检索结果-内蒙古大学图书馆

Personalized Recommendation of Teacher learning Behavior Using Expression Recognition Algorithms Integrated by deep learning and Neural Networks 4

Personalized Recommendation of Teacher Learning Behavior Usi...

引用

4th International Signal processing, Communications and Engineering Management Conference, ISPCEM 2024

作者： Si, Chen Fangna, Chen Wanbing, Ma Xiaolei, Wang Jiahua, Zhu Nini, Cheng School of Art and Technology Chengdu College of UESTC Sichuan Chengdu613000 China College of Education Linyi University Shandong Linyi276000 China School of Life Sciences Central China Normal University Hubei Wuhan430079 China Zhengzhou Fengyang Foreign Language School Henan Zhengzhou450000 China Center for International Education Philippine Christian University Manila1004 Philippines School of Life Sciences Linyi University Shandong Linyi276000 China

ISBN: (纸本)9798331528676

At present, online teacher training platforms represented by Chinese university catechism have become an important way for teachers' professional development. However, the course dropout rate of such platforms is very high. For this reason, this paper intends to utilize a combination of deep learning and augmented learning to monitor the teacher's facial expression in the process of online teaching in real time, so as to realize the analysis of the teacher's learning behavior and emotional state. This paper proposes to investigate a facial expression recognition method based on a combination of Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), using augmented learning strategies to train and optimize the proposed model to better understand the teacher's emotional response and learning needs. In this paper, we use machine learning and other methods to study teachers' emotional changes in the specific teaching process and conduct correlation analysis with them to find out the important factors affecting their learning effectiveness. When the teacher's emotion reaches a pleasant state, the learning completion rate is significantly increased by 20%, indicating that the emotional state has a significant positive impact on learning behavior and performance. Through the in-depth analysis of teachers' facial expressions and learning behaviors, we can optimize the appropriateness and engagement of the training content, as well as provide new research ideas and methods for educational technology research and promote the development of educational informatization. ©2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Low-Cost Deployment Case Study for Industrial IoT Data Collection and processing 4

A Low-Cost Deployment Case Study for Industrial IoT Data Col...

引用

4th International Conference on Intelligent Technology and Embedded Systems, ICITES 2024

作者： Zhang, Jiajun Hu, Biao China Agricultural University College of Engineering Beijing China

ISBN: (纸本)9798331506292

Despite the widespread implementation of SCADA systems in factories for centralized data management, their functionality is restricted to devices equipped with sensors. Manual readings are still prevalent for critical instruments like temperature and pressure gauges, lacking interfaces, leading to inefficiencies and errors. Upgrading production line sensors with IoT-enabled instruments is costly and disruptive. To tackle this, we propose a cost-effective data acquisition system that leverages edge acquisition devices and deep learning algorithms without altering the existing equipment and processes. Our solution involves installing new sensors or cameras in critical areas and employing image recognition algorithms alongside indirect measurement techniques. This approach enhances the real-time accuracy of data collection, significantly boosts the overall efficiency of the production process, and reduces operating costs. © 2024 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Robust occlusion-aware orbital angular momentum feature extraction via all-optical diffractive processing systems

引用

Optics Express 2025年第11期33卷 23053-23064页

作者： Li, Keyao Jia, Yuetian Gu, Min Fang, Xinyuan School of Artificial Intelligence Science and Technology University of Shanghai for Science and Technology Shanghai200093 China Institute of Photonic Chips University of Shanghai for Science and Technology Shanghai200093 China

