Landslide detection from an open satellite imagery and digital elevation model dataset using attention boosted convoluti
- PDF / 13,388,978 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 53 Downloads / 480 Views
Shunping Ji I Dawen Yu I Chaoyong Shen I Weile Li I Qiang Xu
Landslide detection from an open satellite imagery and digital elevation model dataset using attention boosted convolutional neural networks
Abstract Convolution neural network (CNN) is an effective and popular deep learning method which automatically learns complicated non-linear mapping from original inputs to given labels or ground truth through a series of convolutional layers. This study focuses on detecting landslides from high-resolution optical satellite images using CNN-based methods, providing opportunities for recognizing latent landslides and updating large-scale landslide inventory with high accuracy and time efficiency. Considering the variety of landslides and complicated backgrounds, attention mechanisms originated from the human visual system are developed for boosting the CNN to extract more distinctive feature representations of landslides from backgrounds. As deep learning needs a large number of labeled data to train a learning model, we manually prepared a landslide dataset which is located in the Bijie city, China. In the dataset, 770 landslides, including rock falls, rock slides, and a few debris slides, were interpreted by geologists from the satellite images and digital elevation model (DEM) data and further checked by fieldwork. The landslide data was separated into a training set that trains the attention boosted CNN model and a testing set that evaluates the performance of the model with a ratio of 2:1. The experimental results showed that the best F1score of landslide detection reached 96.62%. The results also proved that the performance of our spatial-channel attention mechanism was fairly over other recent attention mechanisms. Additionally, the effectiveness of predicting new potential landslides with high efficiency based on our dataset is demonstrated. Keywords Landslide detection . Satellite optical images . Convolution neural network . Attention mechanism . Remote sensing landslide dataset Introduction As a common and frequent geological disaster, landslide causes severe damages to natural environments, properties, and personal safety all over the world. For example, more than 200 people died from landslide disasters in Guizhou province of China between 2013 and 2017. A landslide may be triggered by several factors, such as seismic shaking, heavy rainfalls, and human activities, and result in the downward and outward movement of slop-forming materials including rock, soil, or their combination (Cruden and Varnes 1996; Cruden 1991). There are various studies based on field surveying or remote sensing data, which apply a wide range of methods on landslide monitory, detection, potential landslide prediction, and hazard mitigation. Field surveying for discovering potential landslide regions and updating landslide inventories are common and reliable strategies, but they are time-consuming, costly, and inefficient. With the rapid advancement of remote sensing technology, automatic landslide detection from aerial InSAR or sa
Data Loading...
