Deep neural network-based spatiotemporal heterogeneous data reconstruction for landslide detection

Landslides could cause huge threats to lives and cause property damages. In the landslide prediction system, environmental information can be collected through sensors to detect the possibility of landslide occurrences. However, the data collected by wireless sensor network systems (WSNs) may be lost due to sensor failures, external interferences, or other environmental factors, which may affect the accuracy of landslide predictions. In order to solve the problem of missing data, we propose a data reconstruction method based on rainfall intensity, soil moisture, slope, and slope direction and reconstruct missing data based on heterogeneous data and temporal and spatial relationships. A convolutional long short-term memory (ConvLSTM) deep neural network is trained to predict the missing time slot data. We use the predicted data to compensate for missing data. The results of the experiments show that the factor of safety of ConvLSTM achieves better RMSE in almost all of the missing data types and rates than LSTM. The mean and stdev forecast error of gradual fading with ConvLSTM at missing rate 30% are -0.001 and 0.033, respectively.