[1]管凤旭,吴卓锋,张雨竹,等.明暗恢复形状算法改进的高精度快速水下图像三维重建方法[J].智能系统学报,2025,20(3):670-678.[doi:10.11992/tis.202401017]
　GUAN Fengxu,WU Zhuofeng,ZHANG Yuzhu,et al.High-precision and fast three-dimensional reconstruction of underwater images driven by the shape-from-shading algorithm[J].CAAI Transactions on Intelligent Systems,2025,20(3):670-678.[doi:10.11992/tis.202401017]
点击复制

明暗恢复形状算法改进的高精度快速水下图像三维重建方法

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 20 期数: 2025年第3期 页码: 670-678 栏目: 学术论文—智能系统 出版日期: 2025-05-05
Title:
High-precision and fast three-dimensional reconstruction of underwater images driven by the shape-from-shading algorithm
作者:
管凤旭1,2, 吴卓锋2, 张雨竹1, 唐世文1, 姚佳豪1, 杜雪1
1. 哈尔滨工程大学 智能科学与工程学院, 黑龙江 哈尔滨150001;
2. 哈尔滨工程大学 南海研究院, 海南 三亚 572024
Author(s):
GUAN Fengxu1,2, WU Zhuofeng2, ZHANG Yuzhu1, TANG Shiwen1, YAO Jiahao1, DU Xue1
1. College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin 150001, China;
2. Nanhai Institute, Harbin Engineering University, Sanya 572024, China
关键词:
侧扫声呐图像; 明暗恢复形状; 三维重建; 线性化法; 海底检测; 高精度成像; 水下图像; 图像处理
Keywords:
side-scan sonar image; shape from shading; 3D reconstruction; linearization method; seabed detection; high-precision imaging; underwater image; image processing
分类号:
TP18
DOI:
10.11992/tis.202401017
摘要:
明暗恢复形状(shape from shading，SFS)算法是三维重建中不可缺少的关键技术。该方法通过单幅图像中物体的灰度的亮度信息，求出每个点的相对高度和表面法向量，现有的SFS线性化算法利用泰勒级数进行展开忽略了高阶部分的影响，导致三维重建的图像不够精确等问题。为解决三维重建精度问题，本文提出一种基于SFS算法改进的高精度快速水下图像三维重建方法，在提高三维重建精度的前提下，解决了对本轮数据利用率不够、迭代速度慢等问题。将其应用在真实图像和虚拟图像上进行实验评估以及三维重建的量化分析，验证了相比于经典 SFS 线性化算法和SFS最小化算法，本文提出的SFS算法改进的高精度快速水下图像三维重建方法，取得了令人满意的信噪比和图像信息熵值。
Abstract:
Shape from shading (SFS) represents an indispensable key technique in three-dimensional (3D) reconstruction. It leverages the luminance information of the gray scale of the object in a single image to determine the relative height and the surface normal vector of each point, and the existing linearization algorithms of SFS leverage a Taylor series approach for the expansion, thereby ignoring the influence of the higher-order part, which causes issues, such as the inaccuracy of the reconstructed image in 3D. Thus, to solve the accuracy problem of 3D reconstruction, a high-precision, fast 3D reconstruction method for underwater images is proposed based on the enhancement driven by the SFS algorithm. This method solves the existing issues regarding insufficient data utilization and low iteration speed, thereby improving the accuracy of 3D reconstruction. Regarding the application to real and virtual images for the experimental evaluation and quantitative analysis of 3D reconstruction, the proposed SFS algorithm-enhanced high-precision and fast 3D reconstruction method for underwater images delivers satisfactory signal-to-noise ratios and entropy values regarding image information compared with classical SFS-linearization and SFS-minimization algorithms.
参考文献/References:
[1] 孙久文, 高宇杰. 中国海洋经济发展研究[J]. 区域经济评论, 2021(1): 38-47.
SUN Jiuwen, GAO Yujie. Research on China’s marine economic development[J]. Regional economic review, 2021(1): 38-47.
[2] 叶鹏程, 郭红梅, 潘光, 等. 海洋强国战略背景下高校海洋意识培育路径研究[J]. 科学大众(科学教育), 2021(2): 145, 120.
[3] 秦琳贵, 沈体雁. 科技创新促进中国海洋经济高质量发展了吗: 基于科技创新对海洋经济绿色全要素生产率影响的实证检验[J]. 科技进步与对策, 2020, 37(9): 105-112.
