检索结果-南通市图书馆

Solar Image Cloud Removal based on Improved pix2pix Network

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第12期73卷 6181-6193页

作者： Xukun Zhang Wei Song Ganghua Lin Yuxi Shi School of Information Engineering Minzu University of ChinaBeijing100081China Key Laboratory of Solar Activity Chinese Academy of Sciences(KLSACAS)Beijing100101China National Language Resource Monitoring and Research Center of Minority Languages Minzu University of ChinaBeijing100081China National Astronomical Observatories Chinese Academy of Sciences(NAOCCAS)Beijing100101China Department of Electrical and Computer Engineering New Jersey Institute of TechnologyNewarkNew Jersey07102USA

In ground-based observations of the Sun,solar images are often affected by appearance of thin clouds,which contaminate the images and affect the scientific results from data *** this paper,the improved pixel to pixel Network(pix2pix)network is used to convert polluted images to clear images to remove the cloud shadow in the solar *** adding attention module to the model,the hidden layer of pix2pix model can infer the attention map of the input feature vector according to the input feature *** then,the attention map is multiplied by the input feature map to give different weights to the hidden features in the feature map,adaptively refine the input feature map to make the model pay attention to important feature information and achieve better recovery *** order to further enhance the model’s ability to recover detailed features,perceptual loss is added to the loss *** model was tested on the full disk H-alpha images datasets provided by Huairou Solar Observing Station,National Astronomical *** experimental results show that the model can effectively remove the influence of thin clouds on the picture and restore the details of solar *** peak signal-to-noise ratio(PSNR)reaches 27.3012 and the learned perceptual image patch similarity(LPIPS)reaches 0.330,which is superior to the existed dehaze algorithms.

关键词： pix2pix decontamination full-disk H-alpha images

Image to Image Translation Based on Differential Image pix2pix Model

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2023年第10期77卷 181-198页

作者： Xi Zhao Haizheng Yu Hong Bian College of Mathematics and System Sciences Xinjiang UniversityUrumqi830017China School of Mathematical Sciences Xinjiang Normal UniversityUrumqi830017China

In recent years,pix2pix,a model within the domain of GANs,has found widespread application in the field of image-to-image ***,traditional pix2pix models suffer from significant drawbacks in image generation,such as the loss of important information features during the encoding and decoding processes,as well as a lack of constraints during the training *** address these issues and improve the quality of pix2pixgenerated images,this paper introduces two key ***,to reduce information loss during encoding and decoding,we utilize the U-Net++network as the generator for the pix2pix model,incorporating denser skip-connection to minimize information ***,to enhance constraints during image generation,we introduce a specialized discriminator designed to distinguish differential images,further enhancing the quality of the generated *** conducted experiments on the facades dataset and the sketch portrait dataset from the Chinese University of Hong Kong to validate our proposed *** experimental results demonstrate that our improved pix2pix model significantly enhances image quality and outperforms other models in the selected ***,the pix2pix model incorporating the differential image discriminator exhibits the most substantial improvements across all *** analysis of the experimental results reveals that the use of the U-Net++generator effectively reduces information feature loss,while the pix2pix model incorporating the differential image discriminator enhances the supervision of the generator during *** of these enhancements collectively improve the quality of pix2pix-generated images.

关键词： Image-to-image translation generative adversarial networks U-Net++ differential image pix2pix

Prediction of the Pore-Pressure Built-Up and Temperature of Fire-Loaded Concrete with pix2pix

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第5期79卷 2907-2922页

作者： Xueya Wang Yiming Zhang Qi Liu Huanran Wang Mechanics and Materials Science Research Center Ningbo UniversityNingbo315211China Jinyun Institute Zhejiang Sci-Tech UniversityLishui321400China School of Civil Engineering and Architecture Zhejiang Sci-Tech UniversityHangzhou310018China School of Computer Science Nanjing University of Information Science and TechnologyNanjing210044China

