Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including lo...Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including low throughput,large computation cost,high latency,and poor flexibility,which limits the efficiency of online PCBA inspection.In this paper,a novel PCBA defect detection method based on a lightweight deep convolution neural network is proposed.In this method,the semantic segmentation model is combined with a rule-based defect recognition algorithm to build up a defect detection frame-work.To improve the performance of the model,extensive real PCBA images are collected from production lines as datasets.Some optimization methods have been applied in the model according to production demand and enable integration in lightweight computing devices.Experiment results show that the production line using our method realizes a throughput more than three times higher than traditional methods.Our method can be integrated into a lightweight inference system and pro-mote the flexibility of AOI.The proposed method builds up a general paradigm and excellent example for model design and optimization oriented towards industrial requirements.展开更多
Low-power and low-variability artificial neuronal devices are highly desired for high-performance neuromorphic computing.In this paper,an oscillation neuron based on a low-variability Ag nanodots(NDs)threshold switchi...Low-power and low-variability artificial neuronal devices are highly desired for high-performance neuromorphic computing.In this paper,an oscillation neuron based on a low-variability Ag nanodots(NDs)threshold switching(TS)device with low operation voltage,large on/off ratio and high uniformity is presented.Measurement results indicate that this neuron demonstrates self-oscillation behavior under applied voltages as low as 1 V.The oscillation frequency increases with the applied voltage pulse amplitude and decreases with the load resistance.It can then be used to evaluate the resistive random-access memory(RRAM)synaptic weights accurately when the oscillation neuron is connected to the output of the RRAM crossbar array for neuromorphic computing.Meanwhile,simulation results show that a large RRAM crossbar array(>128×128)can be supported by our oscillation neuron owing to the high on/off ratio(>10^(8))of Ag NDs TS device.Moreover,the high uniformity of the Ag NDs TS device helps improve the distribution of the output frequency and suppress the degradation of neural network recognition accuracy(<1%).Therefore,the developed oscillation neuron based on the Ag NDs TS device shows great potential for future neuromorphic computing applications.展开更多
In the past few decades,the Moore’s Law has been the revolutionary force for our integrated circuit(IC)industry.However,the tremendous challenges faced in continuous transistor physical down-scaling and the unprecede...In the past few decades,the Moore’s Law has been the revolutionary force for our integrated circuit(IC)industry.However,the tremendous challenges faced in continuous transistor physical down-scaling and the unprecedented demands for computing and storage capabilities require our urgent search for strategies and solutions to integrate diverse materials,devices,circuits,and architectures in a 3D vertically stacked manner so that they can orchestrate in the most effective way to provide significantly enhanced functionalities as well as superior speed,energy,bandwidth,form fact,and cost.展开更多
Traditional charge-based memories,such as dynamic random-access memory(DRAM)and flash,are approaching their scaling limits.A variety of resistance-based memories,such as phase-change memory(PCM),magnetic random-access...Traditional charge-based memories,such as dynamic random-access memory(DRAM)and flash,are approaching their scaling limits.A variety of resistance-based memories,such as phase-change memory(PCM),magnetic random-access memory(MRAM)and resistive random-access memory(RRAM),have been long considered for emerging memory applications thanks to their non-volatility,fast speed,low power,and compact size for potentially high-density integration.展开更多
[Objectives]To explore the effects of injection of oregano oil submicron emulsion on lipopolysaccharide-induced pneumonia in rats.[Methods]Rats induced by lipopolysaccharide were used as animal models of acute pneumon...[Objectives]To explore the effects of injection of oregano oil submicron emulsion on lipopolysaccharide-induced pneumonia in rats.[Methods]Rats induced by lipopolysaccharide were used as animal models of acute pneumonia.The experiment was divided into blank group,model group,administration group and positive drug control group.The morphology of lung tissue and the changes of cells and inflammatory factors in each group were observed,and the anti-inflammatory effects of injection of oregano oil submicron emulsion.[Results]The injection of oregano oil submicron emulsion can improve the pathological injury of rat lung tissue,inhibit the release of IL-6,IL-10,TNF-αcytokines induced by lipopolysaccharide,and significantly reduce the value of CRP.[Conclusions]Oregano oil submicron emulsion has a certain therapeutic effect on lipopolysaccharide-induced pneumonia in rats,and its mechanism may be related to reducing the release of cytoinflammatory factor IL-6,IL-10,and TNF-αand alleviating the injury to tissues and organs.展开更多
