An improved double-branch network method for intelligently extracting marine cage culture area
ZHENG Zhiteng1(), FAN Haisheng2, WANG Jie3, WU Yanlan1,4, WANG Biao1(), HUANG Tengjie2
1. School of Resources and Environmental Engineering, Anhui University, Hefei 230601, China 2. Zhuhai Obit Aerospace Technology Co., Ltd., Zhuhai 519080, China 3. Geodetic Data Processing Center, Ministry of Natural Resources, Xi'an 710054, China 4. Anhui Engineering Research Center for Geographical Information Intelligent Technology, Hefei 230601, China
针对传统基于遥感影像数据提取网箱信息中存在的精度低、“异物同谱”、“椒盐”噪声等问题。基于高分二号卫星(Gaofen-2 satellite,GF-2)数据,提出了一种改进的双支网络模型网箱信息提取方法。该模型在空间编码路径上利用密集连接块提取网箱的空间特征信息,在全局编码路径上利用全局平均池化快速获得网箱的全局上下文信息,最终通过特征融合来丰富网箱空间细节特征信息和深层判别特征信息,提高了网箱的提取精度。本文方法在精确率、交并比(intersection over union,IOU)和F1分数这3个评价指标上分别取得了87.37%,72.56%和82.47%的得分,与精度最高的传统方法相比分别提高了7.82,4.12和4.64百分点,与经典的深度学习模型相比较在IOU和F1上也取得了8.43和8.69百分点的提高。实验表明,这一方法能很好地满足海水网箱养殖区的提取工作,此方法可以为近海海水网箱养殖业的监管和调控提供技术支持。
For the traditional remote sensing image data extraction, the accuracy of the cage information is low, and there exist the problems of “different object with the same spectrum” and “salt and salt” noise. Based on the Gaofen-2 satellite (“GF-2”) data, this paper proposes an improved double-branch network model cage information extraction method. The model uses the dense connection block to extract the spatial feature information of the cage on the spatial coding path, obtains the global context information of the cage quickly by using the global average pooling on the global coding path, and finally enriches the detailed information of the cage space through feature fusion. And deep discriminant feature information improves the extraction accuracy of the cage. The method has achieved scores of 87.37%, 72.56%, and 82.47% on the three evaluation indicators of precision, IOU, and F1, respectively, which are 7.82, 4.12, and 4.64 percentage points higher than the traditional method with the highest accuracy, respectively. The deep learning model has also achieved an increase of 8.43 and 8.69 percentage points in IOU and F1. Experiments show that the method can meet the extraction work of sea cage culture area, and can provide technical support for the regulation and regulation of offshore sea cage culture.
Shen X D. Analysis on the current situation and prospect of sea cage aquaculture in China[C]// 2008 Symposium on Marine Culture and Coastal Economic Development, 2008:178-181.
[3]
郑晶予. 地理信息系统在水产养殖中的发展应用[J]. 河北渔业, 2011(12):55-57.
Zheng J Y. Comparison and analysis of the nutrient components of Larimichthys crocea in seine net and traditional net cage[J]. Hebei Fisheries, 2011(12):55-57.
Xia L H, Liu Z G, Xu S. Estim ation on pollution load and analysis on spatial characteristics of m ariculture in Zhelin Bay[J]. Guangdong Agricultural Science, 2017,44(12):151-157.
Xu Y J, Zhang Z X, Wang X, et al. Remote sensing monitoring and analysis of coastal aquaculture land in Shandong Province in recent 30 years[J]. Journal of Earth Information Science, 2014,16(3):482-489.
Li X G, Jiang N, Yang Y B, et al. Remote sensing investigation and survey of lake reclamation and enclosure aquaculture in lake TaiHu[J]. Transactions of Oceanology and Limnology, 2006(1):93-99.
Wu Y J, Zhang J H, Tian G H, et al. A survey to aquiculture with remote sensing technology in Hainan Province[J]. Chinese Journal of Tropical Crops, 2006(2):108-111.
Ma Y J, Zhao D L, Wang R M, et al. Offshore aquatic farming areas extraction method based on ASTER data[J]. Transactions of the Chinese Society of Agricultural Engineering, 2010,26(s2):120-124.
Chu J L, Zhao D Z, Zhang F S. Wakame raft interpretation method of remote sensing based on association rules[J]. Remote Sensing Technology and Application, 2012,27(6):941-946.
Lu Y W, Li Q Z, Du X, et al. A method of coastal aquaculture area automatic extraction with spatial resolution images[J]. Remote Sensing Technology and Application, 2015,30(3):486-494.
[13]
Rajitha K, Mukherjee C K, Chandran R V, et al. Applications of remote sensing and GIS for sustainable management of shrimp culture in India[J]. Aquacultural Engineering, 2007,36(1):1-17.
