没有合适的资源?快使用搜索试试~ 我知道了~
人工智能-基于卷积神经网络的高分辨率遥感影像建筑物提取方法研究.pdf
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 197 浏览量
2022-06-28
19:05:35
上传
评论
收藏 3.85MB PDF 举报
温馨提示
试读
78页
人工智能-基于卷积神经网络的高分辨率遥感影像建筑物提取方法研究.pdf
资源推荐
资源详情
资源评论
致谢
时光荏苒,岁月如梭,行笔至此,与中国矿业大学的七年缘分行将告一段落,
在镜湖的滋润下,我由一介懵懂无知的求学者,成长为可以独自背起行囊上路的
书客,矿大给予了我生根发芽的土壤和广阔的天地,使我能够充分地汲取养分,
我在这里不仅收获了学术和能力上的提升,更是我人生路上的启迪与顿悟。感谢
矿大,无论今后飘向何方,寓旅何处,这些必将成为我终生受益的财富。
首先我要感谢三年研究生生涯中的两位导师,张秋昭副教授和自然资源部国
土遥感卫星应用中心的王光辉高级工程师,。张教授认真负责,温和谦逊,对学
生教导有方,在我研究生初期,张教授对我的包容、鞭策和鼓励使我受益良多,
并且给予我去自然资源部国土遥感卫星应用中心为期两年的学习机会,这次机会
极大地开阔了我的视野,为我的后续学习奠定了基础。在我实习的这段时间里,
张教授经常以电话或邮件的形式对我的学习和生活进行关心,其敬业的工作精神
和无微不至的关怀我将终身难忘。然后我要感谢王光辉高工,感谢王老师给我提
供了很好的科研环境,对我悉心栽培,让我参与了很多项目,使我的各方面能力
得到了极大的提升。王老师常常在百忙之中抽空指导我的论文和工作,令我受益
匪浅,对于我出现的错误与困惑,都耐心点拨。两位导师的人格魅力,都将成我
在前进道路上不灭的指路明灯。
感谢我的师兄杨威、我的同门王耀兴和马春在我学习和生活上的陪伴与帮助,
在北京实习的日子里,多亏了你们处理我在学校的大小事宜,这些都给予了我极
大的支持与帮助,我没齿难忘。
感谢我的父母,你们不仅是我生活上的依靠,更是我的朋友、我的听众、我
的拥趸,为我付出了巨大牺牲也给予我无穷的力量,感谢你们将我培养成人,在
未来的岁月里我们将继续携手共进。
最后,再一次感谢我的母校,感谢所有帮助过我的人,我无法将你们一一罗
列,但你们对我的支持与帮助都将深深地刻在我的脑海中,我永远心怀感激。
万方数据
I
摘 要
随着高分辨率遥感影像来源不断丰富,影像获取越来越容易,海量数据中
蕴含的地物信息已经在地图测绘、资源勘探、环境监控、国土资源调查、变化
检测和灾害评估等多个领域得到了广泛应用。建筑物作为城镇地物信息的重要
组成部分,人工提取和更新这些信息浪费大量人力、物力和财力,因此如何能
够精确、快速、自动化地从卫星影像中提取建筑物信息,成为遥感领域的重点
研究方向之一。在过去一些方法中,从遥感影像中提取建筑物大多基于人工设
计的特征,如纹理、光谱、阴影和形状等,但是由于建筑物的屋顶覆盖物不
同、结构走向不一致、空间分布各异等特征,使得这些提取方法适用性不强。
近几年,深度学习的迅速发展使得众多学者将其运用到遥感影像处理中,并且
取得了一定的成果。因此本论文对深度学习语义分割领域相关算法进行了深入
研究与分析,主要的研究工作如下:
(1)阐述了卷积神经网络的基本原理以及训练方法。解决了高分辨率遥感
影像数据匮乏的问题,详细介绍了构建遥感影像建筑物数据集的相关操作,主要
分为图像预处理和图像增广两个部分,其中包括图像去噪、图像增强、数据扩充
和数据集划分等,增强了数据集的丰富性和多样性。
(2)研究和探索了现有的几种比较流行的网络结构的特点,并在它们基础
上进行改进,设计了一种多分辨率特征融合的语义分割网络 MRNet。该网络由
并行多分辨率子网结构和多尺度特征融合结构组成,这两种结构可以实现不同层
级的特征并行训练,关注不同尺度的建筑物,使信息融合更加多样,加强了不同
分辨率特征图信息的流动,有利于丢失信息的重建,比金字塔结构更加高效。最
后提出一种边界损失函数,使模型在训练过程中能够更多地关注建筑物边界,有
效地改善了边界锯齿问题。
(3)基于条件 生成对抗网 络的思想,设 计 了一种生成分 割对抗网络
ResUNet-GAN,以对抗的方式实现分割任务。其中生成网络 ResUNet 在编码器
阶段使用残差模块,有效地解决了训练过程梯度消失的问题,同时加深了网络层
数,有利于特征的提取,在中间加入跳跃连接,有利于多尺度信息融合,最后再
加入判别网络 SimNet 进行交替训练,实验证明了判别器对于分割网络的优化起
到了一定的效果,分割对抗架构在一定程度上能够提升建筑物提取的精度。
关键词:遥感影像;建筑物提取;深度学习;语义分割;生成对抗网络
万方数据
II
Abstract
As the sources of high-resolution remote sensing images continue to increase,
image acquisition is becoming easier. The feature information contained in massive
data has been widely used in many fields such as map mapping, resource exploration,
environmental monitoring, land and resources investigation, change detection and
disaster assessment. Buildings are an important part of urban feature information.