Memory cells have always been an important element of information technology.With emerging technologies like big data and cloud computing,the scale and complexity of data storage has reached an unprecedented peak with...Memory cells have always been an important element of information technology.With emerging technologies like big data and cloud computing,the scale and complexity of data storage has reached an unprecedented peak with a much higher requirement for memory technology.As is well known,better data storage is mostly achieved by miniaturization.However,as the size of the memory device is reduced,a series of problems,such as drain gate-induced leakage,greatly hinder the performance of memory units.To meet the increasing demands of information technology,novel and high-performance memory is urgently needed.Fortunately,emerging memory technologies are expected to improve memory performance and drive the information revolution.This review will focus on the progress of several emerging memory technologies,including two-dimensional material-based memories,resistance random access memory(RRAM),magnetic random access memory(MRAM),and phasechange random access memory(PCRAM).Advantages,mechanisms,and applications of these diverse memory technologies will be discussed in this review.展开更多
Training an artificial neural network with backpropagation algorithms to perform advanced machine learning tasks requires an extensive computational process.This paper proposes to implement the backpropagation algorit...Training an artificial neural network with backpropagation algorithms to perform advanced machine learning tasks requires an extensive computational process.This paper proposes to implement the backpropagation algorithm optically for in situ training of both linear and nonlinear diffractive optical neural networlks,which enables the acceleration of training speed and improvement in energy efficiency on core computing modules.We demonstrate that the gradient of a loss function with respect to the weights of diffractive layers can be accurately calculated by measuring the forward and backward propagated optical fields based on light reciprocity and phase conjunction principles.The diffractive modulation weights are updated by programming a high-speed spatial light modulator to minimize the error between prediction and target output and perform inference tasks at the speed of light.We numerically validate the effectiveness of our approach on simulated networks for various applications.The proposed in situ optical learning architecture achieves accuracy comparable to in silico training with an electronic computer on the tasks of object dlassification and matrix-vector multiplication,which further allows the diffractive optical neural network to adapt to system imperfections.Also,the self-adaptive property of our approach facilitates the novel application of the network for all-optical imaging through scattering media.The proposed approach paves the way for robust implementation of large-scale difractive neural networks to perform distinctive tasks all-optically.展开更多
The rapid growth of the Internet of Things(IoTs)has resulted in an explosive increase in data,and thus has raised new challenges for data processing units.Edge computing,which settles signal processing and computing t...The rapid growth of the Internet of Things(IoTs)has resulted in an explosive increase in data,and thus has raised new challenges for data processing units.Edge computing,which settles signal processing and computing tasks at the edge of networks rather than uploading data to the cloud,can reduce the amount of data for transmission and is a promising solution to address the challenges.One of the potential candidates for edge computing is a memristor,an emerging nonvolatile memory device that has the capability of in-memory computing.In this article,from the perspective of edge computing,we review recent progress on memristor-based signal processing methods,especially on the aspects of signal preprocessing and feature extraction.Then,we describe memristor-based signal classification and regression,and end-to-end signal processing.In all these applications,memristors serve as critical accelerators to greatly improve the overall system performance,such as power efficiency and processing speed.Finally,we discuss existing challenges and future outlooks for memristor-based signal processing systems.展开更多
En ergy storage devices with flexible form factor have become critical components of wearable electr onic systems.In spired by methods of monolithic integration in the microelectronics fabrication process,we propose a...En ergy storage devices with flexible form factor have become critical components of wearable electr onic systems.In spired by methods of monolithic integration in the microelectronics fabrication process,we propose a planar flexible full-solid-state lithium-ion battery(FSLB)architecture and a layer-by-layer stencil printing assembly method for fabricating batteries on polyethylene terephthalate(PET)substrate.FSLBs use quasi-solid electrolyte based on LiTFSI and ultraviolet(UV)-curable ethoxylated trimethylolpropane triacrylate(ETPTA)polymeric matrix in combination with Li4Ti50i2(LTO)/LiFePO4(LFP)-based electrodes.Excellent mechanical flexibility(<10 mm bending radius)can be achieved.The electrochemical characteristics of electrolyte,including ion conductivity,physical stability during room-temperature and tender assembly processes,are promising.A complete thin film-shape FSLB demonstrated working operation both under planar and bending conditions.The unique architecture and assembly processes open new ways for planar flexible devices to be integrated with flexible energy devices.展开更多
