检索结果-内蒙古大学图书馆

2023 3rd International Conference on Detection Technology and Intelligence System, DTIS 2023

作者： Yuan, Enze Zhao, Bin Ding, Shuxue School of Artificial Intelligence Guilin University of Electronic Technology Guangxi Guilin541004 China Guangxi Colleges and Universities Key Laboratory of AI Algorithm Engineering Guangxi Guilin541004 China

Accurate and enough labels are crucial to constructing an excellent deep-learning framework in processing medical scans. However, the expensive cost of medical labels greatly impedes this process. In this case, semi-supervised learning has shown great potential due to its efficient use of unlabeled data. Therefore, we present a novel multi-scale reciprocal consistency network (RC-NET) to utilize the unlabeled data more efficiently for more accurate 3D left atrial segmentation. Our model consists of a common encoder and two independent decoders with inconsistent up-sampling. The decoders generate feature maps of hidden layers at different resolutions during the up-sampling process. Lower-resolution features typically contain local and detailed information, while higher-resolution features involve global or abstract information. Subsequently, we applied consistency learning to these feature maps. By combining local and global semantic information, the model can obtain a comprehensive understanding of the segmentation targets. The experimental data indicate the MRC-Net outperforms many semi-supervised learning methods. It achieves more accurate segmentation results by efficiently utilizing unlabeled data. This provides a completely new approach to improving semi-supervised learning. © Published under licence by IOP Publishing Ltd.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

CrossMatch: Enhance Semi-Supervised Medical Image Segmentation with Perturbation Strategies and Knowledge Distillation

引用

IEEE Journal of Biomedical and Health Informatics 2024年 PP卷 PP-PP页

作者： Zhao, Bin Wang, Chunshi Ding, Shuxue Guilin University of Electronic Technology School of Artificial Intelligence Guangxi Colleges and Universities Key Laboratory of AI Algorithm Engineering Guangxi Guilin541004 China

Semi-supervised learning for medical image segmentation presents a unique challenge of efficiently using limited labeled data while leveraging abundant unlabeled data. Despite advancements, existing methods often do not fully exploit the potential of the unlabeled data for enhancing model robustness and accuracy. In this paper, we introduce CrossMatch, a novel framework that integrates knowledge distillation with dual perturbation strategies, image-level and feature-level, to improve the model's learning from both labeled and unlabeled data. CrossMatch employs multiple encoders and decoders to generate diverse data streams, which undergo self-knowledge distillation to enhance the consistency and reliability of predictions across varied perturbations. Our method significantly surpasses other state-of-the-art techniques in standard benchmarks by effectively minimizing the gap between training on labeled and unlabeled data and improving edge accuracy and generalization in medical image segmentation. The efficacy of CrossMatch is demonstrated through extensive experimental validations, showing remarkable performance improvements without increasing computational costs. Code for this implementation is made available at https://***/aiEson/***. © 2013 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Steel Surface Defect Detection Based on Denoising Diffusion Implicit Models with Data Augmentation

Steel Surface Defect Detection Based on Denoising Diffusion ...

引用

International Conference on Imaging, Signal Processing and Communication (ICISPC)

作者： Yuhang Hong Ziwen Wang Wenzhen Wu Zhenbing Liu Benying Tan guangxi Chen Yujie Li Guilin University of Electronic Technology Guilin China Guangxi Colleges and Universities Key Laboratory of AI Algorithm Engineering Guilin China

ISBN: (数字)9798350367157

ISBN: (纸本)9798350367164

Due to the intricacy of steel production, surface defects unavoidably arise during the process. Detecting these defects is crucial for ensuring the production of high-quality steel. While some defect detection methods have been developed based on target detection models, simply applying object detection models without considering their efficiency often leads to subpar performance. This study addresses two key challenges in steel surface defect detection: insufficient data and detection efficiency. To tackle these challenges, we employed denoising diffusion implicit models (DDIMs) to augment the original dataset. Additionally, we introduced a novel method for defect detection in steel, termed YOLO-DSC, built upon YOLOv8 architecture. DDIMs were specifically leveraged to learn and train the publicly available NEU-DET dataset, generating reliable sample data. To enhance feature fusion, we integrated distribution shifting convolution into the bottleneck, effectively reducing the parameter count and computational load of YOLO-DSC. Experimental results conducted on NEU-DET demonstrate that YOLO-DSC exhibits robust performance and is well-suited for deployment on edge devices.

