A Low-Power 12-Bit SAR ADC for Analog Convolutional Kernel of Mixed-Signal CNN Accelerator

As deep learning techniques such as Convolutional Neural Networks(CNNs)are widely adopted,the complexity of CNNs is rapidly increasing due to the growing demand for CNN accelerator system-on-chip(SoC).Although conventional CNN accelerators can reduce the computational time of learning and inference tasks,they tend to occupy large chip areas due to many multiply-and-accumulate(MAC)operators when implemented in complex digital circuits,incurring excessive power consumption.To overcome these drawbacks,this work implements an analog convolutional filter consisting of an analog multiply-and-accumulate arithmetic circuit along with an analog-to-digital converter(ADC).This paper introduces the architecture of an analog convolutional kernel comprised of low-power ultra-small circuits for neural network accelerator chips.ADC is an essential component of the analog convolutional kernel used to convert the analog convolutional result to digital values to be stored in memory.This work presents the implementation of a highly low-power and area-efficient 12-bit Successive Approximation Register(SAR)ADC.Unlink most other SAR-ADCs with differential structure;the proposed ADC employs a single-ended capacitor array to support the preceding single-ended max-pooling circuit along with minimal power consumption.The SARADCimplementation also introduces a unique circuit that reduces kick-back noise to increase performance.It was implemented in a test chip using a 55 nm CMOS process.It demonstrates that the proposed ADC reduces Kick-back noise by 40%and consequently improves the ADC’s resolution by about 10%while providing a near rail-to-rail dynamic rangewith significantly lower power consumption than conventional ADCs.The ADC test chip shows a chip size of 4600μm^(2)with a power consumption of 6.6μW while providing an signal-to-noise-and-distortion ratio(SNDR)of 68.45 dB,corresponding to an effective number of bits(ENOB)of 11.07 bits.

LKAW: A Robust Watermarking Method Based on Large Kernel Convolution and Adaptive Weight Assignment

Robust watermarking requires finding invariant features under multiple attacks to ensure correct extraction.Deep learning has extremely powerful in extracting features,and watermarking algorithms based on deep learning have attracted widespread attention.Most existing methods use 3×3 small kernel convolution to extract image features and embed the watermarking.However,the effective perception fields for small kernel convolution are extremely confined,so the pixels that each watermarking can affect are restricted,thus limiting the performance of the watermarking.To address these problems,we propose a watermarking network based on large kernel convolution and adaptive weight assignment for loss functions.It uses large-kernel depth-wise convolution to extract features for learning large-scale image information and subsequently projects the watermarking into a highdimensional space by 1×1 convolution to achieve adaptability in the channel dimension.Subsequently,the modification of the embedded watermarking on the cover image is extended to more pixels.Because the magnitude and convergence rates of each loss function are different,an adaptive loss weight assignment strategy is proposed to make theweights participate in the network training together and adjust theweight dynamically.Further,a high-frequency wavelet loss is proposed,by which the watermarking is restricted to only the low-frequency wavelet sub-bands,thereby enhancing the robustness of watermarking against image compression.The experimental results show that the peak signal-to-noise ratio(PSNR)of the encoded image reaches 40.12,the structural similarity(SSIM)reaches 0.9721,and the watermarking has good robustness against various types of noise.

A Kernel-Based Convolution Method to Calculate Sparse Aerial Image Intensity for Lithography Simulation

Optical proximity correction (OPC) systems require an accurate and fast way to predict how patterns will be transferred to the wafer.Based on Gabor's 'reduction to principal waves',a partially coherent imaging system can be represented as a superposition of coherent imaging systems,so an accurate and fast sparse aerial image intensity calculation algorithm for lithography simulation is presented based on convolution kernels,which also include simulating the lateral diffusion and some mask processing effects via Gaussian filter.The simplicity of this model leads to substantial computational and analytical benefits.Efficiency of this method is also shown through simulation results.

ON THE SOLUTION OF THE SINGULAR INTEGRAL EQUATIONS WITH BOTH CAUCHY AND CONVOLUTION KERNEL

The following equations are basic forms of C-K equation (which is simplified in the following as singu-lar integral equations with convolution, that is C-K equations):where a,b,a_j,b_j are known constants or known functions, and find its solution f L_P(R), {0} or {α,β}.There were rather complete investigations on the method of solution for equations of Cauchy type aswell as integral equations of convolution type. But there is not investigation to the C-K equations, nodoubt, such that is important.

