检索结果-内蒙古大学图书馆

The study on ultrasound image classification using a dual-branch model based on Resnet50 guided by U-net segmentation results

引用

BMC MEDICAL IMAGING 2024年第1期24卷 1-16页

作者： Yang, Xu Qu, Shuo'ou Wang, Zhilin Li, Lingxiao An, Xiaofeng Cong, Zhibin Changchun Univ Sci & Technol Sch Elect & Informat Engn Changchun 130022 Peoples R China CCUCM Human Resources Dept Affiliated Hosp 3 Changchun 130117 Peoples R China Jilin Engn Normal Univ Educ Qual Monitoring Ctr Changchun 130052 Peoples R China Univ Tradit Chinese Med Dept Electrodiag Affiliated Hosp Changchun Changchun 130021 Peoples R China

In recent years, the incidence of nodular thyroid diseases has been increasing annually. Ultrasonography has become a routine diagnostic tool for thyroid nodules due to its high real-time capabilities and low invasiveness. However, thyroid images obtained from current ultrasound tests often have low resolution and are plagued by significant noise interference. Regional differences in medical conditions and varying levels of physician experience can impact the accuracy and efficiency of diagnostic results. With the advancement of deep learning technology, deep learning models are used to identify whether a nodule in a thyroid ultrasound image is benign or malignant. This helps to close the gap between doctors' experience and equipment differences, improving the accuracy of the initial diagnosis of thyroid nodules. To cope with the problem that thyroid ultrasound images contain complex background and noise as well as poorly defined local features. in this paper, we first construct an improved ResNet50 classification model that uses a two-branch input and incorporates a global attention lightening module. This model is used to improve the accuracy of benign and malignant nodule classification in thyroid ultrasound images and to reduce the computational effort due to the two-branch *** constructed a U-net segmentation model incorporating our proposed ACR module, which uses hollow convolution with different dilation rates to capture multi-scale contextual information for feature extraction of nodules in thyroid ultrasound images and uses the results of the segmentation task as an auxiliary branch for the classification task to guide the classification model to focus on the lesion region more efficiently in the case of weak local features. The classification model is guided to focus on the lesion region more efficiently, and the classification and segmentation sub-networks are respectively improved specifically for this study, which is used to improve the accurac

关键词： deep learning Thyroid ultrasound images Resnet50 U-net: attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Adaptive Edge Systems for Smart IoT Applications

Adaptive Edge Systems for Smart IoT Applications

引用

作者： Liu, Miaomiao University of California Merced

学位级别：Ph.D., Doctor of Philosophy

The proliferation of the Internet of Things (IoT) and cloud services has given rise to the edge computing paradigm, where data is processed partly or entirely at the edge of the network, rather than solely in the cloud. Edge computing can address problems such as latency, limited battery life of mobile devices, bandwidth costs, security, and privacy. Typical applicable scenarios based on edge computing include video analytics, smart home, smart city, and collaborative *** the development of deep learning techniques, research on employing deep learning to develop intelligent edge systems is emerging. In this dissertation, we aim to investigate how deep learning can process data on source-constrained individual edge devices in real time and how deep learning can process data by utilizing collaborative edge devices to provide better *** build several critical systems, including video analytics, driving anomaly detection, arm posture tracking, and device orientation tracking. In the video analytics system, we combine deep learning with traditional image processing techniques to achieve real-time object detection on mobile devices without offloading. In the driving anomaly detection system, we train deep learning models for driving anomaly detection by leveraging the information from collaborative peer devices to provide better accuracy. In the arm posture tracking system, we employ multitask learning to track the orientation and location of the wrist simultaneously, which significantly improves the latency compared to the conventional methods. In the device orientation tracking system, we develop a deep reinforcement learning framework to train an agent that adjusts the parameters of a conventional orientation tracking method in response to changing *** IoT systems continue to grow in complexity and size, preserving training data has become an increasingly important challenge. In our future work, we plan to investigate the use of representation

关键词： deep learning Edge computing Internet of Things Mobile devices real-time performance System adaptation