With the rapid growth of computer science and big data,the traditional von Neumann architecture suffers the aggravating data communication costs due to the separated structure of the processing units and memories.Memr...With the rapid growth of computer science and big data,the traditional von Neumann architecture suffers the aggravating data communication costs due to the separated structure of the processing units and memories.Memristive in-memory computing paradigm is considered as a prominent candidate to address these issues,and plentiful applications have been demonstrated and verified.These applications can be broadly categorized into two major types:soft computing that can tolerant uncertain and imprecise results,and hard computing that emphasizes explicit and precise numerical results for each task,leading to different requirements on the computational accuracies and the corresponding hardware solutions.In this review,we conduct a thorough survey of the recent advances of memristive in-memory computing applications,both on the soft computing type that focuses on artificial neural networks and other machine learning algorithms,and the hard computing type that includes scientific computing and digital image processing.At the end of the review,we discuss the remaining challenges and future opportunities of memristive in-memory computing in the incoming Artificial Intelligence of Things era.展开更多
All memristor neuromorphic networks have great potential and advantage in both technology and computational protocols for artificial intelligence.It is crucial to find suitable elementary units for both performing fea...All memristor neuromorphic networks have great potential and advantage in both technology and computational protocols for artificial intelligence.It is crucial to find suitable elementary units for both performing featured neuromorphic functions and fabrication in large scale.Here a simple memristive structure,Nb/HfOx/Pd,is proposed for this goal.Its two resistive switching mechanisms,Mott transition of NbO2 and oxygen vacancy(Vo)migration,can be controlled by modulating external bias directions.Negative bias activates reversible phase transition and restrains Vo filament formation to allow the memristor to mimic the firing action potential.Positive bias activates Vo filament formation and restrains the other to allow the memristor to mimic synaptic plasticity and learning protocols.The system can respond adaptively to naturally generated action potentials and modified synaptic signals from the same memristive structure.In addition,some special features related to signal encoding and recognition are discovered when the system is settled according to chaos circuit theory.Our study provides a novel approach for designing elementary units for neuromorphic computations.展开更多
Mathematical morphology operations are widely used in image processing such as defect analysis in semiconductor manufacturing and medical image analysis.These data-intensive applications have high requirements during ...Mathematical morphology operations are widely used in image processing such as defect analysis in semiconductor manufacturing and medical image analysis.These data-intensive applications have high requirements during hardware implementation that are challenging for conventional hardware platforms such as central processing units(CPUs)and graphics processing units(GPUs).Computation-in-memory(CIM)provides a possible solution for highly efficient morphology operations.In this study,we demonstrate the application of morphology operation with a novel memristor-based auto-detection architecture and demonstrate non-neuromoq)hic computation on a multi-array-based memristor system.Pixel-by-pixel logic computations with low parallelism are converted to parallel operations using memristors.Moreover,hardware-implemented computer-integrated manufacturing was used to experimentally demonstrate typical defect detection tasks in integrated circuit(IC)manufacturing and medical image analysis.In addition,we developed a new implementation scheme employing a four-layer network to realize small-object detection with high parallelism.The system benchmark based on the hardware measurement results showed significant improvement in the energy efficiency by approximately 358 times and 32 times more than when a CPU and GPU were employed,respectively,exhibiting the advantage of the proposed memristor-based morphology operation.展开更多
基金supported in part by the IoT Intelligent Microsystem Center of Tsinghua University-China Mobile Joint Research Institute.
