检索结果-南通市图书馆

A Remote Sensing Image semantic segmentation Method by Combining Deformable Convolution with Conditional Random Fields

引用

Journal of Geodesy and Geoinformation Science 2020年第3期3卷 39-49页

作者： Zongcheng ZUO Wen ZHANG Dongying ZHANG School of Aeronautics and Astronautics Shanghai Jiao Tong UniversityShanghai 200240China School of Remote Sensing and Information Engineering Wuhan UniversityWuhan 430079China School of Hydropower and Information Engineering Huazhong University of Science and TechnologyWuhan 430074China

Currently,deep convolutional neural networks have made great progress in the field of semantic *** of the fixed convolution kernel geometry,standard convolution neural networks have been limited the ability to simulate geometric ***,a deformable convolution is introduced to enhance the adaptability of convolutional networks to spatial *** that the deep convolutional neural networks cannot adequately segment the local objects at the output layer due to using the pooling layers in neural network *** overcome this shortcoming,the rough prediction segmentation results of the neural network output layer will be processed by fully connected conditional random fields to improve the ability of image *** proposed method can easily be trained by end-to-end using standard backpropagation ***,the proposed method is tested on the ISPRS *** results show that the proposed method can effectively overcome the influence of the complex structure of the segmentation object and obtain state-of-the-art accuracy on the ISPRS Vaihingen 2D semantic labeling dataset.

关键词： high-resolution remote sensing image semantic segmentation deformable convolution network conditions random fields

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

An Appearance Inspection Method for Resistance Spot Welding Based on semantic segmentation

An Appearance Inspection Method for Resistance Spot Welding ...

引用

作者： Tianyu Zheng Ye Yang Peng Zheng Lukas Benz Lei Wang School of Aerospace Engineering Xiamen University Faculty of Mechanics and Electronics Heilbronn University

Resistance spot welding（RSW） plays an important role in *** quality of the welding can be efficiently assessed by its *** segmentation is an important part of the RSW appearance ***,the classical image segmentation algorithms cannot work very well because of the various RSW *** this study,a novel inspection method is proposed based on semantic *** choose MobileNetV2 as the backbone for the semantic *** modification and optimization of the network,our model achieves an accuracy of 89% mean intersection-ofunion（mIOU）,which is averagely 30% higher than the classical image segmentation algorithms.A classifier further evaluates the quality of the RSW according to some geometric features of the segmented regions,and the classification accuracy is improved by 0.79%.This research is of great importance for the high accuracy quality control of the massive production to reduce the producing cost and improve the efficiency of the RSW pipeline.

关键词： Resistance spot welding semantic segmentation Appearance inspection

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Retinal Blood Vessels semantic segmentation Method Based on Modified U-Net

Retinal Blood Vessels Semantic Segmentation Method Based on ...

引用

第30届中国控制与决策会议

作者： Ling Luo Dali Chen Dingyu Xue College of Information Science and Engineering Northeastern University

Automatic segmentation of retinal blood vessels from fundus images plays a key role in the computer aided diagnostic system, which is helpful for the early treatment of many fundus diseases including diabetic retinopathy, glaucoma and hypertension. In this paper, a modified U-Net is proposed to train semantic segmentation models for retinal blood vessels. In addition, we use Condition Random Field to integrate the global information. The comparison between our method and other typical methods is given to evaluate the proposed method. Our experiments shows that our scheme leads to improved accuracy on DRIVE datasets and having obtained an average accuracy of 86.5% for retinal blood vessels segmentation task.

关键词： Retinal blood vessels semantic segmentation U-Net Condition Random Field

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Survey of Weakly Supervised semantic segmentation Methods

Survey of Weakly Supervised Semantic Segmentation Methods

引用

第30届中国控制与决策会议

作者： Zheng Lu Dali Chen Dingyu Xue College of Information Science and Engineering Northeastern University

Weakly supervised image semantic segmentation is quite popular in computer vision and machine learning *** existing supervised semantic segmentation approaches require lots of pixel-based annotation for training,which are labor-and time-consuming to *** comparison,weakly supervised methods don’t need the pixel-level label,and they can do the semantic segmentation work by only using the image-level *** paper provides a review on weakly supervised semantic segmentation ***,the existing methods are ***,the main datasets and evaluating indicator are ***,the results are ***,the conclusions are given.