As a brain-inspired optical computing architectures, diffractive optical neural networks (DONN) harness light’s wave nature for high-speed, energy efficient and parallel information processing, enabling applications such as image classification and wavefront shaping. However, conventional spatially encoded DONNs struggle with robustness in complex and unpredictable environments, where occlusions and distortions degrade processing accuracy. To address these challenges, we propose a robust all-optical feature extraction framework based on orbital angular momentum (OAM). This approach converts optical information into target OAM modes using a diffractive processing framework trained via deep learning, enabling stable and efficient information representation in the OAM domain. Unlike conventional DONNs, our method maintains high performance across diverse and irregular occlusions without requiring network retraining. This self-adaptive occlusion immune operates with zero additional training samples, effectively enhancing optical computing tasks under dynamic and uncertain conditions. By fully utilizing the helical wavefront and orthogonality of OAM, our approach improves the robustness and scalability of DONNs, demonstrating superior performance in challenging optical environments. Our work paves the way for next-generation optical computing systems that can operate reliably in unpredictable and occlusion-rich environment, unlocking what we believe to be new possibilities for robust, real-time processing in a variety of applications. © 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.

关键词： Angular momentum

来源：评论

学校读者我要写书评

暂无评论

Generative Vision Hybrid Model Enhancing realistic Synthetic Media Creation Through Advanced image processing Techniques 4

Generative Vision Hybrid Model Enhancing Realistic Synthetic...

引用

4th International Conference on Sentiment Analysis and deep learning, ICSADL 2025

作者： Thangamayan, S. Lavaraju, B. Adiyattil, Rini Mujesh, Pranav Hrudhya Nivedin, K.Sai Narasimhan, Lakshmi Saveetha School of Law Saveetha Institute of Medical and Technical Sciences Chennai77 India

ISBN: (纸本)9798331523923

In recent years, synthetic media generation has gained significant attention in areas such as entertainment, gaming, and visual content creation. The Generative Vision Hybrid Model leverages advanced generative architectures and image processing techniques to create highly realistic synthetic media. By integrating StyleGAN with a vision enhancement layer comprising Autoencoder-based Attention and Transformer modules, the model achieves exceptional quality in image synthesis. The dataset underwent robust preprocessing, including Wavelet Transform-based Noise Reduction and Adaptive Contrast Enhancement (ACE), ensuring clean and diverse inputs. The model achieved an Inception Score (IS) of 9.5, a Fréchet Inception Distance (FID) of 15.8, and a Structural Similarity Index (SSIM) of 0.94, outperforming existing benchmarks like StyleGAN2. Human evaluators rated the model's output with an average realism score of 4.7/5, highlighting its superior perceptual quality. Additionally, PSNR reached 32.6 dB, demonstrating high fidelity. The results validate the effectiveness of the hybrid model in producing photorealistic synthetic media with fine-grained detail and structural consistency. Future work can extend this approach to multi-modal media synthesis, real-time applications, and ethical watermarking systems to ensure responsible usage. © 2025 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Multilevel Multimodal Framework for Automatic Collateral Scoring in Brain Stroke

引用

IEEE ACCESS 2024年 12卷 33730-33748页

作者： Raj, Rishi Pruthviraja, Dayananda Gupta, Ayush Mathew, Jimson Kannath, Santhosh Kumar Prakash, Adity Rajan, Jeny Indian Inst Technol Patna Dept Comp Sci & Engn Patna 801106 India Manipal Acad Higher Educ Manipal Inst Technol Bengaluru Dept Informat Technol Manipal 576104 Karnataka India Sree Chitra Tirunal Inst Med Sci & Technol Dept Imaging Sci & Intervent Radiol Thiruvananthapuram 695011 Kerala India Banaras Hindu Univ Inst Med Sci Dept Radiodiag & Imaging Varanasi 221005 India Natl Inst Technol Karnataka Dept Comp Sci & Engn Surathkal 575025 India