文摘Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including low throughput,large computation cost,high latency,and poor flexibility,which limits the efficiency of online PCBA inspection.In this paper,a novel PCBA defect detection method based on a lightweight deep convolution neural network is proposed.In this method,the semantic segmentation model is combined with a rule-based defect recognition algorithm to build up a defect detection frame-work.To improve the performance of the model,extensive real PCBA images are collected from production lines as datasets.Some optimization methods have been applied in the model according to production demand and enable integration in lightweight computing devices.Experiment results show that the production line using our method realizes a throughput more than three times higher than traditional methods.Our method can be integrated into a lightweight inference system and pro-mote the flexibility of AOI.The proposed method builds up a general paradigm and excellent example for model design and optimization oriented towards industrial requirements.
基金supported in part by China Key Research and Development Program(2016YFA0201800)the National Natural Science Foundation of China(91964104,61974081)。
文摘Low-power and low-variability artificial neuronal devices are highly desired for high-performance neuromorphic computing.In this paper,an oscillation neuron based on a low-variability Ag nanodots(NDs)threshold switching(TS)device with low operation voltage,large on/off ratio and high uniformity is presented.Measurement results indicate that this neuron demonstrates self-oscillation behavior under applied voltages as low as 1 V.The oscillation frequency increases with the applied voltage pulse amplitude and decreases with the load resistance.It can then be used to evaluate the resistive random-access memory(RRAM)synaptic weights accurately when the oscillation neuron is connected to the output of the RRAM crossbar array for neuromorphic computing.Meanwhile,simulation results show that a large RRAM crossbar array(>128×128)can be supported by our oscillation neuron owing to the high on/off ratio(>10^(8))of Ag NDs TS device.Moreover,the high uniformity of the Ag NDs TS device helps improve the distribution of the output frequency and suppress the degradation of neural network recognition accuracy(<1%).Therefore,the developed oscillation neuron based on the Ag NDs TS device shows great potential for future neuromorphic computing applications.
文摘In the past few decades,the Moore’s Law has been the revolutionary force for our integrated circuit(IC)industry.However,the tremendous challenges faced in continuous transistor physical down-scaling and the unprecedented demands for computing and storage capabilities require our urgent search for strategies and solutions to integrate diverse materials,devices,circuits,and architectures in a 3D vertically stacked manner so that they can orchestrate in the most effective way to provide significantly enhanced functionalities as well as superior speed,energy,bandwidth,form fact,and cost.
文摘Traditional charge-based memories,such as dynamic random-access memory(DRAM)and flash,are approaching their scaling limits.A variety of resistance-based memories,such as phase-change memory(PCM),magnetic random-access memory(MRAM)and resistive random-access memory(RRAM),have been long considered for emerging memory applications thanks to their non-volatility,fast speed,low power,and compact size for potentially high-density integration.
基金Supported by National Natural Science Foundation of China(81860771)Science and Technology Project of the Education Department of Jiangxi Province(GJJ201219,GJJ170722)+2 种基金Students Innovative Entrepreneurial Training Plan Program of Jiangxi University of Traditional Chinese Medicine(202010412006,202010412143)Program of Key Discipline Training of Young Teachers of Jiangxi University of Traditional Chinese Medicine(2017jzzdxk001)Scientific Research Foundation of Jiangxi University of Traditional Chinese Medicine(2018WBZR011).
文摘[Objectives]To explore the effects of injection of oregano oil submicron emulsion on lipopolysaccharide-induced pneumonia in rats.[Methods]Rats induced by lipopolysaccharide were used as animal models of acute pneumonia.The experiment was divided into blank group,model group,administration group and positive drug control group.The morphology of lung tissue and the changes of cells and inflammatory factors in each group were observed,and the anti-inflammatory effects of injection of oregano oil submicron emulsion.[Results]The injection of oregano oil submicron emulsion can improve the pathological injury of rat lung tissue,inhibit the release of IL-6,IL-10,TNF-αcytokines induced by lipopolysaccharide,and significantly reduce the value of CRP.[Conclusions]Oregano oil submicron emulsion has a certain therapeutic effect on lipopolysaccharide-induced pneumonia in rats,and its mechanism may be related to reducing the release of cytoinflammatory factor IL-6,IL-10,and TNF-αand alleviating the injury to tissues and organs.