关键词： semantic segmentation Weakly supervised learning Deep learning Neural networks

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Extracting dispersion curves using semantic segmentation of fully convolutional networks

Extracting dispersion curves using semantic segmentation of ...

引用

信号处理在地球物理——浙江省信号处理学会2018年学术年会

作者： Tianyu Dai Jianghai Xia Ling Ning Subsurface Imaging and Sensing Laboratory Institute of Geophysics and GeomaticsChina University of Geosciences School of Earth Sciences Zhejiang University

High-frequency surface-wave methods have been widely used for surveying s-wave velocities in near surface both active and positive seismic methods.A key step in high-frequency surface-wave methods is to acquire the dispersion curves in images of surface waves in the frequency-velocity *** traditional way to acquire the dispersion curves is to identify the dispersion energy and manually pick phase velocities by following peaks at different frequencies,which is a time consuming *** propose a novel method to extract dispersion curves using semantic segmentation of fully convolutional networks(FCN) without manual *** selected Resnet34 model as a foundation network to build the FCN for dispersion energy areas *** show that the FCN is suitable and efficient for dispersion energy segmentation *** curves can be extracted from the segmented dispersion energy areas easily and accurately.

关键词： Surface waves Dispersion curves semantic segmentation Convolutional networks

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Improved RGBD semantic segmentation Using Multi-Scale Features

Improved RGBD Semantic Segmentation Using Multi-Scale Featur...

引用

第30届中国控制与决策会议

作者： Xiaoning Gao Meng Cai Jianxun Li School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Luoyang Institute of Electro-optical Equipment AVIC

RGBD semantic segmentation is a popular task in computer vision with applications in autonomous vehicles and virtual reality. This problem is challenging due to the cluttered, dense and diverse scenes. To solve the loss of context information in dense semantic scene segmentation, we propose a novel architecture built on multi-scale feature representation that contains more global and local context cues. The multi-scale features, which are generated via aggregating 3 D region features and sparse coding SIFT features extracted from multiresolution RGB and depth images, are fed into a softmax classifier to labeling each region produced by hierarchical segmentation with a predefined class, that is our final semantic scene segmentation. In addition, compared to the rough four categories predefined from the 894 pixel categories in NYUD2 dataset, we define the 40 detailed pixel classes that cover most common object categories and makes a fine-grained semantic segmentation. Extensive experiments on the standard NYUD2 benchmark demonstrate the effectiveness of our method.

关键词： semantic segmentation Hierarchical segmentation RGBD Multi-scale sparse coding

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Tire Defect Detection Based on Improved Hypernetwork semantic segmentation

Tire Defect Detection Based on Improved Hypernetwork Semanti...

引用

第43届中国控制会议

作者： Xu Li Chen Peng Xiaoyu Li Penghui Zhang Department of Automation College of Mechatronics Engineering and Automation Shanghai University

Given the complex and diverse textures of tire X-ray images, traditional manual inspections fall short of meeting production needs, making intelligent vision inspection a highly sought-after solution. This paper proposes an improved hypernetwork semantic segmentation network for detecting internal tire defects. In this network, a nested U-Net within the hypernetwork is designed to extract high-level contextual features. Furthermore, the decoder, composed of multiple dynamic blocks,effectively combines low-level and high-level information to enhance segmentation accuracy for defects of various shapes and sizes. Experimental results show that our proposed network achieves effective detection of different tire defects, particularly complex wrinkle defects in the belt layer. The final model achieved a mIoU score of 82.32%, representing a significant improvement of 13.18% compared to U-Net.

