检索结果-内蒙古大学图书馆

2021 4th International Conference on E-Business, Information Management and Computer Science

作者： Yuehao Tang Chong Guo Changjiang Institute of Technology China Wuhan Business University China

ISBN: (纸本)9781450395687

Abstract: In order to improve the intellective level of water resources management, a real-time water level recognition method based on deep-learning algorithms and image-processing techniques is proposed in this paper. The recognition process is composed of four steps. Firstly, for the purpose of digit detection, YOLO-v3 model is deployed for extracting numbers from the water gauges. Then, the cropped number images are fed into the LSTM + CTC model as training samples so that digits can be recognized. In the third step, Hough transform are adopted to correct the tilt of water gauge in terms of the vertical edge feature. Morphological operation, associated with horizontal projection would position upper and lower edge of water gauge to recognize the scale lines correctly. Water level could be determined correspondingly. Model application shows that the recognition model has satisfying accuracy and efficiency, with potential being applied in practice.

关键词： Water Gauge image processing Number Recognition deep learning Water Level Number Detection

来源：评论

学校读者我要写书评

暂无评论

deep learning based smart traffic light system using image processing with YOLO v7 4

Deep Learning based smart traffic light system using Image P...

引用

4th International Conference on Circuits, Control, Communication and Computing, I4C 2022

作者： Rangari, Avinash Padmakar Chouthmol, Ashwini Ravindra Kadadas, Chaitanya Pal, Prashant Kumar Singh, Shanshank National Institute of Electronics & Information Technology Dept of ESE Aurangabad India

ISBN: (纸本)9798350397475

India is home to 10% of all traffic deaths worldwide and has the second-largest road network in the world. Moreover, in smart cities, traffic congestion, pollutants, and noise pollution have increased due to a constant rise in vehicle kinds, technical problems with traffic signal management equipment, and inefficient road traffic management. Despite the fact that current traffic control systems rely on fixed time-based techniques, conventional traffic control systems are unable to manage the complicated traffic flow at junctions. Roadblocks increase mileage, increase transport costs, and pollute the air in addition to adding to the driver's stress and further delays. Therefore, we designed a smart traffic light management system employing the recently launched YOLO V7. The new version V7 of the YOLO algorithm outperforms all previous object detection models in both speed and accuracy. As it is the fastest and most accurate real-time object detection model hence it is the best algorithm to deploy in traffic controlling system. YOLO V7 is +120% faster than other previous models and shows the best speed to accuracy balance. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Adaptive transfer learning using SegFormer for imbalanced pixel in medical image segmentation

引用

SIGNAL image AND VIDEO processing 2025年第8期19卷 1-12页

作者： El Joudi, Niama Assia Lazaar, Mohamed Delmotte, Francois Allaoui, Hamid Mahboub, Oussama Mohammed V Univ ENSIAS Rabat Morocco Univ Artois UR 3926 Lab Genie Informat & Automat Artois LGI2A Technoparc FUTURA F-62400 Bethune France Abdelmalek Essaadi Univ ENSA Tetouan Morocco

The real-world medical datasets are often inherently challenged by imbalanced classes, which impact the performance of deep learning models, leading to overfitting and limited effectiveness. These limitations are particularly pronounced in image segmentation tasks, where accurate delineation of anatomical structures is essential to support clinical decision-making. In order to match the recent advancements and enhance the model's generalizability and its ability to classify correctly the minor class, specifically the foreground pixels, we applied the generalized dice loss in conjunction with transfer learning, avoiding the redundancy provided by traditional data augmentation techniques and heavy computational data generation strategies. In this paper, we demonstrated that the choice of the loss function plays a pivotal role in optimizing the learning landscape and guiding the model's training process. The proposed approach generated the highest Dice Coefficient value of 98.44% compared with the existing works and augmentation of 5.24% compared with the network that employed the cross-entropy Loss function. Experimental results indicate that the proposed hybrid approach can accurately identify and segment different shapes of the fetal head, enabling real-time processing and providing a significant potential to assist clinical diagnosis for further circumference measurement.

关键词： deep learning Imbalanced data Loss function Transfer learning SegFormer Medical imaging analysis

来源：评论

学校读者我要写书评

暂无评论

Research on Water Gauge Recognition Based on deep learning and image processing 4

Research on Water Gauge Recognition Based on Deep Learning a...

引用

4th International Conference on E-Business, Information Management and Computer Science, EBIMCS 2021

作者： Tang, Yuehao Guo, Chong Changjiang Institute Of Technology School Of Water Conservancy And Electric Power Wuhan China Wuhan Business University Research Center For Chinese And Western Language And Culture Wuhan China

ISBN: (纸本)9781450395687

In order to improve the intellective level of water resources management, a real-time water level recognition method based on deep-learning algorithms and image-processing techniques is proposed in this paper. The recognition process is composed of four steps. Firstly, for the purpose of digit detection, YOLO-v3 model is deployed for extracting numbers from the water gauges. Then, the cropped number images are fed into the LSTM + CTC model as training samples so that digits can be recognized. In the third step, Hough transform are adopted to correct the tilt of water gauge in terms of the vertical edge feature. Morphological operation, associated with horizontal projection would position upper and lower edge of water gauge to recognize the scale lines correctly. Water level could be determined correspondingly. Model application shows that the recognition model has satisfying accuracy and efficiency, with potential being applied in practice. © 2021 ACM.

