没有合适的资源?快使用搜索试试~ 我知道了~
人工智能-基于多尺度神经网络的小区域图像修复.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 138 浏览量
2022-06-27
23:48:39
上传
评论
收藏 4.74MB PDF 举报
温馨提示
试读
64页
人工智能-基于多尺度神经网络的小区域图像修复.pdf
资源推荐
资源详情
资源评论
致谢
时间转瞬即逝,硕士生涯转眼已近尾声,回望三年间的学习经历,感触颇深,
老师的传道授业,同学的帮扶鼓励,家人朋友的关爱照拂以及自己在学业上的提
升,这些都将在日后深深地影响我。在这里,我向这些给过我莫大帮助的人表达
我最真挚的谢意。
首先,非常感谢陈伟教授在研究生期间给我的无私帮助与关怀,老师严谨的
治学态度、精益求精的工作作风、以及孜孜不倦的科研精神时刻鼓励着我前行,
为我树立了学习的榜样。三年来,从研究方向的确定、实验的完成到最终的写作,
陈老师时刻关注并指导着我,也正是在老师默默地付出和帮助下,我才能够在科
研道路上不断取得进步。同时陈老师在我职业规划中也给出了中肯的建议,这些
无私的付出以及为人师表的品行将令我受益终身。在此,我再次向尊敬的陈伟老
师表达我最诚挚的谢意。
其次,我要感谢我的家人和朋友,是他们的支持和包容才成就了今天的我。
感谢各位同学的鼓励,在我遇到挫折和困境的时候,是他们不断地开导鼓励我,
让我坚持努力,是他们为我营造了良好的学习科研环境。
最后,感谢各位参与我论文评审和答辩的老师,你们就像一面镜子,让我对
自己的研究成果有了深刻的审视,能够让我更加清楚自己以后需要进步的方向。
感谢各位老师。
万方数据
I
摘 要
近十年来,计算机视觉在图像分类、目标检测、图像分割等图像处理任务上
取得了巨大的进步,深度网络的性能在这些任务中有了很大的提升,为新的图像
处理任务奠定了基础。尽管基于生成对抗网络(Generative Adversarial Network,
GAN)的图像修复方法近年来在准确性和速度上取得了突破,但是由于内存硬
件条件限制和 GAN 网络训练不平衡,对于高分辨率图像,修复的区域会显得模
糊,存在清晰的边界线,难以修复高频细节。其次,研究发现,这些基于卷积神
经网络的图像修复方法经常生成的边界伪影、变形结构和不清晰的纹理,除了是
网络模型本身的影响,还可能是由于卷积神经网络无法将远距离的图像信息和图
像补孔之间的长期相关性进行建模引起的。同时,图像修复在实际生活中的应用
也十分广泛,例如删除图像中不必要的行人并获得背景恢复的真实感的问题。这
个问题很有挑战性,因为缺少真实的输出样本去定义重构损失。
为解决上述问题。本文基于生成对抗网络,进行如下图像修复的研究:
1)基于多尺度神经网络的高分辨率图像修复:为了修复高频细节,该网络
包含内容重构网络和纹理细节恢复网络。内容重构网络使用 VGG-16 对输入图
像进行多尺度特征提取,并利用四组不同速率的空洞卷积提取多尺度感受野的图
像特征,然后根据提取的特征重构输入图像。纹理细节恢复网络在 SRGAN
(Super-Resolution Generative Adversarial Network,SRGAN)网络的基础上加入
了空洞卷积,并使用跳跃连接融合多尺度特征,进一步恢复更精细的纹理细节,
该方法能够有效的解决结构扭曲和纹理模糊问题,提高图像修复质量。
2)面向小区域图像修复的行人去除:针对滤除破坏图像美感的占据较小背
景区域的多余行人的问题,文章融合已有的实例分割、图像修复的研究成果搭建
行人去除网络框架。并构建掩码数据集解决,训练过程中缺少真实的输出样本去
定义重构损失的问题。提出的网络框架可以轻松的筛选出背景中的一个或多个行
人,并将其去除。
3)具有语境注意的图像修复算法:针对现有语境注意层忽略了生成的图像
块之间的相关性,可能会导致最终结果缺乏延展性和连续性的缺陷。文章基于具
有新颖的图像语境注意层的统一前馈生成网络进行了改进。通过改变语境注意
层的传输机制加强了生成的图像块之间的相关性。并在语境注意层的输出端通过
四组不同速率的空洞卷积进行多尺度感受野的特征聚合,这四组不同速率的空洞
卷积保证了最终重建特征的结构与环境的一致性。改进的基于语境注意的图像修
复网络模型能够加强识别周围环境的图像结构的能力,能够自适应地借用周围环
境的信息来帮助注意图的合成和生成。
关键词:图像修复;生成对抗网络;多尺度神经网络;语境注意
万方数据
II
Abstract
In recent ten years, computer vision has made great progress in image processing
tasks such as image classification, target detection and image segmentation, and the
performance of deep network has been greatly improved in these tasks, laying a
foundation for new image processing tasks. Although the generated against Network
based (Generative Adversarial Network, GAN) image restoration method in recent
years have made a breakthrough in the accuracy and speed, but the memory hardware
conditions and GAN unbalanced Network training, for high resolution image, repair
area will appear blurred, there are clear lines, hard to fix the high frequency detail.