基金This work was supported by the National Natural Science Foundation of China(61622401,61851402,and 61734003)National Key Research and Development Program(2017YFB0405600)+1 种基金Shanghai Education Development Foundation and Shanghai Municipal Education Commission Shuguang Program(18SG01)P.Z.also acknowledges support from Shanghai Municipal Science and Technology Commission(grant no.18JC1410300).
文摘Memory cells have always been an important element of information technology.With emerging technologies like big data and cloud computing,the scale and complexity of data storage has reached an unprecedented peak with a much higher requirement for memory technology.As is well known,better data storage is mostly achieved by miniaturization.However,as the size of the memory device is reduced,a series of problems,such as drain gate-induced leakage,greatly hinder the performance of memory units.To meet the increasing demands of information technology,novel and high-performance memory is urgently needed.Fortunately,emerging memory technologies are expected to improve memory performance and drive the information revolution.This review will focus on the progress of several emerging memory technologies,including two-dimensional material-based memories,resistance random access memory(RRAM),magnetic random access memory(MRAM),and phasechange random access memory(PCRAM).Advantages,mechanisms,and applications of these diverse memory technologies will be discussed in this review.
基金Beijing Municipal Science and Technology Commission(No.Z181100003118014)National Natural Science Foundation of China(No.61722209)Tsinghua University Initiative Scientific Research Program.
文摘Training an artificial neural network with backpropagation algorithms to perform advanced machine learning tasks requires an extensive computational process.This paper proposes to implement the backpropagation algorithm optically for in situ training of both linear and nonlinear diffractive optical neural networlks,which enables the acceleration of training speed and improvement in energy efficiency on core computing modules.We demonstrate that the gradient of a loss function with respect to the weights of diffractive layers can be accurately calculated by measuring the forward and backward propagated optical fields based on light reciprocity and phase conjunction principles.The diffractive modulation weights are updated by programming a high-speed spatial light modulator to minimize the error between prediction and target output and perform inference tasks at the speed of light.We numerically validate the effectiveness of our approach on simulated networks for various applications.The proposed in situ optical learning architecture achieves accuracy comparable to in silico training with an electronic computer on the tasks of object dlassification and matrix-vector multiplication,which further allows the diffractive optical neural network to adapt to system imperfections.Also,the self-adaptive property of our approach facilitates the novel application of the network for all-optical imaging through scattering media.The proposed approach paves the way for robust implementation of large-scale difractive neural networks to perform distinctive tasks all-optically.
基金supported in part by the National Science and Technology Major Project of China(No.2017ZX02315001-005)the National Natural Science Foundation of China(Nos.91964104 and 61974081)。
文摘The rapid growth of the Internet of Things(IoTs)has resulted in an explosive increase in data,and thus has raised new challenges for data processing units.Edge computing,which settles signal processing and computing tasks at the edge of networks rather than uploading data to the cloud,can reduce the amount of data for transmission and is a promising solution to address the challenges.One of the potential candidates for edge computing is a memristor,an emerging nonvolatile memory device that has the capability of in-memory computing.In this article,from the perspective of edge computing,we review recent progress on memristor-based signal processing methods,especially on the aspects of signal preprocessing and feature extraction.Then,we describe memristor-based signal classification and regression,and end-to-end signal processing.In all these applications,memristors serve as critical accelerators to greatly improve the overall system performance,such as power efficiency and processing speed.Finally,we discuss existing challenges and future outlooks for memristor-based signal processing systems.
基金the financial supports from the National Natural Science Foundation of China(51872036,51773025)Dalian Science and Technology Innovation Fund(2018J12GX033)National Key R&D Program of China(2017YFB0405604)
基金This work was supported by the National Key R&D Program of China(No.2017YFB0405604)the Natural Science Foundation of China(No.51502019).