Manually extracting and updating buildings wastes a lot of human, material and
financial resources. Therefore, how to accurately, quickly and automatically extract
building information from satellite images has become one of key research directions
in the field of remote sensing. In the past, the methods for extracting buildings from
remote sensing images were mostly based on artificially designed features, such as
texture, spectrum, shadows, and shapes. However, due to different roof coverings,
inconsistent structural trends, and different spatial distributions of buildings, which
makes these extraction methods unsuitable. In recent years, the rapid development of
deep learning has led many scholars to apply it to remote sensing image processing,
and has achieved certain results. Therefore, this paper makes deep research and analysis
on related algorithms in the field of deep learning semantic segmentation. The main
research work is as follows:
(1) This article introduces the basic principles and training methods of
convolutional neural networks. In order to solve the problem of lack of high-resolution
remote sensing image data, the related operations of building a remote sensing image
building data set are introduced in detail. It is mainly divided into image preprocessing
and image augmentation, including image denoising and image Enhancements, data
expansion, and data set partitioning enhance the richness and diversity of the dataset.
(2) In this paper, the characteristics of several existing popular network structures
are studied and explored. Based on this, a multi-resolution feature fusion semantic
segmentation network named MRNet is designed. The network consists of a parallel
multi-resolution subnet structure and a multi-scale feature fusion structure. These two
structures can achieve parallel training of features at different levels, focusing on
buildings of different scales, making the information fusion more diverse. The
information flow of feature maps with different resolutions is strengthened, which is
beneficial to the reconstruction of lost information and is more efficient than the
pyramid structure. Finally, a boundary loss function is proposed, which enables the
万方数据
III
model to pay more attention to the building boundary during the training process, which
effectively improves the boundary jagged problem.
(3) Based on the idea of conditional generative adversarial network, this paper
designs a generative-segmentation adversarial network named ResUNet-GAN to
achieve the segmentation task in an adversarial manner. Generating network ResUNet
uses the residual module in the encoder stage, which effectively solves the problem of
gradient disappearance during the training process and deepens the number of network
layers at the same time, which is conducive to feature extraction. Adding skip
connections in the middle is beneficial to multi-scale information fusion. Finally,
adding the discriminative network SimNet is used for alternate training. Experiments
show that the discriminator has a certain effect on the optimization of the segmentation
network, and the segmentation adversarial architecture can improve the accuracy of
building extraction to a certain extent.
Keywords: remote sensing image; building extraction; deep learning; semantic
segmentation; generating adversarial network
万方数据
IV
目 录
摘 要 ....................................................................................................................... I
目 录 .................................................................................................................... IV
图清单 ................................................................................................................. VIII
表清单 ......................................................................................................................X
1 绪论 ...................................................................................................................... 1
1.1 选题背景与研究意义 ....................................................................................... 1
1.2 国内外研究现状 ............................................................................................... 2
1.3 研究内容 ........................................................................................................... 6
1.4 论文结构 ........................................................................................................... 7
2 卷积神经网络基本原理 ....................................................................................... 8
2.1 卷积神经网络结构 ........................................................................................... 8
2.2 卷积神经网络训练方法.................................................................................. 12
2.3 经典卷积神经网络介绍.................................................................................. 19
2.4 本章小结 ......................................................................................................... 21
3 实验数据预处理与增广 .......................................... 23
3.1 数据集介绍 ..................................................................................................... 23
3.2 数据预处理 ..................................................................................................... 25
3.3 数据增广 ......................................................................................................... 27
3.4 本章小结 ......................................................................................................... 28
4 基于多分辨率特征融合的语义分割网络 .......................................................... 29
4.1 引言 ................................................................................................................ 29
4.2 语义分割网络结构介绍.................................................................................. 29
4.3 多分辨率特征融合的语义分割网络构建 ....................................................... 34
4.4 实验设置与结果分析 ..................................................................................... 38
4.5 本章小结 ......................................................................................................... 46
5 基于生成对抗网络的语义分割方法 .................................................................. 47
5.1 引言 ................................................................................................................ 47
5.2 生成对抗网络介绍 ......................................................................................... 47
5.3 生成对抗网络的原理 ..................................................................................... 48
5.4 基于 ResUNet-GAN 的生成式对抗网络 .......................................................... 51
万方数据
剩余77页未读,继续阅读
资源评论
programcx
- 粉丝: 40
- 资源: 13万+
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功