基于改进Criminisi和SAM算法有信息损失的筏式养殖要素遥感提取

Remote Sensing Extraction of Raft Aquaculture Elements with Information Loss Based on Improved Criminisi and SAM Algorithms

摘要

筏式养殖是养殖水产品的一种方式,利用遥感技术进行筏式养殖的准确监测是控制养殖密度和科学规划的重要途径.由于筏式养殖的分布区位深受波浪和镜面反射的影响,在遥感影像上很多筏式养殖区存在信息损失.为提高提取精度和准确性,本文针对有信息损失的筏式养殖要素,提出一种基于改进Criminisi和SAM(segment anything model)算法的提取方法.首先利用暗通道原理对信息损失区进行定位和分类;然后通过优化Criminisi算法的优先权计算、样本块生成、匹配准则,有效修复信息损失区;最后采用SAM算法实现零样本情况下的筏式养殖要素提取.此外,采用高分6号影像数据集,同两种先进的基于深度学习的算法---RSMamba(remote sensing Mamba)和Swin Transformer进行了对比验证,结果表明:本文算法提取结果的准确率和完整率均不低于90%,错误率不高于0.02%.

Abstract

Raft aquaculture is a key method for farming aquatic products, and utilizing remote sensing technology for its accurate monitoring is an essential approach for controlling aquaculture density and supporting scientific planning. However, because the spatial distribution of raft aquaculture is heavily influenced by waves and specular reflection, significant information loss occurs in many raft aquaculture areas on remote sensing images. To improve extraction precision and accuracy, this paper proposes an extraction method based on the improved Criminisi and SAM (segment anything model) algorithms specifically for raft aquaculture elements with information loss. First, the dark channel prior is used to locate and classify the areas affected by information loss. The Criminisi algorithm is then improved by optimizing priority calculation, sample block generation, and matching criteria to effectively restore the information loss zones. Finally, the SAM algorithm is applied to achieve zero-shot extraction of raft aquaculture elements. In addition, experiments using a Gaofen 6 image dataset were conducted to compare the proposed method with two advanced deep learning-based algorithms, RSMamba (remote sensing mamba) and Swin Transformer. The experimental results demonstrate that the proposed algorithm achieves both the accuracy rate and completeness rate of at least 90%, with an error rate not exceed 0.02%, indicating superior performance.

关键词

Criminisi / SAM / 信息损失 / 筏式养殖 / 遥感提取

Key words

Criminisi / SAM / information loss / raft aquaculture / remote sensing extraction

参考文献

[1] 初佳兰, 邵光辉, 赵建华, 等. 高分一号的浮筏养殖信息提取方法[J]. 测绘科学, 2020, 45(1): 92-98.

[2] Chu Jialan, Shao Guanghui, Zhao Jianhua, et al. Information Extraction of Floating Raft Aquaculture Based on GF-1[J]. Science of Surveying and Mapping, 2020, 45(1): 92-98.

[3] Virdis S G, Pasquale P. An Object-Based Image Analysis Approach for Aquaculture Ponds Precise Mapping and Monitoring: A Case Study of Tam Giang-Cau Hai Lagoon, Vietnam[J]. Environmental Monitoring and Assessment, 2013, 186(1): 117-133.

[4] Liu Y M, Yang X M, Wang Z H, et al. Aquaculture Area Extraction and Vulnerability Assessment in Sanduao Based on Richer Convolutional Features Network Model[J]. Journal of Oceanology and Limnology, 2019, 37(6): 153-166.

[5] 李凯强. 基于遥感观测的海岸带养殖用海水体光谱特征分析与分类研究[D]. 青岛: 山东科技大学, 2020.

[6] Li Kaiqiang. Remote Sensing Observation of Coastal Aquaculture Water Body Spectral Characteristic Analysis and Classification Research[D]. Qingdao: Shandong University of Science and Technology, 2020.

[7] 赵彬如, 牛思文, 杨晓彤, 等. 基于WorldView-2的冰面水体增强指数构建及精细化提取[J]. 海洋通报, 2022, 41(4): 391-400.

[8] Zhao Binru, Niu Siwen, Yang Xiaotong, et al. The Construction of Normalized Enhanced Water Index and the Extraction of Glacial Water Based on WorldView-2 Imagery[J]. Marine Science Bulletin, 2022, 41(4): 391-400.

[9] Cui B, Fei D, Shao G, et al. Extracting Raft Aquaculture Areas from Remote Sensing Images via an Improved U-Net with a PSE Structure[J]. Remote Sensing, 2019, 11(17): 6296-6309.

[10] 吕伏, 崔向燕, 刘铁. 亮度均衡与边缘强化的多尺度低照度图像增强算法[J]. 激光与光电子学进展, 2024, 61(12): 451-459.

[11] Lü Fu, Cui Xiangyan, Liu Tie. Multiscale Low-Light Image Enhancement Algorithm with Brightness Equalization and Edge Enhancement Algorithm[J]. Laser & Optoelectronics Progress, 2024, 61(12): 451-459.

[12] 李同宇, 陈捷, 张梦悦, 等. 基于局部补偿和NSCT的海洋耀斑光学图像增强[J]. 激光与光电子学进展, 2025, 62(6): 0637008.