关键词： Performance evaluation Computational modeling Noise reduction Production Object detection Feature extraction Data augmentation

来源：评论

学校读者我要写书评

暂无评论

RATLIP: Generative Adversarial CLIP Text-to-Image Synthesis Based on Recurrent Affine Transformations

arXiv

引用

arXiv 2024年

作者： Lin, Chengde Lu, Xijun Chen, guangxi School of Artificial Intelligence Guangxi Colleges Universities Key Laboratory of AI Algorithm Engineering Guilin University of Electronic Technology Guangxi Guilin541004 China

Synthesizing high-quality photorealistic images with textual descriptions as a condition is very challenging. Generative Adversarial Networks (GANs), the classical model for this task, frequently suffer from low consistency between image and text descriptions and insufficient richness in synthesized images. Recently, conditional affine transformations (CAT), such as conditional batch normalization and instance normalization, have been applied to different layers of GAN to control content synthesis in images. CAT is a multi-layer perceptron that independently predicts data based on batch statistics between neighboring layers, with global textual information unavailable to other layers. To address this issue, we first model CAT and a recurrent neural network (RAT) to ensure that different layers can access global information. We then introduce shuffle attention between RAT to mitigate the characteristic of information forgetting in recurrent neural networks. Moreover, both our generator and discriminator utilize the powerful pre-trained model, Clip, which has been extensively employed for establishing associations between text and images through the learning of multimodal representations in latent space. The discriminator utilizes CLIP’s ability to comprehend complex scenes to accurately assess the quality of the generated images. Extensive experiments have been conducted on the CUB, Oxford, and CelebA-tiny datasets to demonstrate the superiority of the proposed model over current state-of-the-art models. The code is https://***/OxygenLu/RATLIP. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

CrossMatch: Enhance Semi-Supervised Medical Image Segmentation with Perturbation Strategies and Knowledge Distillation

arXiv

引用

arXiv 2024年

作者： Zhao, Bin Wang, Chunshi Ding, Shuxue School of Artificial Intelligence Guangxi Colleges Universities Key Laboratory of AI Algorithm Engineering Guilin University of Electronic Technology Guangxi Guilin541004 China

Semi-supervised learning for medical image segmentation presents a unique challenge of efficiently using limited labeled data while leveraging abundant unlabeled data. Despite advancements, existing methods often do not fully exploit the potential of the unlabeled data for enhancing model robustness and accuracy. In this paper, we introduce CrossMatch, a novel framework that integrates knowledge distillation with dual perturbation strategies, image-level and feature-level, to improve the model’s learning from both labeled and unlabeled data. CrossMatch employs multiple encoders and decoders to generate diverse data streams, which undergo self-knowledge distillation to enhance the consistency and reliability of predictions across varied perturbations. Our method significantly surpasses other state-of-the-art techniques in standard benchmarks by effectively minimizing the gap between training on labeled and unlabeled data and improving edge accuracy and generalization in medical image segmentation. The efficacy of CrossMatch is demonstrated through extensive experimental validations, showing remarkable performance improvements without increasing computational costs. Code for this implementation is made available at https://***/aiEson/***. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Gaze estimation based on swin transformer 6

Gaze estimation based on swin transformer

引用

6th International Conference on Advanced Electronic Materials, Computers, and Software engineering, AEMCSE 2023

作者： Chen, Jiahui Ma, Jiaxin Wang, Xiwen Huang, Longzhao Li, Yujie School of Artificial Intelligence Guilin University of Electronic Technology Jinji Road Guilin541004 China Guangxi Colleges and Universities Key Laboratory of AI Algorithm Engineering School of Artificial Intelligence Guilin University of Electronic Technology Guilin541004 China

ISBN: (纸本)9781510668249

The direction of human eye gaze is an important human behavior information that reflects the level of attention and cognitive state of the gazer towards various visual information in the environment. Eye gaze estimation has wide application value in multiple fields such as medical care, market research, and human-computer interaction. In recent years, some studies have introduced Transformer into the task of eye gaze estimation and achieved advanced performance. Although Transformer has better global modeling ability, its structural characteristics are not suitable for multi-scale feature learning in visual tasks. In addition, the global self-Attention calculation for images has high complexity. This paper introduces Swin Transformer into the field of eye gaze estimation, using self-Attention mechanism to perform more flexible and effective global modeling of images. The self-Attention calculation uses Windows Multi-head Self-Attention(W-MSA) and Shifted Windows Multi-head Self-Attention (SW-MSA), which greatly reduces the calculation of image self-Attention. The experimental results demonstrate that the Swin Transformer can obtain good results in the task of eye gaze estimation © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A Multi-Stage Framework for 3D Individual Tooth Segmentation in Dental CBCT

arXiv

引用

arXiv 2024年

作者： Wang, Chunshi Zhao, Bin Ding, Shuxue School of Artificial Intelligence Guilin University of Electronic Technology Guangxi Guilin541004 China Guangxi Colleges Universities Key Laboratory of AI Algorithm Engineering Guangxi Guilin541004 China

Cone beam computed tomography (CBCT) is a common way of diagnosing dental related diseases. Accurate segmentation of 3D tooth is of importance for the treatment. Although deep learning based methods have achieved convincing results in medical image processing, they need a large of annotated data for network training, making it very time-consuming in data collection and annotation. Besides, domain shift widely existing in the distribution of data acquired by different devices impacts severely the model generalization. To resolve the problem, we propose a multi-stage framework for 3D tooth segmentation in dental CBCT, which achieves the third place in the"Semi-supervised Teeth Segmentation" 3D (STS-3D) challenge. The experiments on validation set compared with other semi-supervised segmentation methods further indicate the validity of our approach. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Fabric4show: real-time vision system for fabric defect detection and post-processing