文摘En ergy storage devices with flexible form factor have become critical components of wearable electr onic systems.In spired by methods of monolithic integration in the microelectronics fabrication process,we propose a planar flexible full-solid-state lithium-ion battery(FSLB)architecture and a layer-by-layer stencil printing assembly method for fabricating batteries on polyethylene terephthalate(PET)substrate.FSLBs use quasi-solid electrolyte based on LiTFSI and ultraviolet(UV)-curable ethoxylated trimethylolpropane triacrylate(ETPTA)polymeric matrix in combination with Li4Ti50i2(LTO)/LiFePO4(LFP)-based electrodes.Excellent mechanical flexibility(<10 mm bending radius)can be achieved.The electrochemical characteristics of electrolyte,including ion conductivity,physical stability during room-temperature and tender assembly processes,are promising.A complete thin film-shape FSLB demonstrated working operation both under planar and bending conditions.The unique architecture and assembly processes open new ways for planar flexible devices to be integrated with flexible energy devices.
基金This work was financially supported by the National Key R&D Program of China(Nos.2019YFB2205100 and 2021ZD0201201)the National Natural Science Foundation of China(Grant Nos.92064012 and 61874164).
文摘With the rapid growth of computer science and big data,the traditional von Neumann architecture suffers the aggravating data communication costs due to the separated structure of the processing units and memories.Memristive in-memory computing paradigm is considered as a prominent candidate to address these issues,and plentiful applications have been demonstrated and verified.These applications can be broadly categorized into two major types:soft computing that can tolerant uncertain and imprecise results,and hard computing that emphasizes explicit and precise numerical results for each task,leading to different requirements on the computational accuracies and the corresponding hardware solutions.In this review,we conduct a thorough survey of the recent advances of memristive in-memory computing applications,both on the soft computing type that focuses on artificial neural networks and other machine learning algorithms,and the hard computing type that includes scientific computing and digital image processing.At the end of the review,we discuss the remaining challenges and future opportunities of memristive in-memory computing in the incoming Artificial Intelligence of Things era.
基金the National Natural Science Foundation of China(No.51972192).
文摘All memristor neuromorphic networks have great potential and advantage in both technology and computational protocols for artificial intelligence.It is crucial to find suitable elementary units for both performing featured neuromorphic functions and fabrication in large scale.Here a simple memristive structure,Nb/HfOx/Pd,is proposed for this goal.Its two resistive switching mechanisms,Mott transition of NbO2 and oxygen vacancy(Vo)migration,can be controlled by modulating external bias directions.Negative bias activates reversible phase transition and restrains Vo filament formation to allow the memristor to mimic the firing action potential.Positive bias activates Vo filament formation and restrains the other to allow the memristor to mimic synaptic plasticity and learning protocols.The system can respond adaptively to naturally generated action potentials and modified synaptic signals from the same memristive structure.In addition,some special features related to signal encoding and recognition are discovered when the system is settled according to chaos circuit theory.Our study provides a novel approach for designing elementary units for neuromorphic computations.
基金the National Natural Science Foundation of China(Grants No.92064001,61851404,and 61874169)the IoT Intelligent Microsystem Center of Tsinghua University-China Mobile Joint Research Institute.
文摘Mathematical morphology operations are widely used in image processing such as defect analysis in semiconductor manufacturing and medical image analysis.These data-intensive applications have high requirements during hardware implementation that are challenging for conventional hardware platforms such as central processing units(CPUs)and graphics processing units(GPUs).Computation-in-memory(CIM)provides a possible solution for highly efficient morphology operations.In this study,we demonstrate the application of morphology operation with a novel memristor-based auto-detection architecture and demonstrate non-neuromoq)hic computation on a multi-array-based memristor system.Pixel-by-pixel logic computations with low parallelism are converted to parallel operations using memristors.Moreover,hardware-implemented computer-integrated manufacturing was used to experimentally demonstrate typical defect detection tasks in integrated circuit(IC)manufacturing and medical image analysis.In addition,we developed a new implementation scheme employing a four-layer network to realize small-object detection with high parallelism.The system benchmark based on the hardware measurement results showed significant improvement in the energy efficiency by approximately 358 times and 32 times more than when a CPU and GPU were employed,respectively,exhibiting the advantage of the proposed memristor-based morphology operation.