[1]叶添照,赵少杰,云季彪.基于迁移学习和卷积神经网络的桥梁图像美学评价[J].华侨大学学报(自然科学版),2025,(2):176-182.[doi:10.11830/ISSN.1000-5013.202410008]
　YE Tianzhao,ZHAO Shaojie,YUN Jibiao.Aesthetic Evaluation of Bridge Images Based on Transfer Learning and Convolutional Neural Networks[J].Journal of Huaqiao University(Natural Science),2025,(2):176-182.[doi:10.11830/ISSN.1000-5013.202410008]
点击复制

基于迁移学习和卷积神经网络的桥梁图像美学评价()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
期数:
2025年第2期
页码:
176-182
栏目:
出版日期:
2025-03-20

文章信息/Info

Title:
Aesthetic Evaluation of Bridge Images Based on Transfer Learning and Convolutional Neural Networks
文章编号:
1000-5013(2025)02-0176-07
作者:
叶添照;  赵少杰;  云季彪
湘潭大学 土木工程学院, 湖南 湘潭 411105
Author(s):
YE Tianzhao;  ZHAO Shaojie;  YUN Jibiao
School of Civil Engineering, Xiangtan University, Xiangtan 411105, China
关键词:
桥梁美学;  卷积神经网络;  迁移学习;  美学评价
Keywords:
bridge aesthetics;  convolutional neural networks;  transfer learning;  aesthetic evaluation
分类号:
TU026
DOI:
10.11830/ISSN.1000-5013.202410008
文献标志码:
A
摘要:
为了在桥梁方案设计中实现桥梁美学智能评价,提出一种基于迁移学习和卷积神经网络的桥梁图像美学自动评价方法。首先,通过冻结部分卷积层和修改丢弃率优化VGG16网络模型;其次,利用迁移学习将已知数据集AVA模型运用到桥梁图像评价上,最终可自动输出对应桥梁美学评分值。结果表明:与人工主观评分相比,文中方法的平均吻合度达到90.2%,该智能评价方法具有较好的准确性和工程实用性。
Abstract:
In order to realize intelligent evaluation of bridge schemes in bridge design, an automatic aesthetic evaluation method for bridge images based on transfer learning and convolutional neural networks is proposed. First, the VGG16 network model is optimized by freezing part of convolution layers and modifying the dropout rate. Second, the known data set AVA model is applied to bridge images evaluation by transfer learning, which can automaticlly output the corresponding aesthetic scores. The results show that compared with the manual subjective evaluation, the average coincidence degree of the proposed method is 90.2%, indicating that the intelligent evaluation method has good accuracy and engineering practicability.

参考文献/References:

