检索结果-内蒙古大学图书馆

image encoded time series classification of small datasets: an innovative architecture using deep learning ensembles

引用

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS 2025年 1-16页

作者： Indrasiri, Pubudu L. Kashyap, Bipasha Pathirana, Pubudu N. Deakin Univ Sch Engn 75Pigdons Rd Waurn Ponds Vic 3216 Australia

Convolutional neural networks (CNNs) are often favored for their strong learning abilities in tackling automatic intelligent models. The classification of time series data streams spans across many applications of intelligent systems. However, the scarcity of effective Machine Learning architectures to handle limited time-series data adversely affects the realization of some crucial applications. In particular, healthcare-related applications are inherently concerned with limited time series datasets. Indeed, building effective artificial intelligence (AI) models for rare diseases using conventional techniques can pose a significant challenge. Utilizing recent advances in deep learning and signal processing techniques, this study introduces a new ensemble deep learning (DL) approach for time series categorization in the presence of limited datasets. Physiological data, such as ECG and voice, are used to demonstrate the functionality of the proposed DL architecture with data obtained from IoT and non-IoT devices. The proposed framework comprises a self-designed deep CNN-LSTM along with ResNet50 and MobileNet transfer learning approaches. The CNN-LSTM architecture includes an enhanced squeeze and excitation block that improves overall *** architecture processes time series data transformed into a 3-Channel image structure via improved recurrence plot (RP), Gramian angular field (GAF), and fuzzy recurrence plot (FRP) methods. The proposed model demonstrated superior classification accuracy on the ECG5000 and TESS datasets compared to other state-of-the-art techniques, validating its efficacy for binary and multiclass classification.

关键词： Limited datasets Healthcare time series Squeeze and excitation Ensemble learning

来源：评论

学校读者我要写书评

暂无评论

Enhanced Multimodal Conversational AI Using Speech and image Integration 5th

Enhanced Multimodal Conversational AI Using Speech and Image...

引用

5th International Conference on Computing, Communications, and Cyber Security, IC4S 2023

作者： Dixit, Deepali Tyagi, Mihul Kandoi, Raunak Singh, Raghuraj School of Engineering and Technology Sharda University Greater Noida201310 India

ISBN: (纸本)9789819773701

The development of conversational artificial intelligence (AI) is examined in this research paper, with a focus on how speech and image recognition technologies can be combined to transform and interact with systems. The emphasis is on cutting-edge developments in multimodal conversational AI, particularly the smooth fusion of sophisticated image and speech processing methods to improve user experiences. The study addresses the applications, difficulties and potential applications of multimodal conversational AI by conducting a thorough review of the existing literature and case studies. By emphasizing the significance of enhancing voice recognition algorithms and utilizing advanced image processing techniques, the study promotes contextual understanding and customization. We have developed an innovative multimodal conversational AI system that integrates speech, text and image processing capabilities for seamless human–computer interactions. This study presents an improved multimodal conversational AI system that integrates several techniques, such as Google Text-to-Speech (gtts), speech recognition, sound playback and Convolutional neural networks (CNN). Our model architecture combines RNNs for sequential data analysis, CNNs for image processing and attention mechanisms for context extraction. Trained on diverse datasets, our model is optimized for accuracy and robustness. Evaluation against existing models demonstrates superior performance in terms of accuracy, speed and resource efficiency. Our study highlights how improved multimodal conversational AI can advance the field of human–computer interaction. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Optoelectronic Synapse Enabled by Defect Engineering of Tellurene for Neuromorphic Computing

IEEE ELECTRON DEVICE LETTERS

引用

IEEE ELECTRON DEVICE LETTERS 2025年第1期46卷 68-71页

作者： Guo, Junxiong Huang, Junyan Gu, Shuyi Lin, Lin Zhang, Yafei Wang, Xiang Liu, Yu Gong, Tianxun Lin, Yuan Yu, Bin Huang, Wen Zhang, Xiaosheng Univ Elect Sci & Technol China Sch Integrated Circuit Sci & Engn Chengdu 611731 Peoples R China Sch Elect Informat & Elect Engn Chengdu 610106 Peoples R China Nanjing Univ Posts & Telecommun Coll Integrated Circuit Sci & Engn Nanjing 210023 Peoples R China Univ Elect Sci & Technol China Sch Mat & Energy Chengdu 610054 Peoples R China Zhejiang Univ ZJU Hangzhou Global Sci & Technol Innovat Ctr Hangzhou 311215 Peoples R China

