检索结果-内蒙古大学图书馆

16th Asian Conference on Computer Vision (ACCV)

作者： Sun, Yongqing Wu, Xiaomeng Bandoh, Yukihiro Kitahara, Masaki NTT Comp & Data Sci Labs Tokyo Japan NTT Commun Sci Labs Kyoto Japan

ISBN: (纸本)9783031270659;9783031270666

aerial image segmentation is an essential problem for land management which can be used for change detection and policy planning. However, traditional semantic segmentation methods focus on single-perspective images in road scenes, while aerial images are top-down views and objects are of a small size. Existing aerial segmentation methods tend to modify the network architectures proposed for traditional semantic segmentation problems, yet to the best of our knowledge, none of them focus on the noisy information present in the aerial images. In this work, we conduct an investigation on the effectiveness of each channels of the aerial image on the segmentation performance. Then, we propose a disentangle learning method to investigate the differences and similarities between channels and images, so that potential noisy information can be removed for higher segmentation accuracy.

关键词： aerial image segmentation Disentangle learning Semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Multilevel thresholding aerial image segmentation using comprehensive learning-based Snow ablation optimizer with double attractors

引用

EGYPTIAN INFORMATICS JOURNAL 2024年 27卷

作者： Abd Elaziz, Mohamed Al-qaness, Mohammed A. A. Ibrahim, Rehab Ali Ewees, Ahmed A. Shrahili, Mansour Zagazig Univ Fac Sci Dept Math Zagazig Egypt Zhejiang Normal Univ Coll Phys & Elect Informat Engn Jinhua 321004 Peoples R China Zhejiang Inst Optoelect Jinhua 321004 Peoples R China Damietta Univ Dept Comp Dumyat 34517 Egypt King Saud Univ Coll Sci Dept Stat & Operat Res POB 2455 Riyadh 11451 Saudi Arabia Emirates Int Univ Coll Engn & Informat Technol Sanaa 16881 Yemen Galala Univ Fac Comp Sci & Engn Suze 435611 Egypt Ajman Univ Artificial Intelligence Res Ctr AIRC Ajman 346 U Arab Emirates Lebanese Amer Univ Dept Elect & Comp Engn Byblos 135053 Lebanon Middle East Univ MEU Res Unit Amman 11831 Jordan

aerial photography is a remote sensing technique used for target detection, enabling both qualitative and quantitative analysis. The segmentation process is considered one of the most important processes to improve the analysis of aerial images. In this study, we introduce an alternative multilevel threshold image segmentation based on a modified Snow ablation optimizer (SAO) algorithm. This modification is conducted using the strengths of Comprehensive learning and Double attractors which aims to enhance the exploration and exploitation abilities of the SAO during the process of discovering the optimal threshold levels that are used to segment the aerial photography image. To validate the quality of the modified version of SAO, named DCSAO, a set of experimental series is conducted using the CEC2022 benchmark function and sixteen aerial images at different threshold levels. In addition, we compared the results of DCSAO with different wellknown Metaheuristic techniques. The results show the superior performance of DCSAO in comparison to other algorithms according to the performance metrics.

关键词： aerial image segmentation Snow ablation optimizer Comprehensive learning Double attractors

来源：评论

学校读者我要写书评

暂无评论

Evolutionary NAS for aerial image segmentation with gene expression programming of cellular encoding

引用

NEURAL COMPUTING & APPLICATIONS 2022年第17期34卷 14185-14204页

作者： Broni-Bediako, Clifford Murata, Yuki Mormille, Luiz H. Atsumi, Masayasu Soka Univ Fac Sci & Engn Hachioji Tokyo 1928577 Japan

Recently, neural architecture search (NAS) has gained a lot of attention as a tool for constructing deep neural networks automatically. NAS methods have successfully found convolutional neural networks (CNNs) that exceed human expert-designed networks on image classification in computer vision. However, there are growing demands for semantic segmentation in several areas including remote sensing image analysis. In this paper, we introduce an evolutionary NAS method for semantic segmentation of high-resolution aerial images. The proposed method leverages the complementary strengths of gene expression programming and cellular encoding to develop an encoding scheme, called symbolic linear generative encoding (SLGE), for evolving cells (directed acyclic graphs) as building-blocks to construct modularized encoder-decoder CNNs via an evolutionary process. SLGE can evolve cells with multi-branch and shortcut connections similar to the Inception-ResNet-like modules which can improve training and inference performance in deep neural networks. In experiments, we demonstrate the effectiveness of the proposed method on the challenging ISPRS Vaihingen, Potsdam and UAVid semantic segmentation benchmarks. Compared with recent state-of-the-art systems, our network, dubbed SLGENet, improves the overall accuracy performance on Vaihingen and Potsdam;and achieves a competitive overall accuracy on UAVid using fewer parameters. Our method achieves promising results in a little time of 2.5 GPU days.