A Class of Singular Integral Equation of Convolution Type with CSC（τ- θ） Kernel

In this paper, we propose and discuss a class of singular integral equation of convolution type with csc(τ- θ) kernel in class L2[-π, π]. Using discrete Fourier transform and the lemma, this kind of equations is transformed to discrete system of equations, and then we obtain the solvable conditions and the explicit solutions in class L2[-π, π].

A multi-scale convolutional auto-encoder and its application in fault diagnosis of rolling bearings

Aiming at the difficulty of fault identification caused by manual extraction of fault features of rotating machinery,a one-dimensional multi-scale convolutional auto-encoder fault diagnosis model is proposed,based on the standard convolutional auto-encoder.In this model,the parallel convolutional and deconvolutional kernels of different scales are used to extract the features from the input signal and reconstruct the input signal;then the feature map extracted by multi-scale convolutional kernels is used as the input of the classifier;and finally the parameters of the whole model are fine-tuned using labeled data.Experiments on one set of simulation fault data and two sets of rolling bearing fault data are conducted to validate the proposed method.The results show that the model can achieve 99.75%,99.3%and 100%diagnostic accuracy,respectively.In addition,the diagnostic accuracy and reconstruction error of the one-dimensional multi-scale convolutional auto-encoder are compared with traditional machine learning,convolutional neural networks and a traditional convolutional auto-encoder.The final results show that the proposed model has a better recognition effect for rolling bearing fault data.

Image Augmentation-Based Food Recognition with Convolutional Neural Networks

Image retrieval for food ingredients is important work,tremendously tiring,uninteresting,and expensive.Computer vision systems have extraordinary advancements in image retrieval with CNNs skills.But it is not feasible for small-size food datasets using convolutional neural networks directly.In this study,a novel image retrieval approach is presented for small and medium-scale food datasets,which both augments images utilizing image transformation techniques to enlarge the size of datasets,and promotes the average accuracy of food recognition with state-of-the-art deep learning technologies.First,typical image transformation techniques are used to augment food images.Then transfer learning technology based on deep learning is applied to extract image features.Finally,a food recognition algorithm is leveraged on extracted deepfeature vectors.The presented image-retrieval architecture is analyzed based on a smallscale food dataset which is composed of forty-one categories of food ingredients and one hundred pictures for each category.Extensive experimental results demonstrate the advantages of image-augmentation architecture for small and medium datasets using deep learning.The novel approach combines image augmentation,ResNet feature vectors,and SMO classification,and shows its superiority for food detection of small/medium-scale datasets with comprehensive experiments.

Forest fire smoke recognition based on convolutional neural network

Traditional fire smoke detection methods mostly rely on manual algorithm extraction and sensor detection;however,these methods are slow and expensive to achieve discrimination.We proposed an improved convolutional neural network(CNN)to achieve fast analysis.The improved CNN can be used to liberate manpower.The network does not require complicated manual feature extraction to identify forest fire smoke.First,to alleviate the computational pressure and speed up the discrimination efficiency,kernel principal component analysis was performed on the experimental data set.To improve the robustness of the CNN and to avoid overfitting,optimization strategies were applied in multi-convolution kernels and batch normalization to improve loss functions.The experimental analysis shows that the CNN proposed in this study can learn the feature information automatically for smoke images in the early stages of fire automatically with a high recognition rate.As a result,the improved CNN enriches the theory of smoke discrimination in the early stages of a forest fire.

基于KPCA-CNN-DBiGRU模型的短期负荷预测方法

本文针对已有神经网络模型在短期负荷预测中输入维度过高、预测误差较大等问题,提出了一种结合核主成分分析、卷积神经网络和深度双向门控循环单元的短期负荷预测方法。先运用核主成分分析法对原始高维输入变量进行降维,再通过卷积深度双向门控循环单元网络模型进行负荷预测。以第九届全国电工数学建模竞赛试题A题中的负荷数据作为实际算例,结果表明所提方法较降维之前预测误差大大降低,与已有预测方法相比也有大幅的误差降低。