[13] Li Tongyu, Chen Jie, Zhang Mengyue, et al. Flare-Containing Marine Optical Image Enhancement Based on Local Compensation and NSCT[J]. Laser & Optoelectronics Progress, 2025, 62(6): 0637008.

[14] Zeng Z, Wang D, Tan W, et al. Extracting Aquaculture Ponds from Natural Water Surfaces Around Inland Lakes on Medium Resolution Multispectral Images[J]. International Journal of Applied Earth Observation & Geoinformation, 2019, 80: 13-25.

[15] Duan Y Q, Li X, Zhang L P, et al. Detecting Spatiotemporal Changes of Large-Scale Aquaculture Ponds Regions over 1988-2018 in Jiangsu Province, China Using Google Earth Engine[J]. Ocean & Coastal Management, 2020, 188(4): 105-144.

[16] Cheng G, Xie X, Han J, et al. Remote Sensing Image Scene Classification Meets Deep Learning: Challenges, Methods, Benchmarks, and Opportunities[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 3735-3756.

[17] Fu Y, You S, Zhang S, et al. Marine Aquaculture Mapping Using GF-1 WFV Satellite Images and Full Resolution Cascade Convolutional Neural Network[J]. International Journal of Digital Earth, 2022, 15(1): 2047-2060.

[18] 刘圆月. 基于深度神经网络模型优化的海水养殖遥感信息提取[D]. 重庆: 重庆三峡学院, 2024.

[19] Liu Yuanyue. Extraction of Aquaculture Cages from High-Resolution Remote Sensing Images Based on Deep Learning[D]. Chongqing: Chongqing Three Gorges University, 2024.

[20] Liu C, Jiang T, Zhang Z, et al. Extraction Method of Offshore Mariculture Area Under Weak Signal Based on Multisource Feature Fusion[J]. Journal of Marine Science and Engineering, 2020, 8(2): 99-101.

[21] Gao L, Wan C, Liu K, et al. Extraction of Floating Raft Aquaculture Areas from Sentinel-1 SAR Images by a Dense Residual U-Net Model with Pre-Trained ResNet34 as the Encoder[J]. Remote Sensing, 2020, 14(13): 3003.

[22] Liu J, Lu Y, Guo X, et al. A Deep Learning Method for Offshore Raft Aquaculture Extraction Based on Medium-Resolution Remote Sensing Images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16: 6296-6309.

[23] 王世瑞, 沈芳, 李仁虎, 等. 无人机高光谱影像水面耀光去除及信息重构方法研究[J]. 华东师范大学学报(自然科学版), 2024(1): 36-49.

[24] Wang Shirui, Shen Fang, Li Renhu, et al. Research on Water Surface Glint Removal and Information Reconstruction Methods for Unmanned Aerial Vehicle Hyperspectral Images[J]. Journal of East China Normal University (Natural Science), 2024(1): 36-49.

[25] 李宁宁. 基于卫星影像空谱信息的浅海水深反演与底质分类研究[D]. 青岛: 自然资源部第一海洋研究所, 2023.

[26] Li Ningning. Bathymetric Inversion and Substrate Classification in Shallow Waters Based on Satellite Image Spatial Spectrum Information[D]. Qingdao: First Institute of Oceanography, Ministry of Natural Resources, 2023.

[27] 李奇, 李龙, 王卫, 等. 基于改进Criminisi算法的破损纺织品文物图像修复[J]. 激光与光电子学进展, 2023, 60(16): 173-182.

[28] Li Qi, Li Long, Wang Wei, et al. Image Inpainting of Damaged Textiles Based on Improved Criminisi Algorithm[J]. Laser & Optoelectronics Progress, 2023, 60(16): 173-182.

[29] 黄春. 基于Criminisi的图像修复算法研究及软件实现[D]. 西安: 西安电子科技大学, 2022.

[30] Huang Chun. Research on Image Inpainting Algorithm Based on Criminisi and Software Implementation[D]. Xi’an: Xidian University, 2022.

[31] 曹浩杰, 李郁峰, 张权. 一种改进的Criminisi眼底图像修复算法[J]. 中北大学学报(自然科学版), 2023, 44(2): 176-181, 192.

[32] Cao Haojie, Li Yufeng, Zhang Quan. An Improved Criminisi Fundus Image Restoration Algorithm[J]. Journal of North University of China (Natural Science Edition), 2023, 44(2): 176-181, 192.

[33] 余磊. 光学遥感卫星色彩一致性合成影像生成关键技术研究[D]. 武汉: 武汉大学, 2017.

[34] Yu Lei. Color-Distortion Correction for Jilin-1 KF01 Series Satellite Imagery Using a Data-Driven Method[D]. Wuhan: Wuhan University, 2017.

[35] 王威, 熊艺舟, 王新. NHNet:新型层次化遥感图像语义分割网络[J]. 吉林大学学报(地球科学版), 2024, 54(5): 1764-1772.

[36] Wang Wei, Xiong Yizhou, Wang Xin. NHNet: A Novel Hierarchical Remote Sensing Image Semantic Segmentation Network[J]. Journal of Jilin University (Earth Science Edition), 2024, 54(5): 1764-1772.

基金资助

国家海洋信息中心科技攻关创新基金项目(2301GJZD01)

国家自然科学基金青年基金项目(42206200)