关键词： Evolutionary neural architecture search Convolutional neural networks Remote sensing aerial image segmentation

来源：评论

学校读者我要写书评

暂无评论

BOUNDARY-AWARE BIAS LOSS FOR TRANSFORMER-BASED aerial image segmentation MODEL 47

BOUNDARY-AWARE BIAS LOSS FOR TRANSFORMER-BASED AERIAL IMAGE ...

引用

47th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Zhang, Yan Jiang, Xue Liu, Siqi Hu, Bo Gao, Xinbo Chongqing Univ Posts & Telecommun Chongqing Key Lab Image Cognit Chongqing Peoples R China Peking Univ Sch Software & Microelect Beijing Peoples R China

ISBN: (纸本)9781665405409

Inspired by the tremendous success of the transformer-based model in natural language processing (NLP), many efforts introduce the transformer-based model into the image processing tasks. However, naive transformer models have to downsample the image resolution to satisfy computational restrictions, thus discarding the local information, which is catastrophic for high-performance remote sensing image segmentation. Hence, this paper proposes a novel trainable boundaryaware bias loss function to enhance transformer-based models of extracting local information. On the Challenging ISPRS Potsdam dataset, two representative transformer-based models achieve remarkable performance improvements, proving the effectiveness of the proposed method.

关键词： aerial image segmentation Loss Function Transformer Deep Learning Remote Sensing

来源：评论

学校读者我要写书评

暂无评论

DPNET: END-TO-END aerial image segmentation VIA DEFORMABLE POINT NETWORK

DPNET: END-TO-END AERIAL IMAGE SEGMENTATION VIA DEFORMABLE P...

引用

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

作者： Guo, Yiyou Qin, Zheyun Liu, Yikun Yang, Yongtai Lu, Xiankai Xie, Huan Tong, Xiaohua Tongji Univ Coll Surveying & Geoinformat Shanghai 200092 Peoples R China Shangdong Univ Sch Software Jinan 250101 Peoples R China Wilfrid Laurier Univ Fac Sci Waterloo ON Canada

ISBN: (数字)9781665427920

ISBN: (纸本)9781665427920

aerial image segmentation segmentation faces intrinsic foreground-background imbalance and background clutter distraction. To guide the segmentation model to learn more discriminative foreground ability and more invariant background representation features, we design a Deformable Point Network (DPNet). It is an end-to-end segmentation network and consists of a multi-head deformable attention module that simultaneously considers foreground object information and background suppression. Specifically, we first employ a feature pyramid network to aggregate multiple-layer features to handle scale variants. And then, we further investigate deformable convolution to select some representative points for each layer and propose a differential module to implement it automatically instead of traditional dense fusion. Moreover, we incorporate the multi-head mechanism in the feature fusion to focus on the key contents from different representation regions. Experimental results on the representative iSAID, Vaihingen, and Postdam datasets demonstrate that our DPNet achieves competitive performance. Also, the multiple-head deformable attention facilitates the network convergence significantly.

关键词： aerial image segmentation Multi-head deformable attention Feature enhancement

来源：评论

学校读者我要写书评

暂无评论

DHT: Deformable Hybrid Transformer for aerial image segmentation

引用

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2022年 19卷 1页

作者： Zhang, Yan Gao, Xiyuan Duan, Qingyan Yuan, Lin Gao, Xinbo Chongqing Univ Posts & Telecommun Chongqing Key Lab Image Cognit Chongqing 400065 Peoples R China Xidian Univ State Key Lab Integrated Serv Networks Xian 710071 Peoples R China

Due to the strong ability to model global information, the transformer-based methods have shown remarkable improvements in the image segmentation tasks. However, the self-attention mechanism in the transformer is computationally expensive and relies on pretrained parameters. Moreover, the transformer method is weak in modeling local information, which is unfavorable for accurately segmenting objects from high-resolution aerial images. To this end, an efficient deformable orientational self-attention (DoA) is proposed to simultaneously extract the global information and the local information. Besides, for parameter efficiency, we design a depthwise channel self-attention (DcA) to model the contextual information among channels. Combining with the DoA and DcA, we propose the deformable hybrid transformer (DHT) to perform high-quality object segmentation on aerial images. Experiments on the International Society for Photogrammetry and Remote Sensing (ISPRS) Potsdam and Wuhan University (WHU) building datasets illustrate that the proposed DHT can not only achieve the state-of-the-art (SOTA) results but also markedly reduce the dependence of the transformer on pretrained parameters.