Second, the study found that the image restoration method based on convolution
neural networks often generated boundary artifact, deformation structure and clear
texture, as well as the influence of the network model itself, also may be due to the
convolutional neural network to remote image information and image to fill holes
caused by long-term association between modeling. At the same time, image repair is
widely used in real life, such as deleting unnecessary pedestrians in the image and
obtaining the sense of reality of background restoration. This problem is challenging
because there is a lack of real output samples to define the refactoring losses.
To solve the above problems, based on the generated antagonistic network, this
paper conducts the following research on image repair:
1) High-resolution image repair based on multi-scale neural network: In order to
repair high-frequency details, this network includes content reconstruction network
and texture detail recovery network. The content reconstruction network USES
VGG-16 to extract the multi-scale features of the input image, and USES four groups
of cavity convolution with different rates to extract the image features of the
multi-scale receptive field, and then reconstructs the input image according to the
extracted features. Based on the Network of SRGAN (Super-Resolution Adversarial
Network), the texture detail restoration Network added void convolution, and restored
the texture details by using skip connection and multi-scale feature fusion. This
method can effectively solve the problems of structure distortion and texture blur, and
improve the quality of image restoration.
2) Pedestrian removal for image restoration in small areas: Aiming at the
problem of filtering out redundant pedestrians occupying small background areas that
damage image aesthetics, this paper integrates existing example segmentation and
万方数据
III
research results of image restoration to build a pedestrian removal network framework.
And build mask data set to solve the problem of lack of real output samples to define
the reconstruction loss in the training process. The proposed network framework can
easily screen one or more pedestrians in the background and remove them.
3) Image repair algorithm with context attention: The existing context attention
layer ignores the correlation between generated image blocks, which may lead to the
defects of malleability and continuity of the final result. In this paper, we improve the
unified feed forward generation network based on the novel image context attention
layer. The correlation between the generated image blocks is enhanced by changing
the transmission mechanism of the context attention layer. At the output end of the
contextual attention layer, the feature aggregation of multi-scale receptive fields is
carried out through four groups of cavity convolution with different rates, which
ensures the consistency between the structure of the final reconstructed features and
the environment. The improved image repair network model based on contextual
attention can enhance the ability to recognize the image structure of the surrounding
environment and can adaptively borrow the information of the surrounding
environment to help the synthesis and generation of the image.
Keywords: Image inpainting; generative adversarial network; multi-scale neural
network; contextual attention
万方数据
IV
目 录
U 摘 要 U .......................................................................................................................... I
U 目 录 U ....................................................................................................................... IV
U 图清单 U .................................................................................................................... VIII
U 表清单 U ....................................................................................................................... IX
U 变量注释表 U ................................................................................................................. X
U1 绪论 U ..........................................................................................................................1
U1.1 研究背景与意义 U ...................................................................................................1
U1.2 国内外研究现状 U ...................................................................................................2
U1.3 本文主要工作 U .......................................................................................................6
U1.4 本文组织结构 U .......................................................................................................7
U2 基础理论 U ..................................................................................................................8
U2.1 生成对抗网络的网络结构 U ...................................................................................8
U2.2 深度卷积生成对抗网络 U ..................................................................................... 11
U2.3 生成对抗网络的机遇与挑战 U .............................................................................19
U3 基于多尺度神经网络的图像修复及应用 U ............................................................22
U3.1 引言 U .....................................................................................................................22
U3.2 基于多尺度神经网络的图像修复模型 U .............................................................23
3.3 面向小区域图像修复的行人去除.......................................................................27
U3.4 实验结果与分析 U .................................................................................................30
U3.5 本章小结 U .............................................................................................................34
U4 具有语境注意的图像修复算法 U ............................................................................35
U4.1 引言 U .....................................................................................................................35
U4.2 基于语境注意的图像修复模型 U .........................................................................36
U4.3 实验结果与分析 U .................................................................................................40
U4.4 本章小结 U .............................................................................................................42
U5 结论 U ........................................................................................................................44
U5.1 总结 U .....................................................................................................................44
U5.2 展望 U .....................................................................................................................45
U 参考文献 U ....................................................................................................................46
U 作者简历 U ....................................................................................................................53
万方数据
剩余63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