多向加权Tsallis熵最大化导向的自动阈值分割方法

受噪声或随机细节、目标和背景的大小比例、成像时的点扩散等不同因素的影响,许多图像的灰度直方图呈现为无模态、单模态、双模态或者多模态样式.为了在统一框架内处理这4种不同模态情形下的自动阈值选择问题,本文提出了一种多向加权Tsallis熵最大化导向的自动阈值分割方法(Multi-directional Weighted Tsallis Entropy,MWTE).基于新设计的反正切方向性卷积核的多尺度乘积效应,该方法将不同模态的灰度直方图转化为统一的单模态右偏灰度直方图.在4个不同方向上提取出这种特殊的单模态右偏灰度直方图后,通过多向加权策略构建出与原始图像灰度值紧密相关的加权Tsallis熵目标函数,并以该目标函数取最大值时对应的灰度值作为最终分割阈值.本文将提出的方法和3个阈值分割方法、1个软分割方法、1个活动轮廓分割方法以及1个自动聚类分割方法进行了比较.在4种不同模态情形下的4幅合成图像和50幅真实世界图像上的实验结果表明,本文提出的方法虽然在计算效率方面不占有优势,但它对不同模态的测试图像具有更稳健的分割适应性,且在量化分割精度所用的马修斯相关系数方面优于其他6个分割方法.

改进YOLOv5su模型检测桃树缩叶病

为实现自然环境下桃树缩叶病的检测,该研究提出了一种基于YOLOv5su的桃树缩叶病识别改进模型DLLYOLOv5su。首先,针对桃树缩叶病目标特征变化较大的问题,在骨干网络最后一层C3模块中加入可变形自注意力模块(deformable attention,DA),使模型更加关注目标区域,降低背景对模型的影响,提高模型在复杂背景下的拟合能力。其次在SPPF(fast spatial pyramid pooling)模块中引入LSKA(large separable kernel attention)结构,大核卷积增大了模型的感受野,使模型能够关注更多信息。最后,提出了LAWD(lightweight adaptive weighted downsampling)模块,使用轻量化的下采样结构替换卷积模块,减少计算开销。在桃树缩叶病数据集上进行试验,结果显示,DLL-YOLOv5su模型权重大小为17.6 MB,检测速度为83帧/s。识别准确率P、召回率R和平均精度均值mAP_(50)分别达到了80.7%、73.1%和80.4%,相较于原始YOLOv5su分别提高了4.2、2.4和4.3个百分点。与YOLOv3-tiny、Faster R-CNN、YOLOv7和YOLOv8相比mAP_(50)分别高出了28.5、11.8、2.1和4.1个百分点。改进模型识别精度高,误检、漏检率低,检测速度满足实时检测的要求,可以为桃树缩叶病的实时监测和预警提供参考。

基于改进SKNet-SVM的网络安全态势评估

为提高网络安全态势评估的准确率,增强稳定性与鲁棒性,提出一种基于改进选择性卷积核卷积神经网络和支持向量机的网络安全态势评估模型。首先,使用改进选择性卷积核代替传统卷积核进行特征提取,提高卷积神经网络感受野变化的自适应性,增强特征之间关联性。然后,将提取的特征输入到支持向量机中进行分类,并使用网格优化算法对支持向量机中的参数进行全局寻优。最后,根据网络攻击影响指标计算网络安全态势值。实验表明,基于改进选择性卷积核卷积神经网络和支持向量机的态势评估模型与传统的卷积神经网络搭建的态势评估模型相比,准确率更高,并且具有更强的稳定性和鲁棒性。