Concrete subjected to fire loads is susceptible to explosive spalling, which can lead to the exposure of reinforcingsteel bars to the fire, substantially jeopardizing the structural safety and stability. The spalling of fire-loaded concreteis closely related to the evolution of pore pressure and temperature. Conventional analytical methods involve theresolution of complex, strongly coupled multifield equations, necessitating significant computational efforts. Torapidly and accurately obtain the distributions of pore-pressure and temperature, the pix2pix model is adoptedin this work, which is celebrated for its capabilities in image generation. The open-source dataset used hereinfeatures RGB images we generated using a sophisticated coupled model, while the grayscale images encapsulate the15 principal variables influencing spalling. After conducting a series of tests with different layers configurations,activation functions and loss functions, the pix2pix model suitable for assessing the spalling risk of fire-loadedconcrete has been meticulously designed and trained. The applicability and reliability of the pix2pix model inconcrete parameter prediction are verified by comparing its outcomes with those derived fromthe strong couplingTHC model. Notably, for the practical engineering applications, our findings indicate that utilizing monochromeimages as the initial target for analysis yields more dependable results. This work not only offers valuable insightsfor civil engineers specializing in concrete structures but also establishes a robust methodological approach forresearchers seeking to create similar predictive models.

关键词： Fire loaded concrete spalling risk pore pressure generative adversarial network(GAN) pix2pix

Algorithm Development of Cloud Removal from Solar Images Based on pix2pix Network

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第5期71卷 3497-3512页

作者： Xian Wu Wei Song Xukun Zhang Ganghua Lin Haimin Wang Yuanyong Deng School of Information Engineering Minzu University of ChinaBeijing100081China Key Laboratory of Solar Activity Chinese Academy of Sciences(KLSACAS)Beijing100101China National Language Resource Monitoring and Research Center of Minority Languages Minzu University of ChinaBeijing100081China National Astronomical Observatories Chinese Academy of Sciences(NAOCCAS)Beijing100101China Center for Solar-Terrestrial Research New Jersey Institute of TechnologyUniversity HeightsNewark07102-1982USA Institute for Space Weather Sciences New Jersey Institute of TechnologyUniversity HeightsNewark07102-1982USA Big Bear Solar Observatory New Jersey Institute of Technology40386North Shore LaneBig Bear City92314-9672USA

Sky clouds affect solar observations *** shadows obscure the details of solar features in observed ***-covered solar images are difficult to be used for further research without *** this paper,the solar image cloud removing problem is converted to an image-to-image translation problem,with a used algorithm of the pixel to pixel Network(pix2pix),which generates a cloudless solar image without relying on the physical scattering ***2pix is consists of a generator and a *** generator is a well-designed *** discriminator uses PatchGAN structure to improve the details of the generated solar image,which guides the generator to create a pseudo realistic solar *** image generation model and the training process are optimized,and the generator is jointly trained with the *** the generation model which can stably generate cloudless solar image is *** experiment results on Huairou Solar Observing Station,National Astronomical Observatories,and Chinese Academy of Sciences(HSOS,NAOC and CAS)datasets show that pix2pix is superior to the traditional methods based on physical prior knowledge in peak signal-to-noise ratio,structural similarity,perceptual index,and subjective visual *** result of the PSNR,SSIM and PI are 27.2121 dB,0.8601 and 3.3341.

关键词： pix2pix solar image cloud removal

基于改进生成对抗网络pix2pix的火电厂NO_(x)排放预测

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

中国电机工程学报 2023年第12期43卷 4678-4687页

作者：张志宇李新利杨国田华北电力大学控制与计算机工程学院北京市昌平区102206

火电厂燃烧产生NO_(x)会污染大气环境,设计精确高效的NOx实时软测量方案对于节能减排至关重要。该文提出一种基于pix2pix的NO_(x)排放预测模型。采用互信息对影响NOx生成的主要变量进行重要性排序并选择相关特征变量,构建基于pix2pix生... 详细信息

火电厂燃烧产生NO_(x)会污染大气环境,设计精确高效的NOx实时软测量方案对于节能减排至关重要。该文提出一种基于pix2pix的NO_(x)排放预测模型。采用互信息对影响NOx生成的主要变量进行重要性排序并选择相关特征变量,构建基于pix2pix生成对抗网络模型,通过生成器和判别器的博弈训练优化参数。基于某660MW燃煤机组的运行参数进行仿真验证,并与基于长短期记忆网络、反向传播神经网络和最小二乘支持向量机的典型NO_(x)预测模型进行对比,结果表明所提方法可有效解决数据维度增加带来的预测困难,具有较高的预测精度和较小误差。