来源：评论

学校读者我要写书评

暂无评论

Lightweight TransUNet Informal Residential Area segmentation 2

Lightweight TransUNet Informal Residential Area segmentation

引用

2nd International Conference on Machine Vision, image processing and Imaging Technology, MVIPIT 2024

作者： Zhang, Shuo Jia, Yiru Liu, Zhaoying Yang, Fei School of Computer science Beijing University of Technology Beijing China Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798331543037

Remote sensing image informal residential area segmentation plays an important role in urban planning, environmental detection, and disaster assessment. With the development of deep learning, the performance of informal residential area extraction has been improved. However, trying to accurately extract the boundaries between different buildings in the same complex is still a challenging task, and at the same time, segmentation model training takes a lot of time. To address the above problems, we propose a deep learning-based segmentation method for remote sensing images of informal residential area instances, aiming to further improve the accuracy of informal residential area segmentation. This paper has the following three main contributions. First, this paper designs and implements a remote sensing image informal residential area instance segmentation method called TransUNet-IRA, which is a model based on TransUNet, and reduces the training file weights by reducing the number of encoder layers and changing the format of the dataset, which increases the computing speed and improves the performance of the model. Second, we combine DiceCE Loss with IoU Loss to form DiceCE-IoU Loss to further improve the model’s performance in informal residential area segmentation. Finally, we collect a large amount of meter-level remote sensing image data from Guilin City, Guangxi Province, and Chenzhou City, Hunan Province, and construct a remote sensing image informal residential area instance segmentation dataset for these two regions based on these remote sensing images, which is annotated with three segmentations of the background, informal residential area, and single building in the formal residential area. After experimental validation, our method demonstrates performance advantages on the constructed dataset, not only outperforming classical models such as UNet, ResUNet, AttUNet, and SwinUNet but also improving on the original TransUNet. ©2024 IEEE.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Decentralized Federated deep learning image Recognition Models 4

Decentralized Federated Deep Learning Image Recognition Mode...

引用

4th International Conference on Artificial Intelligence, Robotics and Control, AIRC 2023

作者： Kugan, Sharun Islam, Md Quyyum Ul Kashef, Rasha Toronto Metropolitan University Department of Electrical Computer Biomedical Engineering Toronto Canada

ISBN: (纸本)9798350348248

In the era of IoT, numerous frameworks and cutting-edge models have been introduced to enhance user experience and privacy and reduce the risk of data breaches. Over time, IoT device usage has grown tremendously, and a flood of data has been sent to servers for processing. Federated learning has been deployed for efficient decentralization while preserving privacy. Federated learning has been applied in various IoT-related applications such as image classification, object segmentation, object detection, and sensor analytics. Existing centralized image recognition models fall short of providing accurate image classification with acceptable processing time for real-time deployment while preserving privacy. In this paper, we designed two decentralized deep learning models using federated learning, the CNN-TFF and the VGG16-TFF. With around 250 training iterations, we achieved a high accuracy rate of up to 90% with a decrease in the loss value for the CIFAR-100 dataset using the VGG16-TFF model while maintaining data privacy using federated learning. © 2023 IEEE.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

real-time processing of image, Depth and Video Information 2023

Real-time Processing of Image, Depth and Video Information 2...

引用

real-time processing of image, Depth and Video Information 2023

ISBN: (纸本)9781510662629

The proceedings contain 16 papers. The topics discussed include: evolution of real-time processing of visual information over four decades: a retrospective as outlook to the future of real-time imaging;real-time embedded large-scale place recognition for autonomous ground vehicles using a spatial descriptor;real-time video super-resolution reconstruction using wavelet transforms and sparse representation;development of light-field motion tracking technology for use in laboratory studies of planet formation;towards learning-based denoising of light fields;real-time onboard visual parking space detection: a performance study;an automated AI and video measurement techniques for monitoring social distancing, mask detection, and facial temperature screening for COVID-19;computational efficient deep learning-based super resolution approach;and in-sensor neural network for real-time KWS by image processing.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fourth International Conference on Signal processing and Machine learning, CONF-SPML 2024

Fourth International Conference on Signal Processing and Mac...

