基于白化着色的沉船和失事飞机风格迁移方法

doi:10.13474/j.cnki.11-2246.2025.0613

测绘通报 ›› 2025, Vol. 0 ›› Issue (6): 73-77.doi: 10.13474/j.cnki.11-2246.2025.0613

• 学术研究 • 上一篇

基于白化着色的沉船和失事飞机风格迁移方法

闫白羽¹, 翟国君², 边少锋³

1. 中国地质大学(武汉)地理信息与工程学院, 湖北武汉 430070;
2. 中国地质大学(武汉)地质探测与评估教育部重点实验室, 湖北武汉 430074;
3. 海军工程大学, 湖北武汉 430033

收稿日期:2024-10-29 发布日期:2025-07-04
作者简介:闫白羽(1999—),男,硕士,主要从事侧扫声呐图像分类研究。E-mail:1043494739@qq.com
基金资助:
国家自然科学基金(42374050;42430101)

A style transfer method for shipwrecks and crashed aircraft based on whitening and coloring transformation

YAN Baiyu¹, ZHAI Guojun², BIAN Shaofeng³

1. School of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430070, China;
2. Key Laboratory of Geological Exploration and Evaluation of Ministry of Education, China University of Geosciences (Wuhan), Wuhan 430074, China;
3. Naval University of Engineering, Wuhan 430033, China

Received:2024-10-29 Published:2025-07-04

摘要/Abstract

摘要： 深度学习算法在图像分类领域中得到广泛应用,但由于包含目标的侧扫声呐图像数据较少,难以满足深度学习算法的训练需求,会导致过拟合等问题。风格迁移是扩充训练样本的有效方法之一。本文对WCT、PhotoWCT风格迁移算法的迭代过程进行研究和重建,根据Unpooling和Upsampling的特性对WCT算法在解码器不同特征层上进行了适当改进,编写了更适用于侧扫声呐图像的WCST风格迁移算法。通过WCST算法生成了逼真的含目标伪侧扫声呐图像,满足了图像分类网络的训练要求。在图像分类试验中分别用WCST、PhotoWCT生成的图像集作为训练集对ResNet50进行训练,并用真实侧扫声呐图像验证。在图像分类的各项精度上WCST算法均为最佳,证明了其在扩充侧扫声呐图像训练集方面的高效性和优越性。

关键词: 风格迁移, 侧扫声呐图像, 深度学习, 图像分类

Abstract: Deep learning algorithms have been widely applied in the field of image classification. However, the limited availability of side-scan sonar images containing target objects poses a significant challenge in meeting the training demands of these algorithms, often leading to issues like overfitting. Style transfer has emerged as an effective method for augmenting training samples. This paper investigates and reconstructs the iterative processes of the WCT and PhotoWCT style transfer algorithms. Based on the characteristics of Unpooling and Upsampling, we propose modifications to the WCT algorithm across different feature layers of the decoder, resulting in the development of the WCST style transfer algorithm, which is more suited for side-scan sonar imagery.Using the WCST algorithm, realistic pseudo side-scan sonar images containing target objects were generated to meet the training requirements of image classification networks. In subsequent image classification experiments, datasets generated by WCST and PhotoWCT were used to train ResNet50. The results demonstrated that WCST outperformed the other methods in all accuracy metrics, highlighting its effectiveness in augmenting high-quality training sets for side-scan sonar image classification.

Key words: style transfer, side-scan sonar images, deep learning, image classification

中图分类号:

P237

闫白羽, 翟国君, 边少锋. 基于白化着色的沉船和失事飞机风格迁移方法[J]. 测绘通报, 2025, 0(6): 73-77.

YAN Baiyu, ZHAI Guojun, BIAN Shaofeng. A style transfer method for shipwrecks and crashed aircraft based on whitening and coloring transformation[J]. Bulletin of Surveying and Mapping, 2025, 0(6): 73-77.