关键词： NO_(x)排放预测生成对抗网络 pix2pix 互信息

在线全文

学校读者我要写书评

暂无评论

基于改进pix2pix的红外图像转换技术

激光与红外 2024年第7期54卷 1157-1163页

作者：叶明亮史春景郝永平李大伟沈阳理工大学机械工程学院辽宁沈阳110159 沈阳理工大学装备工程学院辽宁沈阳110159

针对不同波段图像获取代价不同的问题,提出一种基于pix2pix的图像转换方法并进行改进。主要针对生成器和鉴别器两方面进行改进。生成器方面,使用残差结构的生成器替换原来的U-Net生成器以缓解梯度消失问题;引入可变形卷积,提高目标边缘... 详细信息

针对不同波段图像获取代价不同的问题,提出一种基于pix2pix的图像转换方法并进行改进。主要针对生成器和鉴别器两方面进行改进。生成器方面,使用残差结构的生成器替换原来的U-Net生成器以缓解梯度消失问题;引入可变形卷积,提高目标边缘和小目标的生成效果;引入BAM注意力机制,提高了算法对图像中主要目标的特征提取能力以提升生成图像的效果。鉴别器方面:改变PatchGAN中卷积层的层数(原PatchGAN为3层卷积),设置对照实验找到转换效果最好的卷积层数。以可见光图像和红外图像之间的转换为例进行实验。实验结果表明,改进后的算法在生成图像上的均方根误差(MSE)下降了31.4%、结构相似性(SSIM)提高了11.2%,可以更好的实现红外图像和可见光图像之间的转换。

关键词：生成对抗网络 pix2pix 图像转换残差结构

基于生成对抗网络(pix2pix)的家具设计草图渲染

在线全文

学校读者我要写书评

暂无评论

家具与室内装饰 2023年第10期30卷 117-121页

作者：朱文霜王禹钧郑文俊桂林理工大学艺术学院广西桂林541006

家具设计师工作初期通常会借助手绘线稿来描绘创意与灵感,但手绘线稿往往难以快速深化效果。为改善该状态,故尝试运用生成对抗网络(Generative Adversarial Network, GAN)门类中的深度学习图像转译模型(pix2pix)对手绘线稿进行快速渲染... 详细信息

家具设计师工作初期通常会借助手绘线稿来描绘创意与灵感,但手绘线稿往往难以快速深化效果。为改善该状态,故尝试运用生成对抗网络(Generative Adversarial Network, GAN)门类中的深度学习图像转译模型(pix2pix)对手绘线稿进行快速渲染。先以样本采集法创建家具数据集,据此对pix2pix进行训练;再建立交互界面来满足用户操作需求;最后通过两组测试验证其可行性。通过对pix2pix生成的图像效果与使用者的反馈分析可知,pix2pix能有效辅助家具设计师进行草图渲染,同时也能帮助非专业设计人员呈现自身创意。这既为家具设计效果表现探索出一条数字化路径,也是一次设计学和计算机科学与技术的深度融合研究。

关键词：生成对抗网络图像转译深度学习家具设计 pix2pix

一种基于pix2pix改进的工业缺陷数据增强方法

在线全文

学校读者我要写书评

暂无评论

计算机工程与科学 2022年第12期44卷 2206-2212页

作者：罗月童段昶江佩峰周波合肥工业大学计算机与信息学院安徽合肥230601

基于深度学习的目标检测算法在工业检测中应用广泛,为解决工业缺陷数据不足的问题,提出了一种基于pix2pix改进的缺陷数据增强方法。从加强生成器和判别器对图像中缺陷区域的注意力出发,针对pix2pix进行了如下改进:(1)仅将整幅图像的缺... 详细信息

