Deep convolutional neural network (DeCNN) is considered one of promising techniques for classifying the high spatial resolution remote sensing (HSRRS) scenes, due to its powerful feature extraction capabilities. It is well-known that huge high quality labeled datasets are required for achieving the better classification performances and preventing over-fitting, during the training DeCNN model process. However, the lack of high quality datasets limits the applications of DeCNN}. In order to solve this problem, in this paper, we propose a HSRRS image scene classification method using transfer learning and DeCNN (TL-DeCNN) model in few shot HSRRS scene samples. Specifically, three typical DeCNNs of VGG19, ResNet50 and InceptionV3, trained on the ImageNet2015, the weights of their convolutional layer for that of the TL-DeCNN are transferred, respectively. Then, TL-DeCNN just needs to fine-tune its classification module on the few shot HSRRS scene samples in a few epochs. Experimental results indicate that our proposed TL-DeCNN method provides absolute dominance results without over-fitting, when compared with the VGG19, ResNet50 and InceptionV3, directly trained on the few shot samples.

Global navigation satellite system (GNSS)-reflectometry is a type of remote sensing technology and can be applied to soil moisture retrieval. Until now, various GNSS-R soil moisture retrieval methods have been reported. However, there still exist some problems due to the complexity of modeling and retrieval process, as well as the extreme uncertainty of the experimental environment and equipment. To investigate the behavior of bistatic GNSS-R soil moisture retrieval process, two ground-truth measurements with different soil conditions were carried out and the performance of the input variables was analyzed from the mathematical statistical aspect. Moreover, the feature of XGBoost method was utilized as well. As a recently developed ensemble machine learning method, the XGBoost method just emerged for the classification of remote sensing and geographic data, to investigate the characterization of the input variables in the GNSS-R soil moisture retrieval. It showed a good correlation with the statistical analysis of ground-truth measurements. The variable contributions for the input data can also be seen and evaluated. The study of the paper provides some experimental insights into the behavior of the GNSS-R soil moisture retrieval. It is worthwhile before establishing models and can also help with understanding the underlying GNSS-R phenomena and interpreting data.

Urban areas have been focused on recently in remote sensing applications since their function closely relates with the distribution of built-up areas, where reflectivity or scattering characteristics are the same or similar. Traditional pixel-based methods cannot discriminate types of urban built-up areas very well. This paper investigates a deep learning-based classification method for remote sensing images, particularly for high spatial resolution remote sensing (HSRRS) images with various changes and multi-scene classes. Specifically, to help develop corresponding classification methods in urban built-up areas, we consider four deep neural networks (DNN): (1) convolutional neural network (CNN), (2) capsule networks (CapsNet), (3) same model with different training rounding based on CNN (SMDTR-CNN), and (4) same model with different training rounding based on CapsNet (SMDTR-CapsNet). The performances of the proposed methods are evaluated in terms of overall accuracy, kappa coefficient, precision, and a confusion matrix. The results revealed that SMDTR-CNN obtained the best overall accuracy (95.0%) and kappa coefficient (0.944), while also improving the precision of parking lot and resident samples by 1% and 4%, respectively.

