Search

Multitemporal cross-sensor imagery is fundamental for the monitoring of the Earth’s surface over time. However, these data often lack visual consistency because of variations in the atmospheric and surface conditions, making it challenging to compare and analyze images. Various image-normalization methods have been proposed to address this issue, such as histogram matching and linear regression using iteratively reweighted multivariate alteration detection (IR-MAD). However, these methods have limitations in their ability to maintain important features and their requirement of reference images, which may not be available or may not adequately represent the target images. To overcome these limitations, a relaxation-based algorithm for satellite-image normalization is proposed. The algorithm iteratively adjusts the radiometric values of images by updating the normalization parameters (slope (α) and intercept (β)) until a desired level of consistency is reached. This method was tested on multitemporal cross-sensor-image datasets and showed significant improvements in radiometric consistency compared to other methods. The proposed relaxation algorithm outperformed IR-MAD and the original images in reducing radiometric inconsistencies, maintaining important features, and improving the accuracy (MAE = 2.3; RMSE = 2.8) and consistency of the surface-reflectance values (R2 = 87.56%; Euclidean distance = 2.11; spectral angle mapper = 12.60).

Particulate matter less than 2.5 microns in diameter (PM2.5) is an air pollutant that has become a major environmental concern for governments around the world. Management and control require air quality monitoring and prediction. However, previous studies did not fully utilize the spectral information in multispectral satellite images and land use data in geographic datasets. To alleviate these problems, this study proposes the extraction of land use information not only from geographic inventory but also from satellite images with a machine learning-based classification. In this manner, near up-to-date land use data and spectral information from satellite images can be utilized, and the integration of geographic and remote sensing datasets boosts the accuracy of PM2.5 concentration modeling. In the experiments, Landsat-8 imagery with a 30-m spatial resolution was used, and cloud-free image generation was performed prior to the land cover classification. The proposed method, which uses predictors from geographic and multispectral satellite datasets in modeling, was compared with an approach which utilizes geographic and remote sensing datasets, respectively. Quantitative assessments showed that the proposed method and the developed model, with a performance of RMSE = 3.06 µg/m3 and R2 = 0.85 comparatively outperform the models with a performance of RMSE = 3.14 µg/m3 and R2 = 0.68 for remote sensing datasets and a performance of RMSE = 3.47 µg/m3 and R2 = 0.79 for geographic datasets.

Chlorophyll-a (Chla) concentration, which serves as a phytoplankton substitute in inland waters, is one of the leading indicators for water quality. Generally, water samples are analyzed in professional laboratories, and Chla concentrations are measured regularly for the purpose of water quality monitoring. However, limited spatial water sampling and the labor-intensive nature of data collection make global and long-term monitoring difficult. The developments of remote-sensing optical sensors and technologies make the long-term monitoring of Chla concentrations for an entire water body more achievable. Many studies based on machine learning techniques, such as regression and artificial neural network (ANN) methods, have recently been proposed for Chla concentration estimation using optical satellite images. The methods based on machine learning can achieve accurate estimation. However, overfitting problems may arise because the in situ Chla dataset is generally insufficient to train a complicated machine learning model, which makes trained models inapplicable. In this study, an ANN model containing three convolutional and two fully connected layers with 4953 unknown parameters is designed. A transfer learning method, consisting of model pretraining, main-training, and fine-tuning stages, is proposed to ease the problem of insufficient in situ samples. In the model pretraining stage, the ANN model is pretrained and initialized using samples derived from an existing Chla concentration model. The pretrained ANN model is then fine-tuned using the proposed transfer learning technique with in situ samples collected in five different campaigns carried out during early 2019 from Laguna Lake, the Philippines. Before the transfer learning, data augmentation and rebalancing methods are conducted to enrich the variability and to near-uniformly distribute the in situ samples in Chla concentration space, respectively. To estimate the alleviation of model overfitting, the trained ANN model, using an in situ dataset from Laguna Lake, was tested using an in situ dataset from Lake Victoria, Uganda, obtained in 2019, which has a similar trophic state as Laguna Lake. The experimental results from Sentinel-3 imagery indicated that the overfitting problem was significantly alleviated and the trained ANN model outperformed related models in terms of the root-mean-squared error of the estimated Chla concentrations.

The retrieval of chlorophyll-a (Chl-a) concentrations relies on empirical or analytical analyses, which generally experience difficulties from the diversity of inland waters in statistical analyses and the complexity of radiative transfer equations in analytical analyses, respectively. Previous studies proposed the utilization of artificial neural networks (ANNs) to alleviate these problems. However, ANNs do not consider the problem of insufficient in situ samples during model training, and they do not fully utilize the spatial and spectral information of remote sensing images in neural networks. In this study, a two-stage training is introduced to address the problem regarding sample insufficiency. The neural network is pretrained using the samples derived from an existing Chl-a concentration model in the first stage, and the pretrained model is refined with in situ samples in the second stage. A novel convolutional neural network for Chl-a concentration retrieval called WaterNet is proposed which utilizes both spectral and spatial information of remote sensing images. In addition, an end-to-end structure that integrates feature extraction, band expansion, and Chl-a estimation into the neural network leads to an efficient and effective Chl-a concentration retrieval. In experiments, Sentinel-3 images with the same acquisition days of in situ measurements over Laguna Lake in the Philippines were used to train and evaluate WaterNet. The quantitative analyses show that the two-stage training is more likely than the one-stage training to reach the global optimum in the optimization, and WaterNet with two-stage training outperforms, in terms of estimation accuracy, related ANN-based and band-combination-based Chl-a concentration models.