基于1D-WCWKCNN的痕量甲烷气体浓度检测

可调谐激光吸收光谱(TDLAS)技术在检测甲烷气体浓度过程中,甲烷气体透射光强二次谐波信号幅值与痕量甲烷气体浓度值成正比关系。如何准确和快速筛选目标甲烷透射光强二次谐波信号幅值至关重要。通过光电探测器获取1 000个甲烷气体透射光强信号样本,解调该透射光强信号获得二次谐波信号。人工在获得多种痕量甲烷气体透射光强和通过透射光强解调二次谐波信号时,存在噪声和人为操作对二次谐波信号的幅值产生影响,从而造成人工筛选二次谐波信号的时间增加。针对传统TDLAS技术筛选痕量甲烷气体二次谐波信号过程中存在高额时间成本的问题,提出了一种基于宽卷积和宽卷积核一维卷积神经网络(1D-WCWKCNN)的痕量甲烷气体浓度检测方法。首先,借助甲烷气体数据集训练1D-WCWKCNN模型,根据训练结果不断调整模型参数。其次,利用宽卷积层和宽卷积核一维卷积层对痕量甲烷气体二次谐波信号进行特征提取,使网络进行一次卷积后能够获得甲烷气体浓度信号中更长序列以及该序列边界信息与气体浓度之间的特征关系。甲烷透射光强二次谐波信号通过6层卷积层提取该信号与甲烷气体浓度关系的深层特征,然后通过6层最大池化层保留该信号与甲烷气体浓度关系的主要特征,再经过Flatten层将前一层处理的痕量甲烷气体透射光强二次谐波信号数据进行一维化处理。最后,根据训练好的1D-WCWKCNN模型通过Dense层输出痕量甲烷气体浓度。利用基于1D-WCWKCNN的痕量甲烷气体浓度检测模型代替了TDLAS技术中人工花费高额时间成本筛选二次谐波信号进行拟合直线对痕量甲烷气体浓度检测的过程。在实际实验中验证了该方法的有效性,实验结果表明利用该方法能够对50~1 000 mg·L^(-1)的痕量甲烷气体浓度进行有效检测,其准确度达到99.85%,与其他方法相比该方法信号特征提取能力强,检测甲烷气体精度高。该方法有助于气体检测领域中待测气体浓度信号的筛选。

基于电子鼻和一维拉普拉斯卷积核的奶粉基粉产地鉴别

奶粉基粉是配方奶粉的基础原材料,其产地影响到终端乳制品的品质。本文提出了一种电子鼻技术与一维拉普拉斯卷积核相结合的基粉奶源地判别方法。通过电子鼻采集样本数据,经过时域信号对齐,尝试使用不同阶数的一维拉普拉斯卷积核进行特征提取,并对比了统计数字特征、快速傅里叶变换、离散余弦变换等其他特征提取方法,最后使用偏最小二乘及可视化进行可分性分析。实验结果发现快速傅里叶变换、离散余弦变换、二阶的一维拉普拉斯卷积核相对于原始特征均有效提升了可分性,偏最小二乘的R2效应量从0.61分别提升到0.95、0.96和1.00。一维拉普拉斯卷积核特征提取方法能够准确区分产自国内和国外(澳大利亚)的基粉,在案例研究中取得了最优判别效果,说明其能够有效提取到电子鼻各通道序列信号的时间响应特征。该方法能够完成中澳两国奶粉基粉样本的区分工作,为快速鉴定奶粉基粉来源提供技术支撑。

门控机制的图像分类网络

为了提取更具表达能力和区分度的重点特征,减少网络传递时关键特征的流失,提高神经网络图像分类能力,提出一种新的门控机制图像分类网络(image classification Network of Gating Mechanism,GMNet).首先,使用门控卷积提取浅层特征,通过门控机制选择性地进行卷积操作,提高网络对原始图像关键特征的提取能力;其次,设计了一种插值门控卷积(Interpolation Gated Convolution,IGC)模块,利用Lanczos插值与门控卷积相结合,强化浅层特征的同时提取更具区分度的特征,提高特征的非线性表达能力;然后,设计了大核门控注意力机制(Large kernel Gated Attention Mechanism,LGAM)模块,将大核注意力与门控卷积相融合,实现了特征的选择性增强和选择性融合,提高关键区域特征的贡献度;最后,将大核门控注意力机制模块嵌入到残差分支中,让模型更有效地学习输入数据的特征和上下文信息,减少关键特征在网络信息传递时流失,提高网络的分类能力.本文方法在图像数据集CIFAR-10、CIFAR100、SVHN、Imagenette、Imagewoof上分别达到了97.05%、83.68%、97.68%、90.60%、83.05%的分类准确率,与当前先进的方法相比分别平均提高了3.26%、7.08%、3.44%、2.65%、5.02%.与现有主流网络模型相较,本文门控机制图像分类网络能够增强特征的非线性表达能力,提取更具表达能力和区分度的重点特征,减少关键特征流失,提高关键区域特征的贡献度,有效地提高神经网络图像分类能力.