基于深度学习的目标检测算法在工业检测中应用广泛,为解决工业缺陷数据不足的问题,提出了一种基于pix2pix改进的缺陷数据增强方法。从加强生成器和判别器对图像中缺陷区域的注意力出发,针对pix2pix进行了如下改进:(1)仅将整幅图像的缺陷区域作为判别器的输入,以此提升生成器对缺陷区域的注意力,同时,判别器采用了更小的卷积核提取缺陷区域的特征;(2)仅将图像中所有缺陷区域的平均生成对抗损失作为该图像的生成对抗损失,使网络更加关注缺陷区域的特征学习。在工业LED缺陷数据集上的实验结果表明,本方法生成的缺陷具有更逼真的视觉效果和更低的FID指数,同时有效提升了基于RetinaNet算法的缺陷检测精度。

关键词：目标检测数据增强 pix2pix 注意力

基于pix2pix的建筑布局生成设计研究——以大学校园布局为例

在线全文

学校读者我要写书评

暂无评论

基于pix2pix的建筑布局生成设计研究——以大学校园布局为例

作者：赖杨婷华南理工大学

学位级别：硕士

随着我国高等教育进入普及化、经济进入逐优竞争的新阶段,高等教育在当今社会经济发展中的作用日益凸显,高校建设需求迫切。大学校园布局是大学校园方案初期的重要阶段,决定了校园空间组织方式与整体方案的发展方向。近年来人工智能技... 详细信息

随着我国高等教育进入普及化、经济进入逐优竞争的新阶段,高等教育在当今社会经济发展中的作用日益凸显,高校建设需求迫切。大学校园布局是大学校园方案初期的重要阶段,决定了校园空间组织方式与整体方案的发展方向。近年来人工智能技术蓬勃发展,深度学习技术使人们看到了人工智能模仿人类思考的潜力。随着pix2pix等神经网络模型的提出,深度学习在图像处理方面的能力引起了建筑学领域的注意,并逐渐出现人工智能与建筑设计结合的跨专业研究。本文的研究目的是以大学校园布局为例,探索基于小样本实现建筑布局生成的策略,利用深度学习技术中的pix2pix模型实现建筑布局自动生成设计。经过有效训练,pix2pix模型能够在极短时间内,根据用地条件及周边环境迅速生成符合要求的大学校园布局方案草图,提供候选方案,同时启发建筑师在短时间内进行多样化设计思考。深度学习技术的两个关键组成部分是算法框架与数据。根据技术原理,部分深度学习技术的数据需要进行标注。与老师给学生划出考试重点的过程类似,数据标注是为了帮助计算机高效、准确学习目标知识。然而,与人脸识别这一类任务中标注人脸的“简单标注”不同,建筑布局设计中的数据标注是需要专业知识背景的“复杂标注”。在pix2pix的跨界应用中,算法框架由计算机领域学者建立,而建筑学领域学者的主要工作是基于专业知识提供有效的标注数据。可以说,数据标注方法及内容几乎决定了pix2pix模型最终生成的建筑布局效果。本文工作主要分为三个部分:由于该领域研究尚处于探索阶段,本文第一部分通过对人工智能技术在建筑布局生成领域的研究及文献进行梳理,结合既有实验工作,建立实验理论基础,提出实验难点及初步应对思路。本文的第二部分是以中心环线型布局为例,基于pix2pix的大学校园布局生成实验过程详述。该部分首先明确实验思路,收集大学校园优秀实际案例。其次,基于大学校园布局设计理论进行数据标注规则设计,对数量有限的大学校园实际案例进行人工改绘,建立“少而精”的小样本数据集,解决“有效案例数量少、模型数据需求大”的矛盾。与此同时,通过反复实验进行数据标注规则优化,试图将建筑设计的核心内容——布局方法与规律——更清晰地表达在图像数据中,提供给pix2pix模型进行学习,从而达到预期的布局生成设计效果。本文的第三部分是对整个研究的回顾与探讨。首先对最终实验结果进行定性、定量分析,在此基础上对实验的局限性与不足进行反思;其次对基于小样本实现建筑布局生成的策略进行总结;最后将pix2pix生成布局结果与人类建筑师完成方案进行对比探讨,思考人工智能浪潮下建筑师的挑战与机遇。

关键词：大学布局生成深度学习 pix2pix