Remote sensing data imputation using deep learning for multispectral imagery
For water authorities monitoring algal blooms, this work provides a practical method to fill cloud-induced gaps in satellite imagery, though it is an incremental application of existing deep learning architectures to a specific domain.
This study compared deep learning models (CNN, Inception Resnet, Autoencoder, CNN-LSTM, Resnet-LSTM, Autoencoder-LSTM) with linear interpolation for imputing missing spectral bands in PlanetScope SuperDove imagery across four lakes. Deep learning models, particularly CNN, substantially outperformed linear interpolation, enabling more reliable algal bloom monitoring.
Remote sensing techniques have been increasingly utilised in aquatic applications in recent years. A common challenge in using optical satellite data is the presence of missing observations due to cloud cover. These data gaps can lead to missed detection of critical events, such as algal blooms, in lakes of high interest to water authorities. As a result, enhancing the completeness of optical satellite datasets is crucial for improving the monitoring and prediction of algal blooms. In this study, we compared a traditional data imputation method (i.e., linear interpolation) with deep learning models for reconstructing missing spectral bands across four lakes with historical records of algal blooms. The deep learning models adopted include CNN-based architectures (i.e., CNN, Inception Resnet, and Autoencoder) and CNN-LSTM-based architectures (i.e., CNN-LSTM, Resnet-LSTM, and Autoencoder-LSTM). Our results demonstrated that deep learning models substantially outperformed the baseline linear interpolation method in imputing spectral band values within artificially masked regions. Among these models, CNN delivered the best performance across most lakes. Furthermore, we evaluated the performance of algal bloom indices (i.e., Green/Red and NDCI) derived from the imputed imagery by comparing them with the observed data. Our results demonstrate that deep learning models are effective for imputing missing data in PlanetScope SuperDove imagery, enabling more reliable applications in water monitoring.