基于多重机制优化YOLOv8的复杂环境下安全帽检测方法

为了解决建筑工地、隧道、煤矿等施工场景中现有安全帽检测算法对于小目标、密集目标以及复杂环境下的检测精度低的问题,提出了一种基于多重机制的安全帽检测方法。以YOLOv8n为基础将Backbone部分的C2f模块加入可扩张残差(DWR)注意力模块,使得网络能够更灵活地适应不同尺度的特征,以而更准确地识别图像中的物体;采用可变形卷积AKConv模块取代主干部分中的原始Conv,为卷积神经网络带来了显著的性能提升,从而实现更高效的特征提取。此外引用了大型可分离核注意力LSKA模块与SPPF结构相结合,大大增强了模型核心的融合能力。在Safety helmet数据集的实验结果表明,改进后的算法相较于原模型,mAP@0.5指标上提升了10.5个百分点,在mAP@0.5-0.95指标上提升了3.7个百分点,能有效提高复杂场景下的安全帽佩戴检测精度。

基于多核扩展卷积的无监督视频行人重识别

行人重识别旨在跨监控摄像头下检索出特定的行人目标。由于存在姿态变化、物体遮挡和背景干扰的不同成像条件等问题,导致行人特征提取不充分。本文提出一种利用多核扩展卷积的无监督视频行人重识别方法,使得提取到的行人特征能够更全面、更准确地表达个体差异和特征信息。首先,采用预训练的ResNet50作为编码器,为了进一步提升编码器的特征提取能力,引入了多核扩展卷积模块,通过增加卷积核的感受野,使得网络能够更有效地捕获到局部和全局的特征信息,从而更全面地描述行人的外貌特征;其次,通过解码器将高级语义信息还原为更为底层的特征表示,从而增强特征表示,提高系统在复杂成像条件下的性能;最后,在解码器的输出中引入多尺度特征融合模块融合相邻层中的特征,进一步减少不同特征通道层之间的语义差距,以产生更鲁棒的特征表示。在3个主流数据集上进行离线实验,结果表明该方法在准确性和鲁棒性上均取得了显著的改进。

行星齿轮箱降噪能力A-CNN模型及其智能诊断

为了降低随机噪声对行星齿轮箱振动信号造成的干扰,设计了一种采用卷积神经网络(A-CNN)算法的行星齿轮箱故障诊断,可以实现对噪声的良好抗干扰性能,采用A-CNN进行处理时可以通过Dropout实现输入信号的随机干扰,再以多尺度卷积核模块完成干扰信号开展特征分析和多尺度特征学习的过程。研究结果表明:采用Dropout处理信号后能够大幅提升模型抗噪性能,当设置3dB强噪条件时提升近10%。当噪声强度低于6dB时,(15×15)卷积核获得比了比(7×7)卷积核更优的效果。当噪声水平上升后,测试模型准确率降低。与其它算法进行比较可知,设计的A-CNN算法在各噪声水平测试集都达到了最优性能。当受到3dB强噪干扰时,A-CNN可以获得比AlexNet提升20%准确率,并且与VGG相比也可以提升近10%准确率。

基于改进时间卷积网络与藤Copula的短期风速预测

考虑风电场相邻风机风速间以及风速与气象因素间复杂的非线性关系,提出了一种基于改进时间卷积网络与藤Copula相结合的风速预测方法。首先,利用深度残差收缩网络中存在的注意力机制及软阈值化的思想改进时间卷积网络中的残差模块,并进行初步风速预测;然后,考虑到众多气象因素对风速的影响,使用核主成分分析对气象数据进行降维,在保证数据特征的同时,降低数据的复杂度;最后,利用藤Copula在描述非线性相关结构方面的优势构建修正模型,使用降维的气象数据修正初步风速预测值,得到最终的风速预测结果。实验证明,所提方法提高了短期风速预测的精度。

基于多尺度残差注意力域适应的轴承故障诊断

针对滚动轴承待监测样本在跨机器任务中诊断困难的问题,提出一种基于多尺度残差注意力域适应的滚动轴承故障诊断方法。该方法将滚动轴承振动信号直接作为多尺度注意力残差网络模块的输入,为更好提取源域与目标域的共同特征,该模块引入多尺度卷积提取特征信息、注意力机制的压缩激励网络解决数据差异性与残差网络的跨层连接,域自适应部分采用局部最大均值差异度量准则,并选择滚动轴承公开故障数据集进行对比与消融试验。试验结果表明:提出的多尺度残差注意力域适应的滚动轴承故障诊断方法在跨机器任务下平均识别精度达到99.1%,相比于其他方法具有较好的泛化性能。所得结论可为滚动轴承故障监测与诊断提供理论依据。