[1] 王伟凝,王励,赵明权,等.基于并行深度卷积神经网络的图像美感分类[J].自动化学报,2016,42(6):904-914.DOI:10.16383/j.aas.2016.c150718.
[2] 蚁静缄.可计算的图像美学分类与评价系统研究[D].广州: 华南理工大学,2013.
[3] SUN Litian,YAMASAKI T,AIZAWA K.Photo aesthetic quality estimation using visual complexity features[J].Multimedia Tools and Applications,2018,77(5):5189-5213.DOI:10.1007/s11042-017-4424-4.
[4] ZHANG Xiaodan,GAO Xinbo,LU Wen,et al.Beyond vision: A multimodal recurrent attention convolutional neural network for unified image aesthetic prediction tasks[J].IEEE Transactions on Multimedia,2020,23:611-623.DOI:10.1109/TMM.2020.2985526.
[5] SHE Dongyu,LAI Yukun,YI Gaoxiong,et al.Hierarchical layout-aware graph convolutional network for unified aesthetics assessment[C]//Computer Vision and Pattern Recognition.Nashville:IEEE Press,2021:8471-8480.DOI:10.1109/CVPR46437.2021.00837.
[6] MARTIN-RODRIGUEZ F,GARCIA-MOJON R,FERNANDEZ-BARCIELA M.Detection of AI-created images using pixel-wise feature extraction and convolutional neural networks[J].Sensors,2023,23(22):9037.DOI:10.3390/s23229037.
[7] HE Shuai,XIAO Yi,MING Anlong,et al.Prompt-guided image color aesthetics assessment: Models, datasets and benchmarks[J].Information Fusion,2025,114:102706.DOI:10.1016/j.inffus.2024.102706.
[8] 梁艳,何畏,唐茂林.桥梁美学2020年度研究进展[J].土木与环境工程学报(中英文),2021,43(增刊1):234-241.DOI:10.11835/j.issn.2096-6717.2021.226.
[9] 李素梅,常永莉,段志成.基于卷积神经网络的立体图像舒适度客观评价[J].光学学报,2018,38(6):138-144.DOI:10.3788/AOS201838.0610003.
[10] 王伟凝,刘剑聪,徐向民,等.基于构图规则的图像美学优化[J].华南理工大学学报(自然科学版),2015,43(5):51-58.DOI:10.3969/j.issn.1000-565X.2015.05.009.
[11] 王欣,穆绍硕,陈华锋.基于多尺度特征提取网络的图像美学量化评分方法[J].浙江大学学报(理学版),2021,48(1):69-73.DOI:10.3785/j.issn.1008-9497.2021.01.010.
[12] DAICHI S,HIRONORI T,AKIHIRO K.Study on relationship between composition and prediction of photo aesthetics using CNN[J].Cogent Engineering,2022,9(1):2107472.DOI:10.1080/23311916.2022.2107472.
[13] LUO Xiaoyu,WU Yue,CHEN Airong,et al.Form finding and aesthetic design for pylons of cable-supported bridges[J].Structural Engineering International,2021,31(6):468-476.DOI:10.1080/10168664.2020.1870056.
[14] WONG LAIKUAN,LOW K.Saliency-enhanced image aesthetics class prediction[C]//IEEE International Conference on Image Processing.Cairo:IEEE Press,2009:993-996.DOI:10.1109/ICIP.2009.5413825.
[15] DATTA R,JOSHI D,LI Jia,et al.Studying aesthetics in photographic images using a computational approach[C]//9th European Conference on Computer Vision.Graz:Springer-Verlag,2006:288-301.DOI:10.1007/1174407 8_23.
[16] HOU Le,YU Chenping,SAMARAS D.Squared earth mover’s distance-based loss for training deep neural networks[EB/OL].(2016-11-17)[2024-10-10] .https://arxiv.org/abs/1611.05916.
[17] 牛顿,林宁,林振超,等.多特征融合的焊缝图像多标签分类算法[J].华侨大学学报(自然科学版),2024,45(4):514-523.DOI:10.11830/ISSN.1000-5013.202403033.

相似文献/References:

[1]吴琼,陈锻生.多尺度卷积循环神经网络的情感分类技术[J].华侨大学学报(自然科学版),2017,38(6):875.[doi:10.11830/ISSN.1000-5013.201606077]
　WU Qiong,CHEN Duansheng.Sentiment Classification With Multiscale Convolutional Recurrent Neural Network[J].Journal of Huaqiao University(Natural Science),2017,38(2):875.[doi:10.11830/ISSN.1000-5013.201606077]
[2]邹辉,杜吉祥,翟传敏,等.深度学习与一致性表示空间学习的跨媒体检索[J].华侨大学学报(自然科学版),2018,39(1):127.[doi:10.11830/ISSN.1000-5013.201508047]
　ZOU Hui,DU Jixiang,ZHAI Chuanmin,et al.Cross-Modal Multimedia Retrieval Based Deep Learning and Shared Representation Space Learning[J].Journal of Huaqiao University(Natural Science),2018,39(2):127.[doi:10.11830/ISSN.1000-5013.201508047]
[3]王改华,李涛,吕朦,等.采用无监督学习算法与卷积的图像分类模型[J].华侨大学学报(自然科学版),2018,39(1):146.[doi:10.11830/ISSN.1000-5013.201703109]
　WANG Gaihua,LI Tao,Lü Meng,et al.Image Classification Model Using Unsupervised Learning Algorithm and Convolution[J].Journal of Huaqiao University(Natural Science),2018,39(2):146.[doi:10.11830/ISSN.1000-5013.201703109]
[4]郑凌云,柳培忠,汪鸿翔.结合高斯核函数的卷积神经网络跟踪算法[J].华侨大学学报(自然科学版),2018,39(5):762.[doi:10.11830/ISSN.1000-5013.201702123]
　ZHENG Lingyun,LIU Peizhong,WANG Hongxiang.Convolution Neural Networks Tracking Algorithm Combined With Gaussian Kernel Function[J].Journal of Huaqiao University(Natural Science),2018,39(2):762.[doi:10.11830/ISSN.1000-5013.201702123]
[5]聂一亮,杜吉祥,杨麟.卷积特征图融合与显著性检测的图像检索[J].华侨大学学报(自然科学版),2018,39(6):937.[doi:10.11830/ISSN.1000-5013.201706028]
　NIE Yiliang,DU Jixiang,YANG Lin.Image Retrieval Based on Convolution Feature Map Fusion and Saliency Detection[J].Journal of Huaqiao University(Natural Science),2018,39(2):937.[doi:10.11830/ISSN.1000-5013.201706028]
[6]刘群,陈锻生.采用ACGAN及多特征融合的高光谱遥感图像分类[J].华侨大学学报(自然科学版),2019,40(1):113.[doi:10.11830/ISSN.1000-5013.201710006]
　LIU Qun,CHEN Duansheng.Classification of Hyperspectral Remote Sensing Images Using ACGAN and Fusion of Multifeature[J].Journal of Huaqiao University(Natural Science),2019,40(2):113.[doi:10.11830/ISSN.1000-5013.201710006]
[7]张圣祥,郑力新,朱建清,等.采用深度学习的快速超分辨率图像重建方法[J].华侨大学学报(自然科学版),2019,40(2):245.[doi:10.11830/ISSN.1000-5013.201804064]
　ZHANG Shengxiang,ZHENG Lixin,ZHU Jianqing,et al.Fast Super-Resolution Image Reconstruction Method Using Deep Learning[J].Journal of Huaqiao University(Natural Science),2019,40(2):245.[doi:10.11830/ISSN.1000-5013.201804064]
[8]吴晨茜,陈锻生.表情符向量化算法[J].华侨大学学报(自然科学版),2019,40(3):399.[doi:10.11830/ISSN.1000-5013.201803011]
　WU Chenxi,CHEN Duansheng.Emoticon Vectorization Algrorithm[J].Journal of Huaqiao University(Natural Science),2019,40(2):399.[doi:10.11830/ISSN.1000-5013.201803011]
[9]邱德府,郑力新,谢炜芳,等.深度学习下的高效单幅图像超分辨率重建方法[J].华侨大学学报(自然科学版),2019,40(5):668.[doi:10.11830/ISSN.1000-5013.201905029]
　QIU Defu,ZHENG Lixin,XIE Weifang,et al.Efficient Single Image Super-Resolution Reconstruction Method Under Deep Learning[J].Journal of Huaqiao University(Natural Science),2019,40(2):668.[doi:10.11830/ISSN.1000-5013.201905029]
[10]陈剑涛,黄德天,陈健,等.改进的二阶龙格-库塔超分辨率算法[J].华侨大学学报(自然科学版),2022,43(1):127.[doi:10.11830/ISSN.1000-5013.202012009]
　CHEN Jiantao,HUANG Detian,CHEN Jian,et al.Improved Second-Order Runge-Kutta Super-Resolution Algorithm[J].Journal of Huaqiao University(Natural Science),2022,43(2):127.[doi:10.11830/ISSN.1000-5013.202012009]

备注/Memo

备注/Memo:
收稿日期: 2024-10-21
通信作者: 赵少杰(1982-),男,副教授,博士,主要从事桥梁工程安全可靠性的研究。E-mail:shaojiez@qq.com。
更新日期/Last Update: 2025-03-20