引用

4th International Conference on Signal processing and Machine learning, CONF-SPML 2024

ISBN: (纸本)9781510674721

The proceedings contain 28 papers. The topics discussed include: phototropic bionics: realization of intelligent machine detection and obstacle avoidance;a study of model predictive control and reinforcement learning control system;advancements and challenges in speech emotion recognition: a comprehensive review;revolutionizing ADHD diagnosis: deep learning in 3D medical imaging;improving robustness in emotion recognition via adversarial training;real image improvement study based on pivotal tuning inversion;a review of 3D printing slicing algorithms;and analysis of two variants of U-net for pulmonary nodule segmentation: attention U-net and dense-attention U-net.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Detection of Healthy and Diseased Plant Leaf Based On Alexnet-Convolutional Neural Network Using deep learning 9

Detection of Healthy and Diseased Plant Leaf Based On Alexne...

引用

9th International Conference on Science, Technology, Engineering and Mathematics, ICONSTEM 2024

作者： Madireddy, Swathi Shareef, Md. Mohammad Nagendra, Kolluru Venkata Sekhar, Bunga Sushma, E. Sandhyarant Mallareddy College of Engineering And Technology Department of Emerging Technologies Telangana Hyderabad India Cmr Technical Campus Department of Cse Telangana Hyderabad India Sree Venkateswara College of Engineering Department of Cse Nellore India Telangana Hyderabad India

ISBN: (纸本)9798350365092

In current years, the most widely used image processing is to analyze and detect plant leaf diseases. Furthermore, using image processing has an important effect on crop yield and plant leaf quality. This is an important process in addressing the problem of plant leaf diseases in agriculture. Therefore, image processing can enhance the effectiveness and exactness of plant disease detection. However, the detection of plant diseases faces significant challenges such as time consumed, processing and large amounts of data. Therefore, the existing method is inefficient and has low accuracy in detecting plant leaf diseases. To solve this problem, the proposed A-CNN (Alexnet-Convolutional Neural Network) technique is applied to accurately identify healthy and unhealthy plant leaf. Moreover, we utilize HEMF (Histogram Equalization Mean Filter) technique for pre-processing to eliminate noise in plant leaf images, reduce background noise, and enhance image quality. Afterwards, features can be selected using the Enhanced Particle Swarm Optimization (EPSO) algorithm to identify disease behaviour in plant leaves. Finally, the A-CNN method proposed can be utilized to identify healthy and diseased parts in plant leaves using deep learning (DL), resulting in enhanced accuracy. Moreover, test results can be evaluated using accuracy, sensitivity, specificity, and accuracy based on the offered technique. Thus, the experimental effects show that the proposed ACNN technique can detect plant leaf diseases effectively. © 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 deep reinforcement learning based approach for dynamic job shop scheduling considering variable processing time 24

A deep reinforcement learning based approach for dynamic job...

引用

4th International Conference on Artificial Intelligence, Automation and High Performance Computing, AIAHPC 2024

作者： Yang, Shuai Guo, Hongwei Huang, Jiaqi Han, Kexian School of Mechanical Engineering Beijing Institute of Technology China

ISBN: (纸本)9798400710049

The job shop scheduling problem is a common challenge in intelligent manufacturing. In a real workshop environment, parameters like processing time often change dynamically. Scheduling strategies must be adjusted flexibly and quickly to match the current state. However, traditional methods only find the optimal solution for a specific instance. When environment parameters change, recalculations are required, leading to high time costs. To address these problems, a scheduling method called S2S-AC, based on deep reinforcement learning, is proposed to efficiently solve the dynamic job shop scheduling problem with variable processing times. In the proposed method, JSSP is modeled as a sequential decision-making problem using a Markov Decision Process. A new state set model, including four static states and two dynamic states, is designed with operations as the action set. An end-to-end framework that combines the Pointer Network model with the A2C algorithm is used to construct the DRL network, which is trained with multiple samples. The trained network directly outputs the scheduling strategy for new instances without requiring retraining. In static experiments, the effectiveness of S2S-AC is verified by comparing its solution results with those of SPT, LPT, MTWR, and the genetic algorithm on benchmark instances. In dynamic experiments, S2SAC achieved the best solution results in all randomly generated test instances based on instance ft10, compared to the above methods, with relatively short solution times. © 2024 Copyright held by the owner/author(s).