关键词： Tire defect Nested U-Net semantic segmentation Dynamic convolution block

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

semantic Pneumonia segmentation and Classification for Covid-19 Using Deep Learning Network

引用

Computers, Materials & Continua 2022年第10期73卷 1141-1158页

作者： M.M.Lotfy Hazem M.El-Bakry M.M.Elgayar Shaker El-Sappagh G.Abdallah M.I A.A.Soliman Kyung Sup Kwak Department of Mathematics Faculty of ScienceArish UniversityArish45511Egypt Department of Information Systems Faculty of Computers and InformationMansoura UniversityMansoura35511Egypt Department of Information Technology Faculty of Computers and InformationMansoura UniversityMansoura35511Egypt Centro Singular de Investigación en Tecnoloxías Intelixentes(CiTIUS) Universidade de Santiago de CompostelaSantiago de Compostela15705Spain Information Systems Department Faculty of Computers and Artificial IntelligenceBenha UniversityBenha13512Egypt Department of Information and Communication Engineering Inha UniversityIncheon22212Korea

Early detection of the Covid-19 disease is essential due to its higher rate of infection affecting tens of millions of people,and its high number of deaths also by 7%.For that purpose,a proposed model of several stages was *** first stage is optimizing the images using dynamic adaptive histogram equalization,performing a semantic segmentation using DeepLabv3Plus,then augmenting the data by flipping it horizontally,rotating it,then flipping it *** second stage builds a custom convolutional neural network model using several pre-trained ***,the model compares the pre-trained data to the new output,while repeatedly trimming the best-performing models to reduce complexity and improve memory *** experiments were done using different techniques and ***,the proposed model achieved an average accuracy of 99.6%and an area under the curve of 0.996 in the Covid-19 *** paper will discuss how to train a customized intelligent convolutional neural network using various parameters on a set of chest X-rays with an accuracy of 99.6%.

关键词： SARS-COV2 COVID-19 pneumonia deep learning network semantic segmentation smart classification

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

A feasibility study of knee joint semantic segmentation on 3D MR images

A feasibility study of knee joint semantic segmentation on 3...

引用

第十七届中国体视学与图像分析学术会议

作者： SHEN Le LU Qian WU Sha YI Yi SUN Yunda LI Min QIU Qian ZHENG Zhuozhao CAI Xu ZHANG Li Key Laboratory of Particle & Radiation Imaging (Tsinghua University) Ministry of Education Department of Engineering Physics Tsinghua University Nuctech Company Limited Beijing Tsinghua Changgung Hospital

The segmentation of knee joint is of great significance for diagnosis and treatment of knee *** various anatomical structures are involved in the 3 D MRI volume,manual delineation is timeconsuming and *** segmentation of the whole knee joint requires less human efforts,additionally can improve the quality of arthritis diagnosis and treatment effect by describing the details more *** works of knee joint segmentation methods in the literature focus on only one or a few structures,while segmentation of whole knee joint remains a *** this paper,we study the feasibility of knee joint semantic segmentation on MR images based on a 3 D SegResnet and deal with the following challenges:1) end-to-end segmentation of 15 anatomical structures,including bone and soft tissue,of the whole knee on MR images;2) robust segmentation of small structures such as the anterior cruciate ligament,accounting for about 0.036% voxels of the volume data;3) Fast and accurate inference strategy on 3 D medical *** compare our method against nnUNet,a current state-ofart approach on medical image *** is verified that the average segmentation accuracy of our method is 92.92% compared with 91.45% of *** coefficient of each anatomical structure are shown in Figure 1,our method performs better on small structures especially cartilages and cruciate *** Whole knee segmentation result using our method is shown in Figure 2.

关键词： Knee Osteoarthritis semantic segmentation

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Multidimensional attention and multiscale upsampling for semantic segmentation

引用

Journal of Measurement Science and Instrumentation 2022年第1期13卷 68-78页

作者： LU Zhongda ZHANG Chunda WANG Lijing XU Fengxia School of Mechanical and Electrical Engineering Qiqihar University Qiqihar 161000 China Heilongjiang Province Collaborative Innovation Center for Intelligent Manufacturing Equipment Industrialization Qiqihar 161000 China

semantic segmentation is for pixel-level classification tasks,and contextual information has an important impact on the performance of *** order to capture richer contextual information,we adopt ResNet as the backbone network and designs an encoder-decoder architecture based on multidimensional attention(MDA)module and multiscale upsampling(MSU)*** MDA module calculates the attention matrices of the three dimensions to capture the dependency of each position,and adaptively captures the image *** MSU module adopts parallel branches to capture the multiscale features of the images,and multiscale feature aggregation can enhance contextual information.A series of experiments demonstrate the validity of the model on Cityscapes and Camvid datasets.

关键词： semantic segmentation attention mechanism multiscale feature convolutional neural network(CNN) residual network(ResNet)

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：