs.     

Real-time license plate detection and recognition using deep convolutional neural networks

Automatic License Plate Recognition (ALPR) is an important task with many applications in Intelligent Transportation and Surveillance systems. This work presents an end-to-end ALPR method based on a hierarchical Convolutional Neural Network (CNN). The core idea of the proposed method is to identify the vehicle and the license plate region using two passes on the same CNN, and then to recognize the characters using a second CNN. The recognition CNN massively explores the use of synthetic and augmented data to cope with limited training datasets, and our results show that the augmentation process significantly increases the recognition rate. In addition, we present a novel temporal coherence technique to better stabilize the OCR output in videos. Our method was tested with publicly available datasets containing Brazilian and European license plates, achieving accuracy rates better than competitive academic methods and a commercial system.     

细胞神经网络在通信信号处理中的研究进展

基于非线性理论的通信信号处理一直是信号处理领域的热点研究问题。细胞神经网络(CNN)作为最易于VLSI实现的一类神经网络 ,是非线性理论的一个重要研究方向 ,近年来在通信信号处理领域取得了许多重要进展。文中主要介绍了细胞神经网络的基本理论、结构及其在通信信号处理中的研究进展 ,并指出其今后在通信信号处理中的进一步研究方向     

基于卷积神经网络的雷达目标HRRP分类识别方法

卷积神经网络通过卷积和池化操作提取图像在各个层次上的特征进而对目标进行有效识别,是深度学习网络中应用最广泛的一种。文中围绕一维距离像雷达导引头自动目标识别,开展基于卷积神经网络的目标高分辨距离像分类识别方法研究。首先,基于空中目标一维距离像姿态敏感性仿真生成近似平行交会条件下不同类型目标的高分辨距离像数据集;其次,构建一种一维卷积神经网络结构对目标高分辨距离像进行分类识别;作为比较,针对同类高分辨距离像数据集,分析了主成分分析-支持向量机方法的目标分类识别效果。结果表明:基于卷积神经网络的目标分类识别算法有更好的识别能力,对高分辨距离像的姿态敏感性具有较强的适应性。     

基于细胞神经网络的肝脏核磁共振图像自动分割方案

Currently, the processing speed of existing automatic liver segmentation for Magnetic Resonance Imaging (MRI) images is relatively slow. An automatic liver segmentation scheme for MRI images based on Cellular Neural Networks (CNN) is presented in this paper. It ensures the validity of this scheme and at the same time completes the image segmentation faster to accurately calculate the liver volume by using parallel computing in real time. In order to facilitate the CNN image processing, firstly, three-dimensional liver MRI images should be transformed into binary images; secondly, an appropriate template parameter of the Global Connectivity Detection CNN (GCD CNN) shall be selected to probe the connectivity of the liver to extract the entire liver; and then the Hole-Filler CNN (HF CNN) are used to repair the entire extracting liver and improve the accuracy of liver segmentation; finally, the liver volume is obtained. Results show that the scheme can ensure the accuracy of the automatic segmentation of the liver, and it can also improve the processing speed at the same time. The liver volume calculated is in line with the clinical diagnosis.     

基于混沌神经网络的移动通信信道分配方法研究

该文应用混沌神经网络求解信道分配问题,给出了信道分配的能量函数表达式和混沌神经网络模型,研究了判别混沌神经网络混沌特性的Lyapunov指数法,讨论了网络模型参数对网络混沌特性的影响,提出了基于混沌神经网络的信道分配算法.仿真结果表明,混沌神经网络具有复杂的瞬态混沌特性,它比Hopfield网络具有更强的搜索全局最优解的能力,和更快的收敛速度.     

基于轨道空间压缩的混沌神经网络控制

该文提出了基于轨道空间压缩的混沌神经网络控制方法,利用该方法对混沌神经网络进行控 制,使神经网络的输出稳定地收敛于与网络起始模式有最小汉明距离的存储模式或其反相模式上。该控制方法简单易行,物理意义明确。     

Multi-focus image fusion with the all convolutional neural network

A decision map contains complete and clear information about the image to be fused, which is crucial to various image fusion issues, especially multi-focus image fusion. However, in order to get a satisfactory image fusion effect, getting a decision map is very necessary and usually difficult to finish. In this letter, we address this problem with convolutional neural network (CNN), aiming to get a state-of-the-art decision map. The main idea is that the max-pooling of CNN is replaced by a convolution layer, the residuals are propagated backwards by gradient descent, and the training parameters of the individual layers of the CNN are updated layer by layer. Based on this, we propose a new all CNN (ACNN)-based multi-focus image fusion method in spatial domain. We demonstrate that the decision map obtained from the ACNN is reliable and can lead to high-quality fusion results. Experimental results clearly validate that the proposed algorithm can obtain state-of-the-art fusion performance in terms of both qualitative and quantitative evaluations.     

Focal-Plane and Multiple Chip VLSI Approaches to CNNs

In this paper, three alternative VLSI analog implementations of CNNs are described, which have been devised to perform image processing and vision tasks: a programmable low-power CNN with embedded photo-sensors, a compact fixed-template CNN based on unipolar current-mode signals, and basic CMOS circuits to implement an extended CNN model using spikes. The first two VLSI approaches are intended for focal-plane image processing applications. The third one allows, since its dynamics is defined by process-independent local ratios and its input/outputs can be efficiently multiplexed in time, the construction of very large multiple chip CNNs for more complex vision tasks.