参考文献

[1] HUO Guanying,WU Ziyin,LI Jiabiao.Underwater object classification in sidescan sonar images using deep transfer learning and semisynthetic training data[J].IEEE Access,2020,8: 47407-47418.
[2] YE Xiufen,YANG Haibo,LI Chuanlong,et al.A gray scale correction method for side-scan sonar images based on retinex[J].Remote Sensing,2019,11(11): 1281.
[3] 吕思晨,梁红,杨长生.基于迁移学习的声呐图像分类方法研究[C]//2022年中国西部声学学术交流会论文集.哈尔滨:[s.n.],2022.
[4] 陈佳辉,陈岚萍,夏小云,等.基于迁移学习的海底底质声呐图像分类[J].计算机仿真,2022,39(1):229-233.
[5] FU Chengyang,LIU Wei,RANGA A,et al.DSSD: deconvolutional single shot detector[EB/OL].[2024-09-20].https://arxiv.org/abs/1701.06659v1.
[6] 朱世伟,杭仁龙,刘青山.基于类加权YOLO网络的水下目标检测[J].南京师大学报(自然科学版),2020,43(1): 129-135.
[7] 于淼.基于深度学习的侧扫声呐图像目标检测方法研究[D].哈尔滨: 哈尔滨工程大学,2020.
[8] NEUPANE D,SEOK J.A review on deep learning-based approaches for automatic sonar target recognition[J].Electronics,2020,9(11): 1972.
[9] LI Yijun,FANG Chen,YANG Jimei,et al.Universal style transfer via feature transforms[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems.[S.l.]:ACM Press,2017:385-395.
[10] WINNEMÖLLER H,OLSEN S C,GOOCH B.Real-time video abstraction[J].ACM Transactions on Graphics,2006,25(3): 1221-1226.
[11] SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C]//Proceedings of 2016 International Conference on Learning Representations.[S.l.]:Computer Science,2016.
[12] LI Y,LIU M-Y,LI X,et al.A closed-form solution to photorealistic image stylization[C]//Proceedings of 2018 European Conference on Computer Vision (ECCV).[S.l.]:ECCV,2018.
[13] NOH H,HONG S,HAN B.Learning deconvolution network for semantic segmentation[C]//Proceedings of 2015 IEEE International Conference on Computer Vision (ICCV).Santiago:IEEE,2015: 1520-1528.
[14] LI Chuanlong,YE Xiufen,XI Jier,et al.A texture feature removal network for sonar image classification and detection[J].Remote Sensing,2023,15(3): 616.
[15] XIA Guisong,BAI Xiang,DING Jian,et al.DOTA: a large-scale dataset for object detection in aerial images[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City: IEEE,2018: 3974-3983.
[16] ZHU Haigang,CHEN Xiaogang,DAI Weiqun,et al.Orientation robust object detection in aerial images using deep convolutional neural network[C]//Proceedings of 2015 IEEE International Conference on Image Processing (ICIP).Quebec City: IEEE,2015:3735-3739.
[17] CHENG Gong,HAN Junwei.A survey on object detection in optical remote sensing images[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016,117: 11-28.
[18] LONG Yang,GONG Yiping,XIAO Zhifeng,et al.Accurate object localization in remote sensing images based on convolutional neural networks[J].IEEE Transactions on Geoscience and Remote Sensing,2017,55(5): 2486-2498.
[19] CHENG Gong,HAN Junwei,LU Xiaoqiang.Remote sensing image scene classification: benchmark and state of the art[J].Proceedings of the IEEE,2017,105(10): 1865-1883.
[20] ZHANG P,TANG J,ZHONG H,et al.Self-trained target detection of radar and sonar images using automatic deep learning[J].IEEE Transactions on Geoscience and Remote Sensing,2021,60:4701914.
[21] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas:IEEE,2016: 770-778.
[22] FELDENS P.Super resolution by deep learning improves boulder detection in side scan sonar backscatter mosaics[J].Remote Sensing,2020,12(14): 2284.

基于白化着色的沉船和失事飞机风格迁移方法

A style transfer method for shipwrecks and crashed aircraft based on whitening and coloring transformation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张超, 熊春宝, 连思达. 基于深度学习的海洋平台基础环境监测方法[J]. 测绘通报, 2025, 0(6): 12-17.
[2]	许淇, 杨嘉葳, 王继燕. 基于二次平滑和特征加权的高光谱图像分类[J]. 测绘通报, 2025, 0(6): 55-61.
[3]	陈登峰, 何拓航, 杨小燕, 刘世鹏, 孟屯良. 路基病害雷达图像多视图融合识别方法[J]. 测绘通报, 2025, 0(5): 66-73.
[4]	赵辰瑞, 连增增, 田亚林, 贺刘辉, 陈浩, 王鹏辉, 王孟奇. BiLSTM-Chan算法在超宽带室内定位中的应用[J]. 测绘通报, 2025, 0(5): 100-105.
[5]	朱松, 陈安泰, 华远盛, 姜文宇, 元鹏鹏, 朱家松. 面向漂流胶囊机器人图像的管道缺陷检测[J]. 测绘通报, 2025, 0(4): 34-38.
[6]	马于博, 张爱国, 王浩宇, 刘帅琪, 靳镜宇, 沈占锋, 李均力. 基于U-ConvHDNet模型的戈壁砾幕层提取[J]. 测绘通报, 2025, 0(4): 68-74.
[7]	李建, 王健, 王雷, 李敏, 杨立克, 赵艺龙. 双重注意力机制的电力走廊点云语义分割[J]. 测绘通报, 2025, 0(4): 127-133.
[8]	谢巴图, 胡佳睿, 潘俊. 基于双路全局信息优化网络的遥感影像海陆分割算法[J]. 测绘通报, 2025, 0(3): 52-58,86.
[9]	马海志. 基于深度学习的隧道超高分辨率图像病害检测框架[J]. 测绘通报, 2025, 0(3): 105-110.
[10]	张璇, 杨本勇, 文武, 邓维熙, 周何帆. 改进的DeepLabV3+模型用于四川县域大小春作物识别[J]. 测绘通报, 2025, 0(3): 133-137,167.
[11]	苟长龙, 庞敏, 杨扬. 改进的U-Net卷积网络在遥感影像地物分类中的应用[J]. 测绘通报, 2025, 0(3): 150-155.
[12]	宋培焱, 叶勤, 曾亮, 罗俊奇, 张硕, 尹长江. 面向密度明显差异点云的室内场景三维Mesh模型POCO重建方法[J]. 测绘通报, 2025, 0(2): 7-12,47.
[13]	石佳俊, 臧玉府, 肖雄武, 张莹滢. 非刚性匹配的激光点云与多光谱影像深度融合[J]. 测绘通报, 2025, 0(2): 41-47.
[14]	马浩然, 王金华. 多维特征学习与模型融合的地理命名实体识别[J]. 测绘通报, 2024, 0(9): 129-134.
[15]	石一剑, 谭海, 钟旭辉. 改进YOLOv5的国产光学影像辐射异常检测方法[J]. 测绘通报, 2024, 0(8): 60-65,72.