关键词： Deformable models image segmentation Computational modeling Buildings Object segmentation Transformers Sensors aerial image segmentation deep learning remote sensing self-attention transformer

来源：评论

学校读者我要写书评

暂无评论

aerial image segmentation

AERIAL IMAGE SEGMENTATION

引用

作者： Abdulkareem Ali Althwaini Laurentian University

学位级别：硕士

image segmentation plays a vital role in applications such as remote sensing. For this example, remote sensing, aerial image segmentation is a special case of image segmentation. There are some unique features of aerial images, like noise in natural landscapes, which need to be addressed in order to obtain an optimal solution. Bushes and rocks are examples of landscape features with diverse and variable pixel values that need to be distinguished by the segmentation process. Smoothing filters present a common solution to address the problem of noise in images, as does aerial image segmentation. There are several image segmentation techniques used for aerial image segmentation. Some of these techniques are more sensitive to noise problems, and are necessary to discriminate between different smoothing filters. In this thesis, a number of different aspects of aerial image segmentation and their solutions are explained. In addition to this, a novel smoothing filter is introduced and compared with other methods using different segmentation techniques. Finally, all of the previous points are applied to a real world problem.

关键词： aerial image segmentation smoothing filters k-means classifying c-means fuzzy classifying eigenvalues singular values and climate change detection

来源：评论

学校读者我要写书评

暂无评论

Multi - Feature Fusion aerial image segmentation in Complex Background 3

Multi - Feature Fusion Aerial Image Segmentation in Complex ...

引用

3rd International Conference on Vision, image and Signal Processing (ICVISP)

作者： Yang, Rui Qian, Xiao Jun Zhang, Bing Bing Nanjing Normal Univ Dept Comp Sci & Technol Nanjing Peoples R China

ISBN: (纸本)9781450376259

The hybrid method of initial partitioning and region merging is widely used in aerial image segmentation. The existing initial partitioning methods are the watershed transform of the Edge Strength Map (ESM) of aerial images. Therefore, if watershed algorithm is used in images with discontinuous edges and lots of noise, it will be easy to produce "improper segmentation". In order to form high-quality initial partitions, we propose a new MRF (YMRF) image segmentation method from the perspective of fully exploiting the image spatial information. The key points of region merging are region similarity measurement, merging process and merging stopping moment, but the problem of region label selection is ignored after the region pair which will be merged is selected. So, we propose a kind of image scene to reflect the necessity of paying attention to this problem and develop a region merging label selection mechanism for the image scene. To solve the problem that merging stopping moment tends to form the result with high homogeneity in the domain, we propose a optimal merging state, which can weaken the homogeneity in the domain. Experimental results show that our algorithm is more effective than the existing methods, when they are used in our unique dataset.

关键词： aerial image segmentation ESM Watershed algorithm MRF Region Merging Region similarity measurement Region merging label selection mechanism Merging process Optimal merging state

来源：评论

学校读者我要写书评

暂无评论

Multi-Feature Fusion aerial image segmentation in Complex Background 2019

Multi-Feature Fusion Aerial Image Segmentation in Complex Ba...

引用

Proceedings of the 3rd International Conference on Vision, image and Signal Processing

作者： Rui Yang Xiao jun Qian Bing bing Zhang Department of Computer Science and Technology Nanjing Normal University Nanjing China

ISBN: (纸本)9781450376259

关键词： aerial image segmentation Merging process Optimal merging state ESM Region similarity measurement Region Merging MRF Watershed algorithm Region merging label selection mechanism

来源：评论

学校读者我要写书评

暂无评论

Context Enhanced Traffic segmentation: traffic jam and road surface segmentation from aerial image 14

Context Enhanced Traffic Segmentation: traffic jam and road ...

引用

14th IEEE image, Video, and Multidimensional Signal Processing Workshop (IVMSP)

作者： Wang, Yubo Wang, Zhao Nakano, Yuusuke Nishimatsu, Ken Hasegawa, Katsuya Ohya, Jun Waseda Univ Tokyo Japan NTT Labs Tokyo Japan JAXA Sagamihara Kanagawa Japan

ISBN: (纸本)9781665478229

Traffic jam detection and density estimation of aerial images have been widely utilized in various scenarios, such as vehicle routing and city management. Rather than directly detecting traffic jams or estimating density, traffic condition analysis based on traffic jam segmentation could yield more accurate results. Therefore, we propose a Context Enhanced Traffic segmentation Model to simultaneously segment the traffic jam parts and road surface. However, there are two critical issues for traffic jam segmentation in aerial images: one is the scale variation problem and the other is the difficulty of accurately segmenting ambiguous traffic jam boundaries. Thus, we design a traffic estimation module to handle the scale variation problem and present a context attention module to enhance the boundary of traffic jam segmentation. Experimental results demonstrate the superiority of our proposed method.

关键词： traffic jam segmentaion aerial image segmentation self-attention mechanism

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：