检索结果-内蒙古大学图书馆

Novel Dictionary Pair Learning through Locality-Constrained Profiles for the Identification of Chili Pepper Varieties via E-noses

引用

IEEE Sensors Journal 2025年第10期25卷 17530-17540页

作者： Chen, Liang Peng, Guihua Zhang, Changfu Wang, Xueya Chen, Ju Yin, Yong Yan, Jia Southwest University College of Artificial Intelligence Chongqing400715 China Guizhou Academy of Agricultural Sciences Chili Pepper Research Institute Guiyang550006 China Southwest University College of Artificial Intelligence Beibei Chongqing400715 China Chongqing Key Laboratory of Brain-inspired Computing and Intelligent Chips Chongqing400715 China

In modern society, to make it easier for customers to assess product quality and regularize distribution and supply, almost all crops are sorted and labeled. This sorting and labelling process also facilitates packaging and transportation of products and benefits farmers. As an emerging technology, electronic noses (E-noses) can be applied to food quality control. In this study, we proposed a novel dictionary pair learning model named dictionary pair learning based on locality constraint of profiles (DPL-LCP), which achieved better performance in the identification of chili pepper varieties using E-noses. This model incorporates profile vectors to define a profile locality constraint term and integrates it with a dictionary reconstruction error term into a unified dictionary learning model to fully learn the characteristics of samples from various aspects. This approach not only reduces the similarity between dictionary atoms but also avoids the problem of weakening the connections between characteristics of different classes of samples caused by using the block diagonal constraint. In addition, the model utilizes the l2,1 norm to regularize the analysis dictionary matrix, enhancing the sparsity of features and thereby enhancing the flexibility and robustness of the model. The experimental results show that the proposed model can achieve better classification results and higher stability than other advanced dictionary learning models can. With this method, it is possible to reliably identify 13 varieties of chili peppers based on flavor assessment. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

MA-MNN: Multi-flow attentive memristive neural network for multi-task image restoration

引用

Signal Processing: Image Communication 2025年 138卷

作者： He, Peng Zhang, Lin Yang, Yu Zhou, Yue Duan, Shukai Hu, Xiaofang College of Artificial Intelligence Southwest University Chongqing400715 China Brain-inspired Computing & Intelligent Control of Chongqing Key Lab Chongqing400715 China Ministry of Education China

Images taken in rainy, hazy, and low-light environments severely hinder the performance of outdoor computer vision systems. Most data-driven image restoration methods are task-specific and computationally intensive, whereas the capture and processing of degraded images occur largely in end-side devices with limited computing resources. Motivated by addressing the above issues, a novel software and hardware co-designed image restoration method named multi-flow attentive memristive neural network (MA-MNN) is proposed in this paper, which combines a deep learning algorithm and the nanoscale device memristor. The multi-level complementary spatial contextual information is exploited by the multi-flow aggregation block. The dense connection design is adopted to provide smooth transportation across units and alleviate the vanishing-gradient. The supervised calibration block is designed to facilitate achieving the dual-attention mechanism that helps the model identify and re-calibrate the transformed features. Besides, a hardware implementation scheme based on memristors is designed to provide low energy consumption solutions for embedded applications. Extensive experiments in image deraining, image dehazing and low-light image enhancement have shown that the proposed method is highly competitive with over 20 state-of-the-art methods. © 2025 Elsevier B.V.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

Improving E-Nose Performance: A Novel ELM-Based Dual-Level Joint Domain Adaptation Method for Sensor Drift Data

引用

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Wang, Zijian Zhang, Linxia Li, Zhe Duan, Shukai Yan, Jia Southwest University College of Artificial Intelligence Chongqing400715 China Southwest University Chongqing Key Laboratory of Brain-inspired Computing and Intelligent Chips College of Artificial Intelligence Chongqing400715 China