关键词： Water levels

来源：评论

学校读者我要写书评

暂无评论

Intersections and crosswalk detection using deep learning and image processing techniques

引用

PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS 2020年 543卷 123510-123510页

作者： Tumen, Vedat Ergen, Burhan Bitlis Eren Univ Bitlis Turkey Firat Univ Elazig Turkey

Road separations, intersections, and crosswalks, which are important components of highways, are seen as significant areas for autonomous vehicles and advanced driver assistance systems because traffic accident occurrence rate is considerably high in these areas. In this study, an image processing method and a deep learning based approach on real images has been proposed in order to provide instant information for drivers and autonomous vehicles, or to develop warning systems as part of advanced driver assistance systems to prevent or minimize traffic accidents. The information is obtained from the classification of images belonging to the separations, intersections and crosswalks on the road using a new model and VggNet, AlexNet, LeNet based on Convolutional Neural Network(CNN). We have obtained high classification accuracy with our model based on CNN. The result of the study performed on different datasets showed that the proposed method is usable for driver assistance systems and an effective structure that can be used in many areas such as warning both vehicles and drivers. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Road intersection detection Crosswalk detection deep learning Intelligent transportation systems

来源：评论

学校读者我要写书评

暂无评论

Application of Convolutional Neural Networks for Parallel Multi-Scale Feature Extraction in Noise image Denoising

引用

IEEE ACCESS 2024年 12卷 98599-98610页

作者： Li, Yiming Xie, Tao Mei, Dongdong Ningxia Inst Sci & Technol Coll Comp Sci & Engn Shizuishan 753000 Peoples R China

Although deep learning techniques have made significant advances in the field of images, existing methods still face challenges in processing complex, noisy images. In view of the limitation that most denoising models only focus on extracting single scale features, a new denoising network structure is proposed in this paper. Firstly, the channel attention mechanism and convolutional neural network are combined to construct a real image denoising model, and then the parallel multi-scale convolutional neural network is constructed by combining the adaptive dense connected residual block and parallel multi-scale feature extraction module. The results showed that the designed model can reach the stable state only after 121 and 86 iterations on the training set and the test set, and the denoising accuracy of the model is as high as 0.96. In addition, the research model has high computational efficiency and short denoising time when processing noisy images, and the processing time of an image is as low as 0.09s. Therefore, the proposed denoising structure has good denoising performance under different noise levels and types, and this study also provides a new idea for the application of deep learning in image denoising and other image processing tasks.

关键词： Noise measurement Noise reduction image denoising Mathematical models Convolutional neural networks Multi-scale CNN attention mechanism feature extraction noise image residual residual

来源：评论

学校读者我要写书评

暂无评论

Designing deep Reinforcement learning enhanced edge-terminal collaborative AIoT for Intelligent Visitor Management System

引用

AD HOC NETWORKS 2025年 169卷

作者： Liao, Yong Zhu, Zhiyuan Tang, Tong Wu, Dapeng Wang, Ruyan Chongqing Univ Posts & Telecommun Sch Commun & Informat Engn Chongqing 400065 Peoples R China Key Lab Chongqing Educ Commiss China Adv Network & Intelligent Connect Technol Chongqing 400065 Peoples R China Chongqing Key Lab Ubiquitous Sensing & Networking Chongqing 400065 Peoples R China

Intelligent Visitor Management System (IVMS) is crucial for enhancing security and operational efficiency in smart factories and intelligent office buildings. Leveraging AIoT-driven image analysis will facilitate realtime visitor authentication and access control. However, the growing volume of interactions and the limited processing power of local terminals complicate the delivery of timely and accurate image analysis. To address these challenges, we propose an edge-terminal collaborative AIoT framework for real-time visitor management. The framework solves the limitations of traditional approaches, where local terminals are unable to handle the computational load and edge solutions experience high latency due to transmission delays. Specifically, it integrates three key components to improve system performance: a local analysis module for initial processing, an image communication module for efficient data transmission, and an edge analysis module for advanced processing. Moreover, the framework jointly optimizes image task offloading, wireless channel allocation, and image compression, all formulated as an optimization problem to ensure fast and accurate analysis. Additionally, a novel multi-level deep Reinforcement learning (DRL) method is further designed to dynamically refine the selection of compression and offloading strategies. By learning in real-time, the DRL model adapts to network variations, addressing the scalability and adaptability limitations of existing methods. Simulation results show that our proposed edge-terminal collaborative AIoT framework significantly outperforms both edge-only and terminal-only methods in terms of latency and accuracy.