引用

Visual Intelligence 2024年第1期2卷 1-13页

作者： Lin, Huaizhou Cai, Dan Xu, Zengmin Wu, Jinsong Sun, Lixian Jia, Haibin School of Materials Science and Engineering Guangxi Key Laboratory of Information Guilin University of Electronic Technology Guilin China School of Mathematics and Computing Science Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation Guilin University of Electronic Technology Center for Applied Mathematics of Guangxi (GUET) Guilin China Anview.ai Guilin China School of Artificial Intelligence Guilin University of Electronic Technology Guilin China Department of Electrical Engineering University of Chile Santiago Chile

The exploration of computer vision applications for fabric defect detection has immense potential value. However, current relevant research in this area has primarily focused on detection models that aim for high detection accuracy and algorithmic efficiency, while neglecting the practical industrial production requirements. Therefore, we propose a fabric defect detection and post-processing system that integrates an optimized region with convolutional neural network (CNN) features (i.e., Faster R-CNN) for defect detection, defect localization and detection model evaluation. In addition, the proposed intelligent system incorporates novel approaches, such as a rearranged fabric dataset, anomaly detection, recommended clipping region division, and a replenishment device. This study illustrates an example of artificial intelligence (ai)-driven automated technology in fabric manufacturing. The accuracy and detection speed of different detection models under identical hardware conditions are evaluated and compared with related work. Experimental results demonstrate that the proposed approach achieves comparable performance to other models, while significantly reducing computational resource requirements. The potential efficiency of using two-stage networks on hardware systems for fabric defect detection tasks is highlighted, which is likely to have relevant implications for the textile industry. © The Author(s) 2024.

关键词： Automation Computer vision Defect detection Fabric manufacturing Faster R-CNN

来源：评论

学校读者我要写书评

暂无评论

A Novel Two-Stage Approach for 3D Dental Tooth Instance Segmentation 2nd

A Novel Two-Stage Approach for 3D Dental Tooth Instance Seg...

引用

2nd International ToothFairy2: Multi-Structure Segmentation in CBCT Volumes, ToothFairy 2024, 3D Teeth Landmarks Detection Challenge, 3DTeethLand 2024, Semi-supervised Teeth Segmentation, STS 2024 held in conjunction with the 27th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCai 2024

作者： Chen, Yuhan Wang, Chunshi Zhao, Bin School of Computer Science and Information Security Guilin University of Electronic Technology Guilin541004 China School of Artificial Intelligence Guilin University of Electronic Technology Guangxi Guilin541004 China Guangxi Colleges and Universities Key Laboratory of AI Algorithm Engineering Guangxi Guilin541004 China

ISBN: (纸本)9783031889769

Tooth CBCT instance segmentation is a foundational step in advancing digital dental health systems, with the precision of tooth segmentation playing a critical role in accurate medical diagnosis. Deep learning based methods face significant challenges in accurately locating and classifying teeth in 3D dental images, especially given the complex anatomical structures present in CBCT scans. Additionally, manually identifying and labeling each tooth from these scans is an extremely time-consuming and burdensome task for medical professionals. To address these issues, in this paper, we propose a two-stage semi-supervised method for tooth instance segmentation, along with innovative data preprocessing. Our method achieves a distinguished fourth place on the validation set of the "MICCai STS 2024 Challenge Task 2", outperforming other mainstream semi-supervised methods. Code for this implementation is made available at https://***/magic-fortune/MICCai2024. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

A novel fire monitoring system for electric bicycle shed based on YOLOv8

A novel fire monitoring system for electric bicycle shed bas...

引用

IEEE International Symposium on Embedded Multicore Socs (MCSoC)

作者： Hongyang Zhong Ziwen Wang Zhiyi Chen Wenhao Chen Yujie Li Guilin University of Electronic Technology Guilin China School of Artificial Intelligence Guangxi Colleges and Universities Key Laboratory of AI Algorithm Engineering Guilin University of Electronic Technology Guilin China

With the increasing popularity of electric bicycles, safety concerns regarding electric bicycle sheds have garnered public attention. In order to protect people’s life and property safety, an accurate and real-time fire monitoring system for electric bicycle shed is crucial. We proposed a YOLOv8 based fire monitoring system for electric bicycle shed. The system can be conveniently deployed on various end devices, computers, and IoT terminals. The proposed system uses deep learning technology to offer round-the-clock real-time monitoring, early fire detection, and automatic alarm triggering capabilities. The experimental results demonstrated that the proposed system can successfully monitor fire for electric bicycle. It is highly practical and can be widely used for fire monitoring and other similar applications.

关键词：

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：