Cloud Detection Algorithm for Multi-Satellite Remote Sensing Imagery Based on a Spectral Library and 1D Convolutional Neural Network

Abstract

Automatic cloud detection in remote sensing images is of great significance. Deep-learning-based methods can achieve cloud detection with high accuracy; however, network training heavily relies on a large number of labels. Manually labelling pixel-wise level cloud and non-cloud annotations for many remote sensing images is laborious and requires expert-level knowledge. Different types of satellite images cannot share a set of training data, due to the difference in spectral range and spatial resolution between them. Hence, labelled samples in each upcoming satellite image are required to train a new deep-learning-based model. In order to overcome such a limitation, a novel cloud detection algorithm based on a spectral library and convolutional neural network (CD-SLCNN) was proposed in this paper. In this method, the residual learning and one-dimensional CNN (Res-1D-CNN) was used to accurately capture the spectral information of the pixels based on the prior spectral library, effectively preventing errors due to the uncertainties in thin clouds, broken clouds, and clear-sky pixels during remote sensing interpretation. Benefiting from data simulation, the method is suitable for the cloud detection of different types of multispectral data. A total of 62 Landsat-8 Operational Land Imagers (OLI), 25 Moderate Resolution Imaging Spectroradiometers (MODIS), and 20 Sentinel-2 satellite images acquired at different times and over different types of underlying surfaces, such as a high vegetation coverage, urban area, bare soil, water, and mountains, were used for cloud detection validation and quantitative analysis, and the cloud detection results were compared with the results from the function of the mask, MODIS cloud mask, support vector machine, and random forest. The comparison revealed that the CD-SLCNN method achieved the best performance, with a higher overall accuracy (95.6%, 95.36%, 94.27%) and mean intersection over union (77.82%, 77.94%, 77.23%) on the Landsat-8 OLI, MODIS, and Sentinel-2 data, respectively. The CD-SLCNN algorithm produced consistent results with a more accurate cloud contour on thick, thin, and broken clouds over a diverse underlying surface, and had a stable performance regarding bright surfaces, such as buildings, ice, and snow.

A Priori Land Surface Reflectance Synergized With Multiscale Features Convolution Neural Network for MODIS Imagery Cloud Detection

“…Because the surface reflectance database is built using high-temporal-resolution images, their approach is best suited for images acquired in short revisit intervals. In general, parameter adaptation and global optimization are typically difficult to perform for threshold-based methods due to the complexity of the surface environment and the diversity of cloud geometry, leading to varying degrees of cloud cover estimation bias [13].…”

“…Jeppesen et al [36] proposed RS-Net for achieving promising cloud detection results, although it had limited multispectral capabilities. To reduce the annotation of training samples, Ma et al [13] combined ASTER library and AVIRIS spectral images using convolutional neural network (CNN) to achieve cloud detection for multi-sensor remote sensing imageries. Domain adaption [38,39] were also introduced to improve the cloud detection performance.…”

Probabilistic cloud mask from visible and infrared sensors on‐board geostationary platform

“…Machine‐learning algorithms – like pattern recognition (Ebert, 1987), fuzzy logic (Baum et al ., 1997; Ghosh et al ., 2006), decision trees (Scaramuzza et al ., 2012; Kilpatrick et al ., 2019), neural networks (Hughes and Hayes, 2014), support vector machines (Pengfei et al ., 2015), and deep learning (Chen et al ., 2018; Mateo‐Garcia et al ., 2018; Jeppesen et al ., 2019; Ma et al ., 2021) – have also been attempted for the automatic cloud screening of satellite observations. These methods are based on the training of textural and spectral features of the observations for the identification of cloudy pixels.…”

A Supplementary Module to Improve Accuracy of the Quality Assessment Band in Landsat Cloud Images

“…Moreover, some methods further include more other features, such as the lower temperature of clouds estimated from the thermal infrared bands and the geometric relationship between clouds and cloud shadows. Instead of using these well-defined cloud features, some recent studies employ machine-learning algorithms (e.g., deep learning models) to automatically extract cloud and cloud shadow features [12][13][14][15]. The second category is the multi-temporal methods, which employ the temporal information provided by the images acquired at other times (called "reference images") [16][17][18][19][20][21].…”

“…agreement of clouds and shadows agreement of clouds and shadows+commission of clouds and shadows (13) Omission = 1 − agreement of clouds and shadows agreement of clouds and shadows+omission of clouds and shadows (14) F1 = 2 × agreement of clouds and shadows 2 × agreement of clouds and shadows+omission of clouds and shadows+commission of clouds and shadows(15) …”