The theory and technology of electronic nose (E-nose) systems have been vigorously developed, and these systems have achieved success in many practical applications, such as medical diagnosis, food quality inspection, and environmental detection. However, the drift problem of a sensor array affects the industrialization and commercialization of E-nose systems. In this article, a novel ELM-based dual-level joint domain adaptation method (JDAELM) is proposed to effectively suppress drift and address the distribution discrepancy issue. Specifically, the proposed method implements joint domain adaptation (DA) at the feature level and label level. For source domain data without drift, the information of the data could be preserved as much as possible. Considering the inconsistent distribution caused by drift, the marginal and conditional distribution discrepancies are reduced to a minimum at the feature level to achieve domain alignment. To reduce the impact of pseudolabels on the model, we align the label space to achieve DA at the label level. By maximizing the Hilbert-Schmidt independence criterion, the relationship between the feature projection space and label projection space is strengthened in this model. The joint learning model is effectively solved by an efficient alternative optimization strategy. The average accuracy of the proposed method is 88.30% and 87.21% under long-term drift and short-term drift, respectively, and 96.41% on the instrument variation dataset, which is superior to that of other comparison methods in terms of accuracy. This proves that the JDAELM can be well adapted to long-term and short-term drift scenarios, and can effectively compensate for instrument variation drift caused by inherent differences. © 1963-2012 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A 3–8 decoder of terahertz metamaterials and its sensing application

引用

Diamond and Related Materials 2025年 156卷

作者： Zhu, Aijun Wei, Wenrui Hou, Weigang Cheng, Lei Hu, Cong Guangxi Key Laboratory of Automatic Detecting Technology and Instruments Guilin University of Electronic Technology Guilin541004 China Institute of Intelligent Communication and Network Security Chongqing University of Posts and Telecommunications Chongqing400000 China Jincheng Research Institute of Opto-mechatronics Industry Shanxi Jincheng048000 China Shanxi Key laboratory of Advanced semiconductor Optoelectronic Devices and Integrated Systems Shanxi Jincheng048000 China Guangxi Key Laboratory of Brain-inspired Computing and Intelligent Chips Guangxi Normal University Guilin541004 China

A 3–8 decoder based on terahertz metamaterial was proposed for the first time, which can achieve both 3–8 decoding function and dual band selectable 2–4 decoding function. The function of this device mainly relies on VO2 acting as a conductor or insulator. The finite integration in time domain (FITD) method was used to simulate and optimize the performance of the metamaterial absorber (MMA), and its working mechanism was analyzed through electric field distribution diagram and equivalent circuit model (ECM). When the device is used as a 3–8 decoder, the maximum modulation depth (MD) is 99.60 %, the minimum extinction ratio (ER) is 17.7 dB, the maximum insertion loss (IL) is 0.76 dB, and it has the property of being insensitive to the incident angle, with stable decoding ability. In addition, the device can also be used as a sensor with a refractive index sensitivity (S) of 0.55 THz/RIU, a quality factor (Q) of 24.714, and a FOM of 3.93RIU−1, demonstrating excellent sensing capabilities. This work provides new ideas for improving the decoding capability of terahertz devices and designing dynamic devices. © 2025

关键词： Refractive index

来源：评论

学校读者我要写书评

暂无评论

MindEye-OmniAssist: A Gaze-Driven LLM-Enhanced Assistive Robot System for Implicit Intention Recognition and Task Execution

arXiv

引用

arXiv 2025年

作者： Zhang, Zejia Yang, Bo Chen, Xinxing Shi, Weizhuang Wang, Haoyuan Luo, Wei Huang, Jian Hubei Key Laboratory of Brain-Inspired Intelligent Systems Huazhong University of Science and Technology Wuhan430074 China Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China State key laboratory of intelligent vehicle safety technology chongqing changan automobile co ltd Chongqing400023 China Science and technology innovation center China ship development and design centre Wuhan430060 China

A promising effective human-robot interaction in assistive robotic systems is gaze-based control. However, current gaze-based assistive systems mainly help users with basic grasping actions, offering limited support. Moreover, the restricted intent recognition capability constrains the assistive system’s ability to provide diverse assistance functions. In this paper, we propose an open implicit intention recognition framework powered by Large Language Model (LLM) and Vision Foundation Model (VFM), which can process gaze input and recognize user intents that are not confined to predefined or specific scenarios. Furthermore, we implement a gaze-driven LLM-enhanced assistive robot system (MindEye-OmniAssist) that recognizes user’s intentions through gaze and assists in completing task. To achieve this, the system utilizes open vocabulary object detector, intention recognition network and LLM to infer their full intentions. By integrating eye movement feedback and LLM, it generates action sequences to assist the user in completing tasks. Real-world experiments have been conducted for assistive tasks, and the system achieved an overall success rate of 41/55 across various undefined tasks. Preliminary results show that the proposed method holds the potential to provide a more user-friendly human-computer interaction interface and significantly enhance the versatility and effectiveness of assistive systems by supporting more complex and diverse task. Copyright © 2025, The Authors. All rights reserved.