Global navigation satellite system-reflectometry (GNSS-R) can retrieve Earth's surface parameters, such as soil moisture (SM) using the reflected signals from GNSS constellations with advantages of noncontact, all-weather, real-time, and continuity, particularly the space-borne cyclone GNSS (CYGNSS) mission. However, the accuracy and efficiency of SM estimation from CYGNSS still need to improve. In this article, the global SM is estimated using machine learning (ML) regression aided by a preclassification strategy. The total observations are classified by land types and corresponding subsets are built for constructing ML regression submodels. Ten-fold cross-validation technique is adopted. The overall performance of SM estimation with/without preclassification is compared, and the results show that the SM estimations using different ML algorithms all have substantial improvement with the preclassification strategy. Then, the optimal XGBoost predicted model with root-mean-square error (RMSE) of 0.052 cm ³ /cm ³ is adopted. In addition, the satisfactory daily and seasonal SM prediction outcomes with an overall correlation coefficient value of 0.86 and an RMSE value of 0.056 cm ³ /cm ³ are achieved at a global scale, respectively. Furthermore, the extensive temporal and spatial variations of CYGNSS SM predictions are evaluated. It shows that the reflectivity plays a main role among the predictors in SM estimation, and the next is vegetation. In some extremely dry places, the roughness may become more important. The value of SM is positively correlated with RMSE and also another limit condition that will constrain the variation of predictors, thus affecting correlation coefficient R and RMSE. Also, we compare both SMAP and CYGNSS SM predictions against in situ SM measurements from 301 stations. Similar low-median unbiased RMSEs are obtained, and the daily averaged CYGNSS-based SM against the in situ networks is 0.049 cm ³ /cm ³ . The presented approach succeeds in providing SM estimation at a global scale with employing the least ancillary data with superior results and this article reveals the spatio-temporal heterogeneity for SM estimation using CYGNSS data.

Global Navigation Satellite System-Reflectometry (GNSS-R) as a microwave remote sensing technique can retrieve the Earth’s surface parameters using the GNSS reflected signal from the surface. These reflected signals convey the surface features and therefore can be utilized to detect certain physical properties of the reflecting surface such as soil moisture content (SMC). Up to now, a serial of electromagnetic models (e.g., bistatic radar and Fresnel equations, etc.) are employed and solved for SMC retrieval. However, due to the uncertainty of the physical characteristics of the sites, complexity, and nonlinearity of the inversion process, etc., it is still challenging to accurately retrieve the soil moisture. The popular machine learning (ML) methods are flexible and able to handle nonlinear problems. It can dig out and model the complex interactions between input and output and ultimately make good predictions. In this paper, two typical ML methods, specifically, random forest (RF) and support vector machine (SVM), are employed for SMC retrieval from GNSS-R data of self-designed experiments (in situ and airborne). A comprehensive simulated dataset involving different types of soil is constructed firstly to represent the complex interactions between the variables (reflectivity, elevation angle, dielectric constant, and SMC) for the requirement of training ML regression models. Correspondingly, the main task of soil moisture retrieval (regression) is addressed. Specifically, the post-processed data (reflectivity and elevation angle) from sensor acquisitions are used to make predictions by these two adopted ML methods and compared with the commonly used GNSS-R retrieval method (electromagnetic models). The results show that the RF outperforms the SVM method, and it is more suitable for handling the inversion problem. Moreover, the RF regression model built by the comprehensive dataset demonstrates satisfactory accuracy and strong universality, especially when the soil type is not uniform or unknown. Furthermore, the typical task of detecting water/soil (classification) is discussed. The ML algorithms demonstrate a high potential and efficiency in SMC retrieval from GNSS-R data.

Accurate prediction of future observations based on past data is the key to near real-time disturbance detection using satellite image time series (SITS). To overcome the limitations of existing methods, we present an attention-based LSTM Encoder-Decoder model in which the historical time series of a pixel is encoded with a bidirectional LSTM encoder while the future time series is produced by another LSTM decoder. An attention mechanism is integrated into the encoder-decoder model to align the input time series with the output time series and to dynamically choose the most relevant contextual information while forecasting. Based on the proposed model, we develop a framework for near real-time disturbance detection and verify its effectiveness in the case of burned area mapping. The prediction accuracy of the proposed model is evaluated using Moderate Resolution Imaging Spectroradiometer (MODIS) time series and compared with state-of-the-art models. Experimental results show that our model achieves the best results in terms of lower prediction error and higher model fitness. We also evaluate the disturbance detection ability of the proposed framework. The proposed approach improves the detection rate of disturbances while suppressing false alarms, and increases the temporal accuracy. We suggest that the proposed methods provide new tools for enhancing current early warning systems in real time.