Emerging optoelectronic synapses hold immense potential for advancing neuromorphic computing systems. However, achieving precise control over selective responses in optoelectronic memory and clarifying tunable synaptic weights has remained challenging. This study reports an optoelectronic synapse utilizing oxygen plasma-assisted defect engineering in tellurene for artificial neural networks. Through DFT calculations and experimental analyses, we demonstrate that tellurene conductance can be modulated by controlling plasma-defined defect engineering, allowing a transition from short-term to long-term synaptic plasticity, largely determined by intrinsic large-lattice-relaxation effects. Our artificial synapses exhibit high linearity, a broad dynamic range, and tunable synaptic weights. Additionally, our optoelectronic synapses display selective sensitivity to multi-spectral light and achieve a pattern recognition accuracy of up to 96.7% across five typical datasets, surpassing even the ideal synapse. These tunable spectral responses, combined with high-performance neuromorphic applications using spike coding, establish a foundation for developments in brain-inspired machine learning, robotics, and real-time data processing.

关键词： Optoelectronic synapse defect engineering image recognition neural network Optoelectronic synapse defect engineering image recognition neural network

来源：评论

学校读者我要写书评

暂无评论

Global-Local Multigranularity Transformer for Hyperspectral image Classification

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2025年 18卷 112-131页

作者： Meng, Zhe Yan, Qian Zhao, Feng Chen, Gaige Hua, Wenqiang Liang, Miaomiao Xian Univ Posts & Telecommun Sch Commun & Informat Engn Xian 710121 Peoples R China Xian Univ Posts & Telecommun Sch Artificial Intelligence Xian 710121 Peoples R China Xian Univ Posts & Telecommun Sch Comp Sci & Technol Xian 710121 Peoples R China Jiangxi Univ Sci & Technol Sch Informat Engn Jiangxi Prov Key Lab Multidimens Intelligent Perce Ganzhou 341000 Peoples R China

Hyperspectral image (HSI) classification is a challenging task in remote sensing applications, aiming to determine the category of each pixel by utilizing rich spectral and spatial information in HSI. Convolutional neural networks (CNNs) have been effective in processing HSI data by extracting local features, but they are deficient in capturing global contextual information. Recently, transformer has become proficient in attending to global information due to their self-attention mechanisms, yet they may fall short in capturing multiscale features of HSI. To address these limitations, a global-local multigranularity transformer (GLMGT) network is proposed for HSI classification. The GLMGT combines CNN with the transformer to comprehensively capture multigranularity spectral and spatial features across global and local scales. Specifically, we introduce a multigranularity spatial feature extraction block to extensively extract spatial information at different granularities, including multiscale local spatial features and global spatial features. In addition, we introduce a multigranularity spectral feature extraction block to fully leverage spectral information across different granularities. The validity of the proposed method is demonstrated through experimental validation using seven publicly available datasets, which include two Chinese satellite hyperspectral datasets (ZY1-02D Huanghekou and GF-5 Yancheng) and one UAV-based hyperspectral dataset.

关键词： Feature extraction Transformers Convolution Convolutional neural networks Data mining Three-dimensional displays Hyperspectral imaging Kernel Spectral analysis Computer vision CNN multigranularity transformer hyperspectral image (HSI) classification

来源：评论

学校读者我要写书评

暂无评论

Learning Truly Monotone Operators with applications to Nonlinear Inverse Problems

引用

SIAM JOURNAL ON IMAGING SCIENCES 2025年第1期18卷 735-764页

作者： Belkouchi, Younes Pesquet, Jean-Christophe Repetti, Audrey Talbot, Hugues Univ Paris Saclay CentraleSupelec Inria F-91190 Gif Sur Yvette France Heriot Watt Univ Sch Math & Comp Sci Edinburgh EH14 4AS Scotland Heriot Watt Univ Sch Engn & Phys Sci Edinburgh EH14 4AS Scotland Bayes Ctr Maxwell Inst Math Sci Edinburgh EH9 3FD Scotland

This article introduces a novel approach to learning monotone neural networks (NNs) through a newly defined penalization loss. The proposed method is particularly effective in solving classes of variational problems, specifically monotone inclusion problems, commonly encountered in image processing tasks. The forward-backward-forward (FBF) algorithm is employed to address these problems, offering a solution even when the Lipschitz constant of the NN is unknown. Notably, the FBF algorithm provides convergence guarantees under the condition that the learned operator is monotone. Building on plug-and-play methodologies, our objective is to apply these newly learned operators to solving nonlinear inverse problems. To achieve this, we initially formulate the problem as a variational inclusion problem. Subsequently, we train a monotone NN to approximate an operator that may not inherently be monotone. Leveraging the FBF algorithm, we then show simulation examples where the nonlinear inverse problem is successfully solved.

关键词： monotone operator optimization inverse problem deep learning forward-backward-forward plug and play (PnP)

来源：评论

学校读者我要写书评

暂无评论

Deep Residual Learning for Fruits of Ceremonial Plants Recognition 2

Deep Residual Learning for Fruits of Ceremonial Plants Recog...

