顾及冰源噪声影响的冰下侧扫声纳图像去噪方法

doi:10.12305/j.issn.1001-506X.2026.11.03

系统工程与电子技术 ›› 2026, Vol. 48 ›› Issue (11): 3719-3726.doi: 10.12305/j.issn.1001-506X.2026.11.03

顾及冰源噪声影响的冰下侧扫声纳图像去噪方法

叶宇浩(), 柴洪洲(), 明璐璐(), 王志豪(), 王亦乔(), 马洪磊()

网络空间部队信息工程大学地理空间信息学院，河南郑州 450001

收稿日期:2025-09-24 修回日期:2026-02-26 接受日期:2026-03-26 出版日期:2026-10-25 发布日期:2026-06-01
通讯作者: 柴洪洲 E-mail:18705798703@163.com;chaihz1969@163.com;mll9805@163.com;ZhihaoWang1997@outlook.com;15732139087@163.com;315154754@qq.com
作者简介:叶宇浩（2000—），男，博士研究生，主要研究方向为海底地形地貌与环境感知，极地环境探测
明璐璐（1998—），女，博士研究生，主要研究方向为水下目标检测与识别
王志豪（1997—），男，博士研究生，主要研究方向为GNSS水汽反演
王亦乔（2000—），女，博士研究生，主要研究方向为遥感影像处理、图像分割、极地环境探测
马洪磊（1989—），男，讲师，博士研究生，主要研究方向为倒置式超短基线水声定位辅助无人水下航行器水下导航
基金资助:
国家自然科学基金青年科学基金（42404010）资助课题

Under-ice side-scan sonar image denoising method considering ice-generated noise influence

Yuhao Ye(), Hongzhou Chai(), Lulu Ming(), Zhihao Wang(), Yiqiao Wang(), Honglei Ma()

Institute of Geospatial Information，Network Space Force Information Engineering University，Zhengzhou 450001，China

Received:2025-09-24 Revised:2026-02-26 Accepted:2026-03-26 Online:2026-10-25 Published:2026-06-01
Contact: Hongzhou Chai E-mail:18705798703@163.com;chaihz1969@163.com;mll9805@163.com;ZhihaoWang1997@outlook.com;15732139087@163.com;315154754@qq.com

摘要/Abstract

摘要：

针对目前图像脉冲噪声模型未考虑声纳成像机制与海洋环境噪声时频特性，导致基于其训练的声纳图像去噪模型在噪声特性改变时性能下降的问题，面向具有特殊海洋环境的北极海域，提出侧扫声纳冰源噪声添加模型。首先，基于第9次北极科考采集的冰源噪声数据实现噪声仿真。然后，结合声纳方程计算其环境噪声级并融入声纳回波强度，实现与真实成像高度匹配的冰下侧扫声纳噪声仿真并构建成对训练集。最后，实验验证基于所提模型生成的成对训练集能显著优化卷积稀疏编码网络的冰源噪声去除效果，进一步验证了所提模型的正确性和有效性。

关键词: 冰源噪声, 噪声添加模型, 侧扫声纳, 声纳图像去噪, 卷积稀疏编码

Abstract:

To address the issue that existing image impulse noise models fail to consider the sonar imaging mechanism and the time-frequency characteristics of marine environmental noise, leading to performance degradation of sonar image denoising models trained on them when noise characteristics change, a side-scan sonar ice-generated noise addition model for the Arctic sea area with a special marine environment is proposed. Firstly, noise simulation is implemented based on the ice-generated noise data collected during the 9th Chinese Arctic Scientific Expedition. Then, the environmental noise level is calculated by combining the sonar equation and integrated into the sonar echo intensity. The under-ice side-scan sonar noise simulation highly matching real imaging is achieved, and paired training sets are constructed. Eventually, experiments verify that the paired training sets generated based on the proposed model can significantly optimize the ice-generated noise removal effect of the convolutional sparse coding network, further confirming the correctness and effectiveness of the proposed model.