QIN Lingui, SHEN Tiyan. Does technological innovation promotethe high quality development of China’s marine economy: empirical test based on effect of technological innovation on GTFP[J]. Science & technology progress and policy, 2020, 37(9): 105-112.
[4] 许瑞恒, 林欣月. 多元补偿主体、环境规制与海洋经济可持续发展[J]. 经济问题, 2020(11): 58-67.
XU Ruiheng, LIN Xinyue. Multiple subjects of ecological compensation, environmental regulation and sustainable development of marine economy[J]. On economic problems, 2020(11): 58-67.
[5] LIU Yifei, YE Xiufen. A gray scale correction method for side-scan sonar images considering rugged seafloor[J]. IEEE transactions on geoscience and remote sensing, 2023, 61: 4209810.
[6] ZHANG Yang, LI Haisen, ZHU Jianjun, et al. Contrast study of side scan sonar image enhancement methods[C]//2021 OES China Ocean Acoustics. Harbin: IEEE, 2021: 995-999.
[7] YIN Jiye, WANG Yu, LYU Jiaqi, et al. Study on underwater simultaneous localization and mapping based on different sensors[C]//2021 IEEE 10th Data Driven Control and Learning Systems Conference. Suzhou: IEEE, 2021: 728-733.
[8] 何勇光. 海洋侧扫声呐探测技术的现状及发展[J]. 工程建设与设计, 2020(4): 275-276.
HE Yongguang. Present situation and development of ocean side scan sonar detection technology[J]. Construction & design for engineering, 2020(4): 275-276.
[9] 王晨, 陈晶晶, 王润田. 利用侧扫声呐实现船舶目标的快速识别[J]. 声学技术, 2022, 41(2): 199-204.
WANG Chen, CHEN Jingjing, WANG Runtian. Fast recognition of ship targets by using side-scan sonar[J]. Technical acoustics, 2022, 41(2): 199-204.
[10] JU Y C, BRUHN A, BREU? M. Variational perspective shape from shading[M]//Scale Space and Variational Methods in Computer Vision. Cham: Springer International Publishing, 2015: 538-550.
[11] PINK T H, RAMANATHAN U. Extraction of height information from target shadow for applications in ATC[C]//IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS’99. Hamburg: IEEE, 1999: 351-353.
[12] WANG Zheyin, SHEN Liquan, WANG Zhengyong, et al. Generation-based joint luminance-chrominance learning for underwater image quality assessment[J]. IEEE transactions on circuits and systems for video technology, 2023, 33(3): 1123-1139.
[13] HUANG Nuo, GONG Chen, FU Chunfang, et al. Preliminary investigation of air-to-water visible light communication link under strong ambient light[C]//2021 IEEE 94th Vehicular Technology Conference. Norman: IEEE, 2021: 1-5.
[14] 杜召平, 陈刚, 王达. 国外声呐技术发展综述[J]. 舰船科学技术, 2019, 41(1): 145-151.
DU Zhaoping, CHEN Gang, WANG Da. Foreign sonar technology development research summary[J]. Ship science and technology, 2019, 41(1): 145-151.
[15] 董鹏辉, 柯良军. 基于图像的三维重建技术综述[J]. 无线电通信技术, 2019, 45(2): 115-119.
DONG Penghui, KE Liangjun. Overview of 3D reconstruction techniques based on images[J]. Radio communications technology, 2019, 45(2): 115-119.
[16] HORN B K P. Height and gradient from shading[J]. International journal of computer vision, 1990, 5(1): 37-75.
[17] 郑太雄, 黄帅, 李永福, 等. 基于视觉的三维重建关键技术研究综述[J]. 自动化学报, 2020, 46(4): 631-652.
ZHENG Taixiong, HUANG Shuai, LI Yongfu, et al. Key techniques for vision based 3D reconstruction: a review[J]. Acta automatica sinica, 2020, 46(4): 631-652.