关键词： Eye movements

来源：评论

学校读者我要写书评

暂无评论

Multi-bit nonvolatile memory device based on the reverse polarization under the channel of organic ferroelectric thin-film transistor

引用

Surface and Coatings Technology 2025年 509卷

作者： Chen, Qiusong Zhou, Yulan Sanchez, Juan E. Lin, Dong Zhu, Guodong Xiong, Zuhong School of Physics and Electronics Science & Guizhou Key Laboratory of Artificial Intelligence and Brain-inspired Computing Guizhou Education University Guiyang550018 China DEVSIM LLC PO Box 50096 AustinTX78763 United States Fujian Provincial Key Laboratory of Oceanic Information Perception and Intelligent Processing Jimei University Xiamen361021 China Department of Materials Science Fudan University No. 220 Handan Road Shanghai200433 China Chongqing Key Laboratory of Micro & Nano Structure Optoelectronics School of Physical Science and Technology Southwest University Chongqing400715 China

Ferroelectric field thin-film transistors (Fe-TFTs) are a promising candidate for multi-bit nonvolatile memory devices. Different theories and methods have been proposed to realize this kind of device. This paper introduces a novel operation procedure that utilizes reverse polarization under the organic ferroelectric thin-film transistors (OFe-TFTs) channel to achieve the multi-bit nonvolatile memory device. The potential difference between the source and the drain contact of Fe-TFT was utilized to implement a reverse distribution of the polarization under the channel of Fe-TFT. A self-designed technology computer-aided design platform of DEVSIM was used to exemplify our method of reverse polarization distribution in the bottom-gate top-contact Fe-TFTs. Then, an organic Fe-TFT using poly (vinylidene fluoride-trifluoroethylene) (PVDF-TrFE) as the dielectric and Poly [2,5-(2-octyldodecyl) -3, 6- diketopyrrolopyrrole -alt-5,5- (2,5-di (thien-2-yl) thieno [3, 2-b] thiophene)] (DPP-DTT) as the semiconductor layers was fabricated to realize a 2-bit memory device. The measured transfer characteristics and retention properties perform a good verification of our originality. © 2025 Elsevier B.V.

关键词： Ferroelectric thin films

来源：评论

学校读者我要写书评

暂无评论

Physical mechanisms and integration design of memristors

引用

Materials Today Nano 2025年 30卷

作者： Wang, Mengna Wang, Kun Sun, Bai Zhou, Guangdong Cao, Zelin Gao, Kaikai Ren, Fenggang Chen, Xiaoliang Li, Xiangming Shao, Jinyou Frontier Institute of Science and Technology and Interdisciplinary Research Center of Frontier Science and Technology Xi'an Jiaotong University Shaanxi Xi'an710049 China Micro-and Nano-technology Research Center State Key Laboratory for Manufacturing Systems Engineering Xi'an Jiaotong University Shaanxi Xi'an710049 China College of Artificial Intelligence Brain-inspired Computing & Intelligent Control of Chongqing Key Lab Southwest University Chongqing400715 China National Local Joint Engineering Research Center for Precision Surgery & Regenerative Medicine The First Affiliated Hospital of Xi'an Jiaotong University Shaanxi Xi'an710049 China

Memristors with threshold transition behavior and rich dynamics are ideal candidates for simulating biological pulse neurons and constructing efficient neuromorphic systems. Based on changes in material structure, charge distribution, molecular polarization and other physical states, new integrated memory-computing devices have been constructed and applied to the sensor-memory-computing integrated systems. In this review, we first discuss the classification of memristive materials from the perspective of device structure design and focus on introducing the working mechanisms of valence variation, polarization, and carrier transfer covered internally. Further, the latest progress of memristors with multi-physical mechanisms is comprehensively summarized, including the formation/fracture of conductive filaments, carrier capture/release, polarization/depolarization. In particular, it is discussed in detail the principles and applications of memristors with different working mechanisms for constructing neuromorphic systems. Finally, the potential challenges and opportunities that may exist in the development of memristors in the coming years are discussed in depth. © 2025 Elsevier Ltd

关键词： Molecular structure

来源：评论

学校读者我要写书评

暂无评论

KFF: K-Feature Fusion Token Merging for Vision Transformer

引用

Expert Systems with Applications 2025年

作者： Yu Yang Yue Zhou Xiaofang Hu Shukai Duan College of Artificial Intelligence Southwest University ChongQing 400715 China Chongqing Key Laboratory of Brain-inspired Computing and Intelligent Chips Southwest University ChongQing 400715 China