引用

2nd International Conference on Data, Information and Computing Science, CDICS 2024

作者： Negara, Ida Bagus Kade Dwi Suta Putra, I. Ketut Gede Darma Sudarma, Made Sukarsa, I. Made Udayana University Denpasar Indonesia

ISBN: (纸本)9781643685762

Fruit detection refers to a method within image processing and computer vision that focuses on automatically recognizing and distinguishing different types of fruits using advanced algorithms and techniques. Fruits are essential and widely used in ceremonial offerings (Banten). They are pivotal in ensuring the completeness of ceremonial offerings. The main goal of fruit detection is to recognize fruits in image form, making it applicable in various applications, including inventory management, automatic classification in the agricultural industry, and medical applications for monitoring dietary patterns. Commonly used methods include artificial neural networks (deep learning), digital image processing, and feature extraction techniques to distinguish between shapes, colours, textures, and other visual features of different types of fruits. In this research, the deep learning algorithm CNN (Convolutional neural Network) is employed. This technique, a deep learning algorithm, excels in accurate image recognition and classification. A total of 500 fruit images are used for five types of ceremonial fruits commonly used in religious activities, namely coconut, banana, candlenut, nutmeg, and areca nut. The residual learning method RESNET152 is used as the CNN algorithm architecture. According to the test findings, the highest accuracy rate achieved was 93% in correctly identifying ceremonial fruit images. © 2025 The Authors.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

TensorCIM: Digital Computing-in-Memory Tensor Processor With Multichip-Module-Based Architecture for Beyond-NN Acceleration

引用

IEEE JOURNAL OF SOLID-STATE CIRCUITS 2025年第2期60卷 734-747页

作者： Wang, Yiqi Wu, Zihan Wu, Weiwei Liu, Leibo Hu, Yang Wei, Shaojun Tu, Fengbin Yin, Shouyi Tsinghua Univ Beijing Innovat Ctr Future Chip Sch Integrated Circuits Beijing 100084 Peoples R China Tsinghua Univ Beijing Natl Res Ctr Informat Sci & Technol Beijing 100084 Peoples R China Hong Kong Univ Sci & Technol Dept Elect & Comp Engn Hong Kong Peoples R China

While neural networks (NNs) have achieved great results in various intelligent tasks like image classification and speech recognition, real-world scenarios have more applications beyond just NN processing like graph convolutional network (GCN) and deep-learning recommendation model (DLRM), which typically consist of sparse gathering (SpG) and sparse algebra (SpA). Their large application size leads to substantial data movement. Although the fusion of digital computing-in-memory (CIM) and multichip-module (MCM) can reduce data movement efficiently and scale out CIM's capacity in a high-yield solution, the MCM-CIM system raises new challenges for beyond-NN acceleration: SpG involves repeated off-chip DRAM access, interchiplet access, and redundant reduction operations;SpA suffers from inter-CIM workload imbalance and intra-CIM under-utilization. Thus, we design TensorCIM as the CIM processor chiplet with three corresponding features: 1) the redundancy-eliminated gathering manager (REGM) dynamically maintains frequently accessed features and reduction results in the CIM to eliminate redundant accesses and reductions;2) the equal operation-based CIM initializer (EOCI) calculates effective multiply-accumulation (MAC) operations and initializes CIM macros with a balanced inter-CIM workload at the subarray level;and 3) the input-lookahead CIM (ILA-CIM) architecture looks ahead at future inputs to fully utilize CIM logic. The fabricated MCM-CIM system consumes only 3.7 nJ/Gather for the GCN model, achieving 8.3-TFLOPS/W algebra efficiency at FP32.

关键词： In-memory computing Tensors Common Information Model (computing) Computer architecture artificial neural networks Random access memory Vectors Computing-in-memory (CIM) floating-point graph neural network (NN) multichip-module (MCM) recommendation system sparsity

来源：评论

学校读者我要写书评

暂无评论

image-Based Air Quality Estimation Using Convolutional neural Network Optimized by Genetic Algorithms: A Multi-Dataset Approach

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND applications 2025年第3期16卷 1177-1185页

作者： Khan, Arshad Ali Majid, Mazlina Abdul Dandoush, Abdulhalim Univ Malaysia Pahang Al Sultan Abdullah Fac Comp Pekan Malaysia Ctr Artificial Intelligence & Data Sci Gambang Malaysia Univ Doha Sci & Technol UDST Doha Qatar