Key words: ice-generated noise, noise addition model, side-scan sonar, sonar image denoising, convolutional sparse coding

中图分类号:

P229

叶宇浩, 柴洪洲, 明璐璐, 王志豪, 王亦乔, 马洪磊. 顾及冰源噪声影响的冰下侧扫声纳图像去噪方法[J]. 系统工程与电子技术, 2026, 48(11): 3719-3726.

Yuhao Ye, Hongzhou Chai, Lulu Ming, Zhihao Wang, Yiqiao Wang, Honglei Ma. Under-ice side-scan sonar image denoising method considering ice-generated noise influence[J]. Systems Engineering and Electronics, 2026, 48(11): 3719-3726.

图/表 15

图1

图2

图3

图4

图5

图6

图7

表1

图8

图9

表2

图10

表3

图11

表4

参考文献 30

1	李家彪, 冷疏影, 张亮, 等. 国际治理背景下北极海底科学探索的进展与展望[J]. 中国科学基金, 2025 (6): 934.
2	Wadhams P. Sidescan sonar imagery of sea ice in the Arctic Ocean[J]. Canadian Journal of Remote Sensing, 1978, 4 (2): 161.
3	Gwyther D E, Spain E A, King P, et al. Cold ocean cavity and weak basal melting of the Sorsdal ice shelf revealed by surveys using autonomous platforms[J]. Journal of Geophysical Research: Oceans, 2020, 125 (6): e2019JC015882.
4	Fan S S, Zhang X Y, Zeng G X, et al. Underwater ice adaptive mapping and reconstruction using autonomous underwater vehicles[J]. Frontiers in Marine Science, 2023, 10, 1124752. doi: 10.3389/fmars.2023.1124752
5	Fan S S, Bose N, Liang Z M. Polar AUV challenges and applications: a review[J]. Drones, 2024, 8 (8): 413.
6	Huang C, Zhao J H, Yu Y C, et al. Comprehensive sample augmentation by fully considering SSS imaging mechanism and environment for shipwreck detection under zero real samples[J]. IEEE Trans. on Geoscience and Remote Sensing, 2022, 60, 5906814.
7	Blondel P. The handbook of sidescan sonar[M]. Berlin: Springer, 2009.
8	Li M M, Liang H Y, Liu S X, et al. Asymptotically optimal algorithms for running max and min filters on random inputs[J]. IEEE Trans. on Signal Processing, 2018, 66 (13): 3421.
9	韩冷, 强发军, 倪旭, 等. 一种基于小波图像分解的侧扫声纳电源周期性噪声的去噪方法[J]. 数字技术与应用, 2020, 38 (5): 121.
10	Papyan V, Romano Y, Elad M. Convolutional dictionary learning via local processing[C]// IEEE International Conference on Computer Vision, 2017: 5306.
11	Wang Y Q, Yao Q M, Kwok J T, et al. Scalable online convolutional sparse coding[J]. IEEE Trans. on Image Processing, 2018, 27 (10): 4850.
12	Simon D, Elad M. Rethinking the CSC model for natural images[C]// 32nd Conference on Neural Information Processing Systems, 2019.
13	Li Z Y, Wang Z S, Chen D S, et al. RepDNet: a re-parameterization despeckling network for autonomous underwater side-scan sonar imaging with prior-knowledge customized convolution[J]. Defence Technology, 2024, 35, 259.
14	Zhang K, Zuo W M, Chen Y J, et al. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising[J]. IEEE Trans. on Image Processing, 2017, 26 (7): 3142.
15	Yin H T, Chen H. Deep content-dependent 3-D convolutional sparse coding for hyperspectral image denoising[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17, 4125.
16	Vishwakarma A. Denoising and inpainting of sonar images using convolutional sparse representation[J]. IEEE Trans. on Instrumentation and Measurement, 2023, 72, 5007709.
17	Bristow H, Eriksson A, Lucey S. Fast convolutional sparse coding[C]// IEEE Conference on Computer Vision and Pattern Recognition, 2013: 391.
18	Wang Z Y, Xue T, Li J Y, et al. Can optical denoising clean sonar images? A benchmark and fusion approach[EB/OL]. [2026-03-03]. https://arxiv.org/abs/2501.13116.
19	Zhou X T, Mizuno K, Zhang Y L. A self-supervised denoising strategy for underwater acoustic camera imageries[EB/OL]. [2026-03-03]. https://arxiv.org/abs/2407.17121.
20	李启虎, 黄海宁, 尹力, 等. 北极水声学研究的新进展和新动向[J]. 声学学报, 2018, 43 (4): 420.
21	Kinda G B, Simard Y, Gervaise C, et al. Arctic underwater noise transients from sea ice deformation: characteristics, annual time series, and forcing in Beaufort Sea[J]. Journal of the Acoustical Society of America, 2015, 138 (4): 2034.
22	谭靖骞, 曹宇, 黄海宁, 等. 北极海域海洋环境噪声建模与特性分析[J]. 应用声学, 2020, 39 (5): 690.
23	宋国丽, 郭新毅, 马力. 确定峰度非高斯海洋环境噪声模型研究[J]. 声学学报, 2019, 44 (5): 887.
24	莫雪晶. 北极楚科奇海台海洋环境噪声研究[D]. 厦门: 自然资源部第三海洋研究所, 2024.
25	Chen R, Schmidt H. Temporal and spatial characteristics of the Beaufort Sea ambient noise environment[J]. The Journal of the Acoustical Society of America, 2020, 148 (6): 3928.
26	曹天宇. 嵌入式侧扫声纳模块化硬件设计与实现[D]. 哈尔滨: 哈尔滨工程大学, 2019.
27	吕晓蕊. Alpha稳定分布的模型仿真及参数估计[D]. 武汉: 华中科技大学, 2008.
28	赵建虎, 王晓, 张红梅. 侧扫声呐图像海底线自动提取方法研究[J]. 武汉大学学报（信息科学版）, 2017, 42 (12): 1797.
29	Wadhams P, Wilkinson J P, Kaletzky A. Sidescan sonar imagery of the winter marginal ice zone obtained from an AUV[J]. Journal of Atmospheric and Oceanic Technology, 2004, 21 (9): 1462.
30	Mittal A, Soundararajan R, Bovik A C. Making a "completely blind" image quality analyzer[J]. IEEE Signal Processing Letters, 2013, 20 (3): 209.