Xie Y L, Wang M, Zhang X Y. An object-oriented approach for extracting farm waters within coastal belts[J]. Remote Sensing Technology and Application, 2009,24(1):68-72.
Wu Y T. Research on coastal aquaculture detection using remote sensing images. Beijng:The University of Chinese Academy of Sciences(Institute of Remote Sensing and Digital Earth,Chinese Academy of Sciences), 2017.
Sun Y Y, Ying J L, Wang M, et al. Object-oriented classification and dynamic monitoring of sea information for seawater culture[J]. Ocean Information, 2016(2):6-11.
Guang X B, Zhang C P, Jiang J S, et al. Remote sensing monitoring of aquaculture and automatic information extraction[J]. Remote Sensing for Land and Resources, 2009,21(2):44-47.doi: 10.6046/gtzyyg.2009.02.09.
[18]
王晓轩. 基于面向对象的海岸带水产养殖模式识别[D]. 广州:广州大学, 2011.
Wang X X. Recognition of aquaculture types in coastal zone based on object-oriented method[D]. Guangzhou:Guangzhou University, 2011.
[19]
Long J, Shelhamer E, Darrell T, et al. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014,39(4):640-651.
pmid: 27244717
[20]
Fu G, Liu C J, Zhou R, et al. Classification for high resolution remote sensing imagery using a fully convolutional network[J]. Remote Sensing, 2017,9,498:1-21.
[21]
Jia S, Liu H, Sun F. Aerial scene classification with convolutional neural networks[J]. Advances in Neural Networks, 2015,9377:258-265.
Wei Q, Ai L M, Ye X N. An automatic road extraction method for high-resolution remote sensing images[J]. Computer Technology and Development, 2019(6):130-133.
He H, Wang S C, Yang D F, et al. An road extraction method for remote sensing image based on Encoder Decoder network[J]. Acta Geodaetica et Cartographica Sinica, 2019,48(3):66-74.
[24]
Xu Y Y, Wu L, Xie Z, et al. Building extraction in very high resolution remote sensing imagery using deep learning and guided filters[J]. Remote Sensing, 2018,10(1),144:1-18.
[25]
Yang H, Wu P H, Yao X D. Building extraction in very high resolution imagery by Dense-Attention networks[J]. Remote Sensing, 2018,10(11),1768:1-23.
[26]
Yao X D, Yang H, Wu Y L, et al. Land use classification of the deep convolutional neural network method reducing the loss of spatial features[J]. Sensors, 2019,19(12),2792:1-16.
[27]
Yu C Q, Wang J B, Peng C, et al. BiSeNet:Bilateral segmentation network for real-time semantic segmentation[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2018:1-17.
[28]
Wang H Z, Wang Y, Zhang Q, et al. Gated convolutional neural network for semantic segmentation in high-resolution images[J]. Remote Sensing, 2017,9(5),446:1-15.
[29]
Gao H, Liu Z, M LV D, et al. Densely connected convolutional networks[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2017:1-9.
[30]
Jegou S, Drozdzal M, Vazquez D, et al. The one hundred layers tiramisu:Fully convolutional DenseNets for semantic segmentation[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2016:1-9.
[31]
Li L, Jian L, Min W, et al. A multiple-feature reuse network to extract buildings from remote sensing imagery[J]. Remote Sensing. 2018,10(9):1350:651-663.
[32]
Ioffe S, Szegedy C. Batch normalization:Accelerating deep network training by reducing internal covariate shift[C]// Machine Learning, 2015:1-11.
[33]
Chen L C, Papandreou G, Schroff F, et al. Rethinking atrous convolution for semantic image segmentation.[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2017:1-14.
[34]
Zhao H S, Shi J P, Qi X J, et al. Pyramid scene parsing network[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2017:1-11.
[35]
Liu W, Rabinovich A, Berg A C, et al. ParseNet:Looking wider to see better[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2015:1-11.
[36]
Awrangjeb M, Fraser C S. An automatic and threshold-free performance evaluation system for building extraction techniques from airborne LIDAR data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014,7:4184-4198.
[37]
Ronneberger O, Fischer P, Brox T, et al. U-Net:Convolutional networks for biomedical image segmentation[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2015:234-241.
[38]
Badrinarayanan V, Kendall A, Cipolla R, et al. SegNet:A deep convolutional encoder-decoder architecture for image segmentation[C]// IEEE Computer Vision and Pattern Recognition. IEEE, 2015:1-14.
Wang F, Xia L H, Chen Z B, et al. Remote sensing identification of coastal zone mariculture modes based on association-rules object-oriented method[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018,34(12):210-217.