Shallowwater depth plays an important role in marine development, navigation safety, and environmental protection. It is an efficient and economical way to obtain water depth by remote sensing technology. At present, most empirical models based on multispectral image usually obtain water depth by the relationship between the sea surface reflectance (SSR) (a single pixel) and in situ water depth, it is a one-to-one correspondence between the reflectance and depth. However, seafloor substrate and inherent optical properties (IOP) will also have contribution to the SSR. In this article, we propose an adjacent pixels multilayer perceptron model (APMLP) model using adjacent pixels to weaken the influence of seafloor substrate and IOP.Datasets on Oahu Island (Sentinel-2B, LIDAR in situ data) and Saint Thomas Island (Sentinel-2A, LIDAR in situ data) are used to establish and verify the model. The APMLP model are also compared with the multilayer perceptron model (MLP) model, BP neural network model, and Log-ratio model. The overall root-mean-square error (RMSE) of APMLP model on Oahu Island is 0.72 m, which is much better than the other three models (MLP 1.07 m, BP 1.05 m, Log-ratio 1.52 m). Similar results are obtained from the Saint Thomas Island dataset, RMSE of APMLP model is 1.56 m, better than the other three (MLP 1.91 m, BP 1.89 m, Log-ratio 2.39 m). The study confirms that considering adjacent pixels in an artificial neural network model can effectively improve the performance of water depth retrieval.
20 November 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Remote sensing of chlorophyll-a (Chl-a) concentrations in coastal waters is of great importance for the assessment of the marine ecological conditions. However, due to the complex water body optical properties, accurate selection of optimal feature bands is limited, which poses a great challenge for high-precision retrieval. The purpose of this research is to address the problem of high-precision retrieval of the Chl-a concentration in small coastal waters. In this article, a singular value decomposition and deep neural network (SVD-DNN) Chl-a inversion model for Hong Kong coastal waters were constructed. Other machine learning methods, such as random forest (RF) and the support vector machine (SVM) are also used to establish the inversion models for the comparison. At the same time, a comparative analysis was performed with Chl-a retrieval models created using a feature selection method based on the correlation of band combinations. These models are validated using the Landsat 8 OLI and synchronously measured Chl-a dataset (N = 149 samples). Results show that the developed SVD-DNN model has the best retrieval accuracy [mean R = 0.90, root mean square error (RMSE) = 1.21, mean absolute error (MAE) = 1.05], outperforming the SVD-RF and SVD-SVM models. The SVD-DNN model shows superior retrieval performance when the Chl-a concentration is below 6 micrograms per liter (RMSE = 0.66, MAE = 0.67). The proposed model also shows better temporal generalization ability in 2013, 2014, and 2016 compared to the other models. This study demonstrates that the developed and validated SVD-DNN model has excellent robustness and generalization ability and can be used in combination with Landsat data for long-term retrieval of Chl-a across different time series.
20 October 2021 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
The log-ratio model (LRM) with three model parameters ( ${\boldsymbol{n}}$ , ${{\boldsymbol{m}}_1},$ and ${{\boldsymbol{m}}_0}$ ) has been widely used in shallow water bathymetry of multispectral images. At present, obtaining LRM model parameters by optimization algorithms depends on the setting of initial values [ ${\boldsymbol{n}}(0)$ , ${{\boldsymbol{m}}_1}(0),$ and ${{\boldsymbol{m}}_0}(0)$ ] or initial value ranges. However, how to set the initial value can obtain model parameters more effectively and credibly, and the significance of each initial value in the optimization process is very important for LRM bathymetric retrieval method. In this article, remote sensing images and bathymetric data were used to determine the initial value ranges of credible model parameters, and then the sensitivity of each parameter was analyzed in the algorithm optimization process from the partial differential perspective, and the bathymetry was retrieved in the sea area near Ganquan Island. The results show that according to the partial differential analysis of model parameters, ${\boldsymbol{n}}$ , with the smallest change rate in three LRM parameters and the rate of ${{\boldsymbol{m}}_1}$ and ${{\boldsymbol{m}}_0}$ is at least five times of it. So ${\boldsymbol{n}}(0)$ needs more iterations to achieve the optimal. Therefore, the initial value ${\boldsymbol{n}}(0)$ determines whether the optimization method can obtain credible LRM model parameters and the algorithm convergence speed. And when ${\boldsymbol{n}}(0) \in ({1/{{\boldsymbol{R}}_{{{\bf min}}}},\ 1000} ]$ ( ${{\boldsymbol{R}}_{{{\bf min}}}}$ is the minimum reflectance in blue and green bands), there must be a credible model parameter solution. The bathymetric retrieval results of the sea area with coral and sand substrates show the applicability of our initial value ranges in different substrate area.