[18] HORN B K P, SZELISKI R S, YUILLE A L. Impossible shaded images[J]. IEEE transactions on pattern analysis and machine intelligence, 1993, 15(2): 166-170.
[19] ZHANG Ruo, TSAI P S, CRYER J E, et al. Shape-from-shading: a survey[J]. IEEE transactions on pattern analysis and machine intelligence, 1999, 21(8): 690-706.
[20] WANG Shuo, WU Tonghai, WANG Kunpeng, et al. 3-D particle surface reconstruction from multiview 2-D images with structure from motion and shape from shading[J]. IEEE transactions on industrial electronics, 2021, 68(2): 1626-1635.
[21] PENTLAND A P. Local shading analysis[J]. IEEE transactions on pattern analysis and machine intelligence, 1984, PAMI-6(2): 170-187.
[22] HEO S, SONG H, KANG J, et al. High-quality single image 3D facial shape reconstruction via robust albedo estimation[C]//2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference. Tokyo: IEEE, 2021: 1428-1432.
[23] 赵建虎, 尚晓东, 张红梅. 侧扫声呐图像反演海底地形的一种线性算法[J]. 哈尔滨工业大学学报, 2017, 49(5): 80-86.
ZHAO Jianhu, SHANG Xiaodong, ZHANG Hongmei. Obtaining high-resolution seafloor topography from side scan sonar image using a linear algorithm[J]. Journal of Harbin Institute of Technology, 2017, 49(5): 80-86.
[24] NEUPANE D, SEOK J. A review on deep learning-based approaches for automatic sonar target recognition[J]. Electronics, 2020, 9(11): 1972.
[25] TSAI P S, SHAH M. Shape from shading using linear approximation[J]. Image and vision computing, 1994, 12(8): 487-498.
[26] 邓学良, 何扬波, 周建丰. 基于光度立体的三维重建方法综述[J]. 现代计算机, 2021, 27(23): 133-143.
DENG Xueliang, HE Yangbo, ZHOU Jianfeng. Review of three-dimensional reconstruction methods based on photometric stereo[J]. Modern computer, 2021, 27(23): 133-143.
[27] BASRI R, JACOBS D. Lambertian reflectance and linear subspaces[C]//Proceedings Eighth IEEE International Conference on Computer Vision. Vancouver: IEEE, 2001: 383-390.
[28] TSAI P S, SHAH M. A fast linear shape from shading[C]//Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Champaign: IEEE, 1992: 734-736.
[29] 刘浩林, 覃珊珊, 李鹏鸽, 等. 基于SFS方法的侧扫声呐图像三维重构[J]. 舰船科学技术, 2021, 43(15): 125-130.
LIU Haolin, QIN Shanshan, LI Pengge, et al. Three-dimensional reconstruction of side scan sonar imagebased on SFS method[J]. Ship science and technology, 2021, 43(15): 125-130.
[30] XI Zhao, ZHAO Jianhu, ZHU Weiqiang. Side-scan sonar image simulation considering imaging mechanism and marine environment for zero-shot shipwreck detection[J]. IEEE transactions on geoscience and remote sensing, 2023, 61: 4209713.
相似文献/References:
[1]盛子旗,霍冠英.样本仿真结合迁移学习的声呐图像水雷检测[J].智能系统学报,2021,16(2):385.[doi:10.11992/tis.202101030]
　SHENG Ziqi,HUO Guanying.Detection of underwater mine target in sidescan sonar image based on sample simulation and transfer learning[J].CAAI Transactions on Intelligent Systems,2021,16():385.[doi:10.11992/tis.202101030]

备注/Memo

收稿日期:2024-1-15。
基金项目:国家自然科学基金项目(62101156).
作者简介:管凤旭，副教授，博士，主要研究方向为无人系统自主控制、机器视觉目标检测与跟踪、计算机控制及应用。获发明专利授权近20项，出版教材5部，发表学术论文40余篇。E-mail：guanfengxu@hrbeu.edu.cn。;吴卓锋，硕士研究生，主要研究方向为无人系统自主控制。E-mail：1157991141@qq.com。;张雨竹，硕士研究生，主要研究方向为计算机视觉。E-mail：1260641705@qq.com。
通讯作者:管凤旭. E-mail：guanfengxu@hrbeu.edu.cn

更新日期/Last Update: 1900-01-01
Copyright © 《 智能系统学报》 编辑部
地址：（150001）黑龙江省哈尔滨市南岗区南通大街145-1号楼 电话：0451- 82534001、82518134　邮箱：tis@vip.sina.com