噪声类型/指标	均值	中位数
实测噪声信号	335	120
仿真噪声信号	273	197

模型	类项	参数
DnCNN	步长	1
	层数	17
	裁剪尺寸	40
	优化器	Adam
	数据增强	翻转、旋转
	学习率衰减迭代次数	50
	学习率	0.000 1
	学习率策略	ExpLR
CSCNet	步长	8
	滤波器数量	175
	裁剪尺寸	256
	优化器	Adam
	LISTA迭代次数	12
	学习率衰减迭代次数	50
	学习率	0.000 2
	学习率策略	StepLR

去噪算法	SSIM	PSNR/dB	耗时/$ \mu \mathrm{s} $
中值滤波	0.3459±0.0076	20.43±0.46	28
高斯双边滤波	0.6183±0.0142	21.27±0.77	239
小波去噪	0.7278±0.0372	21.49±1.36	1 569
DnCNN	0.5148±0.0105	24.00±0.61	2 262
CSCNet	0.8679±0.0196	29.34±1.06	3 2159

去噪算法	NIQE
中值滤波	11.00
高斯双边滤波	11.42
小波去噪	5.63
椒盐噪声CSCNet	5.56
冰源噪声CSCNet	5.51

顾及冰源噪声影响的冰下侧扫声纳图像去噪方法

Under-ice side-scan sonar image denoising method considering ice-generated noise influence

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 30

相关文章 1

编辑推荐

Metrics

本文评价