Recently, Vision Transformer (ViT) has achieved better performance than Convolutional Neural Networks (CNNs) in various vision applications. However, they are usually more computationally expensive than CNNs and face challenges in training and inference efficiency. Token merging is an effective and training-free way to reduce model complexity. However, since few tokens are exactly the same, prevalent similarity-based merging methods are challenging to avoid feature information loss and accuracy degradation. To address this issue, we propose a novel K-feature fusion token merging algorithm that significantly reduces the similarity metric error and token merging error with almost no accuracy loss. Specifically, we first reveal that similarity measurement errors and merging strategies have a significant impact on the performance of token merging algorithms, and the currently popular K-based similarity method will cause obvious feature shifts during the merging process. Based on this observation, we present a new feature-enhanced K-feature fusion token similarity calculation method. By combining the keys (K), which summarize the information contained in each token, and the more detailed intermediate features, the error of similarity measurement is greatly reduced. Then, we design a similarity-weighted average token merging algorithm to combine tokens that is faster and more accurate than ordinary average token merging. Extensive experiments show that our approach yields better model performance when reducing comparable computational effort and improving throughput without extra training. For example, for ViT-B on ImageNet, our method reduces 49.58% of tokens and improves throughput by 30% with only a 0.44% drop in accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Memristor-based multi-synaptic spiking neuron circuit for spiking neural network

引用

Chinese Physics B 2022年第4期31卷 225-233页

作者： Wenwu Jiang Jie Li Hongbo Liu Xicong Qian Yuan Ge Lidan Wang Shukai Duan College of Artificial Intelligence Southwest UniversityChongqing 400715China National&Local Joint Engineering Laboratory of Intelligent Transmission and Control Technology Chongqing 400715China Brain-inspired Computing and Intelligent Control of Chongqing Key Laboratory Chongqing 400715China Chongqing Brain Science Collaborative Innovation Center Chongqing 400715China

Spiking neural networks(SNNs) are widely used in many fields because they work closer to biological ***,due to its computational complexity,many SNNs implementations are limited to computer ***,this paper proposes a multi-synaptic circuit(MSC) based on memristor,which realizes the multi-synapse connection between neurons and the multi-delay transmission of pulse *** synapse circuit participates in the calculation of the network while transmitting the pulse signal,and completes the complex calculations on the software with ***,a new spiking neuron circuit based on the leaky integrate-and-fire(LIF) model is designed in this *** amplitude and width of the pulse emitted by the spiking neuron circuit can be adjusted as *** combination of spiking neuron circuit and MSC forms the multi-synaptic spiking neuron(MSSN).The MSSN was simulated in PSPICE and the expected result was obtained,which verified the feasibility of the ***,a small SNN was designed based on the mathematical model of *** the SNN is trained and optimized,it obtains a good accuracy in the classification of the IRIS-dataset,which verifies the practicability of the design in the network.

关键词： memristor multi-synaptic circuit spiking neuron spiking neural network(SNN)

来源：评论

学校读者我要写书评

暂无评论

A spintronic memristive circuit on the optimized RBF-MLP neural network

引用

Chinese Physics B 2022年第11期31卷 272-283页

作者： Yuan Ge Jie Li Wenwu Jiang Lidan Wang Shukai Duan School of Artificial Intelligence Southwest UniversityChongqing 400715China Chongqing Brain Science Collaborative Innovation Center Chongqing 400715China Brain-inspired Computing and Intelligent Control of Chongqing Key Laboratory Chongqing 400715China National&Local Joint Engineering Laboratory of Intelligent Transmission and Control Technology Chongqing 400715China

A radial basis function network(RBF)has excellent generalization ability and approximation accuracy when its parameters are set ***,when relying only on traditional methods,it is difficult to obtain optimal network parameters and construct a stable model as *** view of this,a novel radial basis neural network(RBF-MLP)is proposed in this *** connecting two networks to work cooperatively,the RBF’s parameters can be adjusted adaptively by the structure of the multi-layer perceptron(MLP)to realize the effect of the backpropagation updating ***,a genetic algorithm is used to optimize the network’s hidden layer to confirm the optimal neurons(basis function)number *** addition,a memristive circuit model is proposed to realize the neural network’s operation based on the characteristics of spin *** is verified that the network can adaptively construct a network model with outstanding robustness and can stably achieve 98.33%accuracy in the processing of the Modified National Institute of Standards and Technology(MNIST)dataset classification *** experimental results show that the method has considerable application value.

关键词： radial basis function network(RBF) genetic algorithm spintronic memristor memristive circuit

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：