In patients with ischemic brain stroke, collateral circulation plays a crucial role in selecting patients suitable for endovascular therapy. The presence of well-developed collaterals improves the patient's chances of recovery. In clinical practice, the presence of collaterals is diagnosed on a Computed Tomography Angiography scan. The radiologist grades it on the basis of subjective visual assessment, which is prone to interobserver and intraobserver variability. Computer-based methods of collateral assessment face the challenge of non-uniform scan volume, leading to manual selection of slices, meaning that the most imperative slices have to be manually selected by the radiologist. This paper proposes a multilevel multimodal hierarchical framework for automated collateral scoring. Specifically, we propose deploying a Convolutional Neural Network for image selection based on the visibility of collaterals and a multimodal model for comparing the occluded and contralateral sides of the brain for collateral scoring. We also generate a patient-level prediction by integrating automated machine learning in the proposed framework. While the proposed multimodal predictor contributes to Artificial Intelligence, the proposed end-to-end framework is an application in engineering. The proposed framework has been trained and tested on 116 patients, with five-fold cross-validation, achieving an accuracy of 91.17% for multi-class collateral scores and 94.118% for binary class collateral scores. The proposed multimodal predictor achieved a weighted F1 score of 0.86 and 0.95 on multi-class and binary-class collateral scores, respectively. The proposed framework is fast, efficient, and scalable for real-world deployments. Automated evaluation of collaterals with attention maps for explainability would complement radiologists' efforts. Code for the proposed framework is available

关键词： Biomedical imaging Stroke (medical condition) deep learning Feature extraction Task analysis Numerical models Manuals Medical expert systems Circulatory system Patient monitoring Surgery Convolutional neural networks Medical treatment biomedical image processing medical expert systems stroke (medical condition)

来源：评论

学校读者我要写书评

暂无评论

Priority-Based deep learning Inference On Edge Devices for Military Surveillance

Priority-Based Deep Learning Inference On Edge Devices for M...

引用

2025 International Conference on Computing for Sustainability and Intelligent Future, COMP-SIF 2025

作者： Tiwari, Utkarsh Tripathy, Aayushi Jeeban Ashwani, Siddhant Supriya, M. Bhaskaran, Sreebha Shinu, M.R. Amrita Vishwa Vidyapeetham Department of Cs&e Amrita School of Computing Bengaluru India

ISBN: (纸本)9798331538538

This work proposes a priority scheduling formula for Quick Task Execution in a military battlefield including;image segmentation, classification, and detection which are deep learning tasks that face challenges in edge AI due to their computational requirements. Comparisons using real-world simulation scenarios show the efficiency of the proposed formula over traditional scheduling techniques by proving that it can immediately allocate higher priority tasks to resource- limited endpoints on the edge of a network. In a way, the marked approach ensures optimized task execution within queues as well as task importance matched with the computational capacity, which reduces processing time and corresponding delays, and thus improves the general efficiency. This method is scalable and reliable for risk management in military surveillance and is also suitable for any field such as disaster response and smart city planning. © 2025 IEEE.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

6th IEEE International Conference on image processing, Applications and Systems, IPAS 2025 - Proceedings

6th IEEE International Conference on Image Processing, Appli...

引用

6th IEEE International Conference on image processing, Applications and Systems, IPAS 2025

ISBN: (纸本)9798331506520

The proceedings contain 86 papers. The topics discussed include: robust real-time monitoring of complex human activities using multi modal video analytics;a robust approach for classifying laparoscopic video distortions using ResNet-50;enhancing x-ray image classification through neural architecture;revolutionary MRI imaging for Alzheimer’s: cutting-edge GANs and vision transformer solutions;advanced deep learning strategies for breast cancer image analysis;identifying surgical instruments in pedagogical cataract surgery videos through an optimized aggregation network;enhancing auxiliary cancer classification task for multi-task breast ultrasound diagnosis network;and bioinspired computer vision for effective extended reality applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ABDF²-Net: an adaptive bi-directional features fusion network for real-time detection of threat object

引用

JOURNAL OF real-time image processing 2022年第4期19卷 739-749页

作者： Wei, Yiru Zhu, Zhiliang Yu, Hai Zhang, Wei Shenyang Univ Technol Sch Artificial Intelligence Shenyang 110870 Peoples R China Northeastern Univ Software Coll Shenyang 110819 Peoples R China

