Search

Finding a means to extract water body information efficiently and accurately from high-resolution remote sensing images has been an important research direction in the field of water body extraction in recent years. However, shadows from buildings and other obstacles interfere with the accuracy of water body extraction. To address this problem, this paper proposes a neural network method incorporating an attention mechanism for water body extraction. This paper is based on the U-Net convolutional neural network and adds the squeeze-and-excitation module of SENet, an attention mechanism, to the downsampling process of the U-Net network. The module weights the feature maps so that the network focuses more on the features of the water body information and thus reduces attention to the shadow features from buildings and other features, thus improving the accuracy of image segmentation. The dropout and batch normalization layers are also added to improve the generalization ability and stability of the model. In this paper, a water extraction network SE-CU-Net model is presented to overcome the shadowing effect from buildings and other features. Using GF-2 images of Jiangsu province as the data source, the recognition results of this paper are compared with Dense-Net, Res-Net, Seg-Net, U-net, SVM, and RF. Through the comparison experiments, the model of this paper can not only better overcome the influence of shadows from buildings and other features, but it also has a stronger recognition ability and recognition effect. The average ASCR, Precision, mIoU, OA, F1-Score and kappa coefficients in the three tested areas reached 98.27%, 97.17%, 89.33%, 98.2%, 89.3% and 0.883, respectively, with significantly higher accuracy than the other six classical methods, verifying the effectiveness of the model in overcoming the influence of shadows from buildings and other features in water body extraction research. [ABSTRACT FROM AUTHOR]

The frequency of marine oil spills has increased in recent years. The growing exploitation of marine oil and continuous increase in marine crude oil transportation has caused tremendous damage to the marine ecological environment. Using synthetic aperture radar (SAR) images to monitor marine oil spills can help control the spread of oil spill pollution over time and reduce the economic losses and environmental pollution caused by such spills. However, it is a significant challenge to distinguish between oil-spilled areas and oil-spill-like in SAR images. Semantic segmentation models based on deep learning have been used in this field to address this issue. In addition, this study is dedicated to improving the accuracy of the U-Shape Network (UNet) model in identifying oil spill areas and oil-spill-like areas and alleviating the overfitting problem of the model; a feature merge network (FMNet) is proposed for image segmentation. The global features of SAR image, which are high-frequency component in the frequency domain and represents the boundary between categories, are obtained by a threshold segmentation method. This can weaken the impact of spot noise in SAR image. Then high-dimensional features are extracted from the threshold segmentation results using convolution operation. These features are superimposed with to the down sampling and combined with the high-dimensional features of original image. The proposed model obtains more features, which allows the model to make more accurate decisions. The overall accuracy of the proposed method increased by 1.82% and reached 61.90% compared with the UNet. The recognition accuracy of oil spill areas and oil-spill-like areas increased by approximately 3% and reached 56.33%. The method proposed in this paper not only improves the recognition accuracy of the original model, but also alleviates the overfitting problem of the original model and provides a more effective monitoring method for marine oil spill monitoring. More importantly, the proposed method provides a design principle that opens up new development ideas for the optimization of other deep learning network models. [ABSTRACT FROM AUTHOR]