关键词： Intelligent Visitor Management System (IVMS) Terminal-edge collaborative AIoT deep Reinforcement learning (DRL)

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Traditional and deep learning Approaches for Underwater Remote Sensing image Enhancement: A Quantitative Study

引用

JOURNAL OF MARINE SCIENCE AND ENGINEERING 2025年第5期13卷 899-899页

作者： Ma, Yunsheng Cheng, Yanan Zhang, Dapeng Guangdong Ocean Univ Ship & Maritime Coll Zhanjiang 524005 Peoples R China Guangdong Ocean Univ Sch Elect & Informat Engn Zhanjiang 524088 Peoples R China NJUST Taizhou Inst Sci & Technol Coll Business Taizhou 225300 Peoples R China

Underwater remote sensing image enhancement is complicated by low illumination, color bias, and blurriness, affecting deep-sea monitoring and marine resource development. This study compares a multi-scale fusion-enhanced physical model and deep learning algorithms to optimize intelligent processing. The physical model, based on the Jaffe-McGlamery model, integrates multi-scale histogram equalization, wavelength compensation, and Laplacian sharpening, using cluster analysis to target enhancements. It performs well in shallow, stable waters (turbidity < 20 NTU, depth < 10 m, PSNR = 12.2) but struggles in complex environments (turbidity > 30 NTU). deep learning models, including water-net, UWCNN, UWCycleGAN, and U-shape Transformer, excel in dynamic conditions, achieving UIQM = 0.24, though requiring GPU support for real-time use. Evaluated on the UIEB dataset (890 images), the physical model suits specific scenarios, while deep learning adapts better to variable underwater settings. These findings offer a theoretical and technical basis for underwater image enhancement and support sustainable marine resource use.

关键词： underwater remote sensing deep learning algorithms multi-scale fusion-enhanced physical model underwater image enhancement

来源：评论

学校读者我要写书评

暂无评论

Non-invasive respiratory infection monitoring using AI-driven thermal imaging and signal classification

引用

SIGNAL image AND VIDEO processing 2025年第7期19卷 1-14页

作者： Abisha, D. Natl Engn Coll Dept Comp Sci & Engn Kovilpatti Tamil Nadu India

The COVID-19 pandemic has highlighted the need for efficient and non-contact health screening methods. Signal-based infrared imaging is an emerging field in biomedical engineering that enables remote monitoring of vital signs. While fever is a common symptom, respiratory abnormalities often appear earlier, necessitating advanced screening systems that monitor both body temperature and respiratory patterns. This research presents an artificial intelligence-based screening device for health that identifies human respiratory patterns based on a deep learning model. The device is built with a Convolutional Neural Network (CNN) to extract features and a Long Short-Term Memory (LSTM) network to classify time-series patterns. The Softmax classifier accurately classifies respiratory patterns. It is learned on a specialized dataset of six breathing signal patterns, making it an effective model for real-time public health surveillance. The experimental result demonstrates that the proposed CNN-LSTM model achieves 91% accuracy, 90% precision, 93% recall, and an F1-score of 91%. It can be scaled up even further for medical real-time applications, paves the way to even greater future advancements in automated health surveillance.

关键词： Respiratory signals Health screening Feature extraction Signal processing deep learning techniques Classification of patterns

来源：评论

学校读者我要写书评

暂无评论

An efficient single-stage ISP for smartphones using global context residual dense and residual channel attention modules

引用

JOURNAL OF real-time image processing 2025年第3期22卷 1-14页

作者： Bansal, Roli Pal, Anjali Sehgal, Priti Univ Delhi Fac Math Sci Dept Comp Sci New Delhi 110007 India Keshav Mahavidyalaya Dept Comp Sci New Delhi 110034 India

The utilization of smartphone cameras to capture photographs is immensely popular in the world. Smartphone image signal processors are used to produce high-quality images. The field of image Signal processing (ISP) in smartphone cameras involves the application of various techniques and algorithms that process the raw image data acquired by the smartphone camera sensor into a high-quality Red, Green, and Blue (RGB) image. The visual difference between images captured by smartphone cameras and Digital Single Lens Reflex (DSLR) cameras can be attributed to the constrained sizes of smartphone camera sensors and lenses. To address the existing disparity in visual differences of these devices, there is a need to redesign the smartphone ISP with the aim of reconstructing the good quality of the captured images in real time. This work proposes a single-stage end-to-end deep-learning model that can replace most complex ISP pipelines of smartphone cameras. The training of the proposed model is independent of the sensor and optics employed in a specific device. The proposed single-stage ISP pipeline for smartphone cameras uses the Global Context Residual Dense (GCRD) module, the Multiple Convolution Block (MCB) module, and the Residual Channel Attention (RCA) module. The GCRD module is used to learn the residual information which helps in color mapping of the raw image to the corresponding RGB image. At the same time, the MCB module with multiple convolution blocks consisting of layers of different kernel sizes focuses on the fine-grained details of the image. Further, the RCA module used in the proposed work consists of a very high deep trainable network that adaptively learns more beneficial channel-wise features simultaneously. By combining these modules, the pipeline achieves a synergistic effect that helps in balancing global context, local refinement, and feature prioritization, enabling superior performance in complex ISP operations. This work evaluates the proposed mo

关键词： image signal processing (ISP) Smartphones image restoration Sensors Lens Residual channel attention (RCA) Global context residual dense network (GCRD)

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：