Air pollution poses significant threats to human health and the environment, making effective monitoring increasingly essential. Traditional methods using fixed monitoring stations have challenges related to high costs and limited coverage. This paper proposes a new approach using convolutional neural networks with genetic algorithms for estimating air quality directly from images. The convolutional neural network is optimized using genetic algorithms, which dynamically tune hyper-parameters such as learning rate, batch size, and momentum to improve performance and generalizability across diverse environmental conditions. Our approach improves performance and reduces the risk of overfitting, thus ensuring balanced and robust results. To mitigate overfitting, we implemented dropout layers, batch normalization, and early stopping, significantly enhancing the model's generalization capability. Specifically, three different open-access datasets were combined into a single training dataset, capturing extensive temporal, spatial, and environmental variability. Extensive testing of the model performance was conducted with a broad set of metrics, including precision, recall, and F1 score. The results demonstrate that our model not only achieves high accuracy but also maintains well-balanced performance across all metrics, ensuring robust classification of different air quality levels. For instance, the model achieved a precision of 0.97, a recall of 0.97, and an overall accuracy of 0.9544 percent, outperforming baseline methods significantly in all metrics. These improvements underscore the effectiveness of Genetic Algorithms in optimizing the model.

关键词： Convolutional neural network Genetic Algorithm air quality estimation image processing

来源：评论

学校读者我要写书评

暂无评论

An end-to-end deep convolutional neural network-based data-driven fusion framework for identification of human induced pluripotent stem cell-derived endothelial cells in photomicrographs

引用

ENGINEERING applications OF artificial INTELLIGENCE 2025年 139卷

作者： Iqbal, Imran Ullah, Imran Peng, Tingying Wang, Weiwei Ma, Nan Inst Act Polymers Helmholtz Zentrum Hereon Dept PLR D-14513 Teltow Germany German Res Ctr Environm Hlth Helmholtz Munich D-85764 Munich Germany

Deep learning is a very powerful analytic tool to recognize the patterns in data to make appropriate predictions. It has tremendous potential in data analyses, particularly for cell biology domain, caused by the growing scale and inherent complexity of biological data. The core purpose of this research work is to design, implement, and calibrate an efficient deep convolutional neural network (DCNN) architecture in the context of binary-class classification problem. This diversified network is developed to precisely identify human induced pluripotent stem cell-derived endothelial cells (hiPSC-derived EC) based on photomicrograph. The proposed architecture is cerebrally developed with numerous convolutional modules, multiple kernel sizes, various pooling layers, activation functions and strides, nevertheless fewer trainable parameters to strengthen the network and enhance its performance. The proposed feature fusion framework is compared with the classifier fusion approach in terms of Matthews's correlation coefficient (MCC), training time, inference time, number of layers, number of parameters, graphics processing unit (GPU) memory utilization, and floating-point operations (FLOPS). Specifically, it achieves 94.6% sensitivity, 94.5% specificity, and 94.7% precision. Induced pluripotent stem cell (iPS) dataset is also introduced in this research work that has 16278 images which are labelled by three independent and experienced human experts of cell biology domain to facilitate future research. Experimental results show that the proposed framework offers an innovative and attainable algorithm for accelerating and systematizing the classification task along with saving time and effort.

关键词： Computer vision Deep convolutional neural networks Endothelial cells Human induced pluripotent stem cells image processing Information fusion Machine learning Photomicrograph

来源：评论

学校读者我要写书评

暂无评论

Oxide Phototransistor Array With Multiply-and-Accumulation Functions for In-Sensor image processing

引用

IEEE Journal of the Electron Devices Society 2025年 13卷 378-382页

作者： Wang, Saisai Jing, Xiaotao Zhang, Wanlin Wang, Rui Wang, Hong Huang, Qi Westlake Institute for Optoelectronics Zhejiang Key Laboratory of 3D Micro/Nano Fabrication and Characterization Hangzhou311421 China Xidian University State Key Laboratory of Wide Band Gap Semiconductor Devices and Integrated Technology School of Microelectronics Xi’an710071 China

Advanced in-sensor computing paradigm has gradually become a research hotspot in this IOT era of sensor data proliferation. However, most existing in-sensor computing devices are plagued by a complex structure, and the uniformity of the sensor array is difficult to ensure. Moreover, photoconductive devices are incapable of achieving on-chip current summation because the regulation of photoresponsivity usually leads to inconsistent dark current, thereby impeding the practical implementation of artificial neural networks (ANNs) on the sensor array. In this work, we developed a 3 × 3 ultraviolet (UV) image sensor array based on solution-processed indium oxide (In2O3) phototransistors. The devices exhibit tunable responsivity and unified dark current under negative gate voltage (Vgs), enabling the direct execution of multiply-and-accumulation (MAC) operations. Consequently, two key applications of ANN were successfully demonstrated: image convolution and classification. © 2013 IEEE.

关键词： Phototransistors

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：