关键词： deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Spain on fire: A novel wildfire risk assessment model based on image satellite processing and atmospheric information

引用

KNOWLEDGE-BASED SYSTEMS 2024年 283卷

作者： Liz-Lopez, Helena Huertas-Tato, Javier Perez-Aracil, Jorge Casanova-Mateo, Carlos Sanz-Justo, Julia Camacho, David Univ Politecn Madrid Comp Syst Dept Calle Alan Turing S-N Madrid 28031 Spain Univ Alcala Signal Theory & Commun Dept Plaza San Diego S-N Alcala De Henares 28801 Spain Univ Valladolid Remote Sensing Lab LATUV Plaza Santa Cruz Valladolid 47002 Spain

Each year, wildfires destroy larger areas of Spain, threatening numerous ecosystems. Humans cause 90% of them (negligence or provoked) and the behaviour of individuals is unpredictable. However, atmospheric and environmental variables affect the spread of wildfires, and they can be analysed by using deep learning. In order to mitigate the damage of these events, we proposed the novel Wildfire Assessment Model (WAM). Our aim is to anticipate the economic and ecological impact of a wildfire, assisting managers in resource allocation and decision-making for dangerous regions in Spain, Castilla y Leon and Andalucia. The WAM uses a residual-style convolutional network architecture to perform regression over atmospheric variables and the greenness index, computing necessary resources, the control and extinction time, and the expected burnt surface area. It is first pre-trained with self-supervision over 100,000 examples of unlabelled data with a masked patch prediction objective and fine-tuned using a very small dataset, composed of 445 samples. The pretraining allows the model to understand situations, outclassing baselines with a 1,4%, 3,7% and 9% improvement estimating human, heavy and aerial resources;21% and 10,2% in expected extinction and control time;and 18,8% in expected burnt area. Using the WAM we provide an example assessment map of Castilla y Leon, visualizing the expected resources over an entire region.

关键词： Wildfire risk assessment deep learning Autoencoder Regression model Fusion Atmospheric variables Few-shot learning

来源：评论

学校读者我要写书评

暂无评论

EF-Yolo: An Efficient and Lightweight Network for real-time Components Detection of Freight Trains

引用

IEEE SENSORS JOURNAL 2024年第21期24卷 35872-35888页

作者： Feng, Jiachen Shi, Hongmei Qiu, Ji Yu, Zujun He, Chao Beijing Jiaotong Univ State Key Lab Adv Rail Autonomous Operat Sch Mech Elect & Control Engn Beijing 100044 Peoples R China Beijing Jiaotong Univ Collaborat Innovat Ctr Railway Traff Safety Beijing 100044 Peoples R China

Component detection is crucial to the trouble of moving freight car detection system based on image recognition and analysis, serving as the foundational step for downstream fault analysis. However, existing detection algorithms struggle to meet stringent accuracy and speed requirements in real-world high-density vehicle dynamic inspection scenarios. Efficient (EF)-you only look once (YOLO) is proposed to enhance detection speed, whereas preserving high accuracy. EF-Yolo is a lightweight train component detection network that incorporates two efficient gradient-based feature extraction modules, where C2f-Ghost focuses on lightweighting and C2f-SE-Ghost focuses on feature extraction. Furthermore, we introduce a novel module deployment strategy that focuses on optimizing the arrangement of feature extraction modules based on the network's architectural characteristics, thereby maximizing model inference efficiency and detection performance. Experimental results on a freight train components dataset demonstrate that EF-YOLO outperforms state-of-the-art detectors in competitive accuracy and achieves a remarkable detection speed of 70 frames/s, whereas maintaining a modest model size of only 9.15 MB, and can meet the efficiency requirement of completing the detection of 15 000 images of a single train within ten minutes.

关键词： deep learning freight train images lightweight object detection real-time

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：