To achieve automatic detection of threat objects for X-ray baggage screening, we propose an adaptive bi-directional features fusion network (ABDF(2)-Net) to detect threat objects on X-ray images. In ABDF(2)-Net, an adaptive bi-directional feature fusion module (ABDF(2)M) is introduced to fuse the multi-scale features from two directions, and the adaptive function is used to control the features passing rate. Besides, an atrous convolutional pyramid pooling (ACPP) is employed to capture global contextual information, which can provide global semantic guidance for multi-scale features. Finally, the fused multi-scale features are used to predict the final detection results through prediction modules. Experiments on the GDXray database demonstrate the effectiveness and superiority of our proposed method against the other four object detection methods.

关键词： X-ray baggage screening Detection model of threat objects Anchor-free deep learning

来源：评论

学校读者我要写书评

暂无评论

MFP-YOLO: a multi-scale feature perception network for CT bone metastasis detection

引用

MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING 2024年第3期63卷 629-640页

作者： Lu, Wenrui Zhang, Wei Liu, Yanyan Xu, Lingyun Fan, Yimeng Meng, Zhaowei Jia, Qiang Tianjin Univ Sch Microelect Tianjin 300072 Peoples R China Nankai Univ Coll Elect Informat & Opt Engn Tianjin 300072 Peoples R China Tianjin Med Univ Gen Hosp Dept Nucl Med Tianjin 300052 Peoples R China

Bone metastasis is one of the most common forms of metastasis in the late stages of malignancy. The early detection of bone metastases can help clinicians develop appropriate treatment plans. CT images are essential for diagnosing and assessing bone metastases in clinical practice. However, early bone metastasis lesions occupy a small part of the image and display variable sizes as the condition progresses, which adds complexity to the detection. To improve diagnostic efficiency, this paper proposes a novel algorithm-MFP-YOLO. Building on the YOLOv5 algorithm, this approach introduces a feature extraction module capable of capturing global information and designs a new content-aware feature pyramid structure to improve the network's capability in processing lesions of varying sizes. Moreover, this paper innovatively applies a transformer-structure decoder to bone metastasis detection. A dataset comprising 3921 CT images was created specifically for this task. The proposed method outperforms the baseline model with a 5.5% increase in precision and a 7.7% boost in recall. The experimental results indicate that this method can meet the needs of bone metastasis detection tasks in real scenarios and provide assistance for medical diagnosis.

关键词： deep learning Bone metastasis detection YOLOv5 Multi-scale Transformer

来源：评论

学校读者我要写书评

暂无评论

Analysis of Onground to Onboard processing Capabilities for Flood Segmentation using deepLabv3+ 3

Analysis of Onground to Onboard Processing Capabilities for ...

引用

3rd International Conference for Innovation in Technology, INOCON 2024

作者： Varshit, Namburu Chaitanya Ramesh Kumar, P. Rishi, Gudimetla Hanumath V.R.Siddhartha Engineering College Department of CSE Vijayawada India

ISBN: (纸本)9798350381931

Flood segmentation is a crucial aspect of disaster management and remote sensing applications. deep learning models have shown remarkable proficiency in semantic segmentation tasks. However, adapting these models for flood segmentation within an onboard processing framework presents unique challenges and opportunities. This analysis focuses on onground and onboard capabilities, carefully examining the advantages and implications of using single-board computers for real-time processing. The shift from onground to onboard processing, though promising, requires careful consideration of resource constraints. Onboard deployment can greatly improve the speed of flood detection and response, enabling faster and more informed decision-making. Additionally, it reduces dependence on external computational resources, making the system more resilient in disaster-prone areas where infrastructure may be compromised. To compare the performance of flood segmentation onground and onboard, the processing time, memory usage, and CPU consumption can be measured. Based on the results, the analysis led to the conclusion that the model produced an accuracy of 82% on a single-board computer. © 2024 IEEE.

关键词： Floods

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：