Predicting the shape evolution and movement of remote sensing satellite cloud images is a difficult task requiring the effective monitoring and rapid prediction of thunderstorms, gales, rainstorms, and other disastrous weather conditions. We proposed a generative adversarial network (GAN) model for time series satellite cloud image prediction in this research. Taking time series information as the constraint condition and abandoning the assumption of linear and stable changes in cloud clusters in traditional methods, the GAN model is used to automatically learn the data feature distribution of satellite cloud images and predict time series cloud images in the future. Through comparative experiments and analysis, the Mish activation function is selected for integration into the model. On this basis, three improvement measures are proposed: (1) The Wasserstein distance is used to ensure the normal update of the GAN model parameters; (2) establish a multiscale network structure to improve the long-term performance of model prediction; (3) combined image gradient difference loss (GDL) to improve the sharpness of prediction cloud images. The experimental results showed that for the prediction cloud images of the next four times, compared with the unimproved Mish-GAN model, the improved GDL-GAN model improves the PSNR and SSIM by 0.44 and 0.02 on average, and decreases the MAE and RMSE by 18.84% and 7.60% on average. It is proven that the improved GDL-GAN model can maintain good visualization effects while keeping the overall changes and movement trends of the prediction cloud images relatively accurate, which is helpful to achieve more accurate weather forecast. The cooperation ability of satellite cloud images in disastrous weather forecasting and early warning is enhanced.
20 September 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Many current satellite-derived bathymetry (SDB) methods usually rely on in-situ water depth, which limits their applications. In this study, an ICESat-2 assisted Dual-Band Model (IDBM) is proposed, which can be used to derive bathymetry independent of in-situ water depth. First, the Sentinel-2 blue and green band diffuse attenuation coefficient is derived using ICESat-2 lidar photon data from adjacent deep water according to the Jerlov Kd spectral curves and the sensor spectral response function. Then, these two coefficients are applied to the IDBM model to derive the water depth. Datasets for four study areas are used to validate the feasibility of the proposed IDBM model, in addition, the original QAA (quasi-analytical algorithm) assisted Dual-Band Model (QDBM) is also used to derive bathymetry for comparison purposes. Experiment results show that the proposed IDBM model is effective and can achieve higher or similar accuracy compared to the original QDBM model. The average MAE and RMSE of the IDBM model reach 1.23m and 1.67m, while those of the QDBM model are 1.44m and 1.91m, respectively. The performance of the IDBM model has improved to some extent. Especially for the SI case, the MAE and RMSE of the QDBM model are 1.46m and 2.00m, while those of the IDBM model are 0.99m and 1.55m, which are reduced by 0.47m and 0.45m. In conclusion, the proposed IDBM model is feasible and effective to derive bathymetry without in-situ water depth.
Landfast sea ice (LFSI) refers to sea ice attached to the shoreline with little or no horizonal motion in contrast to drifting sea ice. The LFSI plays an important role in the Arctic marine environmental and biological systems. Therefore, it is crucial to accurately monitor the spatiotemporal changes in the LFSI distribution. Here we present an automatic LFSI retrieval method for the Laptev Sea, eastern Arctic Ocean, based on a conditional generative adversarial network Pix2Pix using the true-color images of Moderate Resolution Imaging Spectroradiometer (MODIS). The spatial resolution of the derived product is 1.25 km, with a temporal interval of 7 days. Compared to the manually identified data from the true-color images of MODIS, the average precision of the LFSI area derived from LFSI mapping model reaches 91.4%, with the recall reaching 98.7% and F1-score reaching 94.5%. The LFSI coverage is consistent with the traditional large-scale LFSI products, but provides more details. Intraseasonal and interannual variations in LFSI area of the Laptev Sea in spring (March–May) during the period of 2002–2021 are investigated using the new product. The spring LFSI area in this region decreases at a rate of 0.67 × 103 km2 per year during this period (R2 = 0.117, p < 0.01). According to the spatial and temporal changes, we conclude that the LFSI is becoming more stable while the area is shrinking. The method is fully-automatic and computationally efficient, which can be further applied to the entire Arctic Ocean for LFSI identification and monitoring.