scholarly journals Void Filling of Digital Elevation Models with a Terrain Texture Learning Model Based on Generative Adversarial Networks

2019 ◽  
Vol 11 (23) ◽  
pp. 2829 ◽  
Author(s):  
Qiu ◽  
Yue ◽  
Liu

Digital elevation models (DEMs) are an important information source for spatial modeling. However, data voids, which commonly exist in regions with rugged topography, result in incomplete DEM products, and thus significantly degrade DEM data quality. Interpolation methods are commonly used to fill voids of small sizes. For large-scale voids, multi-source fusion is an effective solution. Nevertheless, high-quality auxiliary source information is always difficult to retrieve in rugged mountainous areas. Thus, the void filling task is still a challenge. In this paper, we proposed a method based on a deep convolutional generative adversarial network (DCGAN) to address the problem of DEM void filling. A terrain texture generation model (TTGM) was constructed based on the DCGAN framework. Elevation, terrain slope, and relief degree composed the samples in the training set to better depict the terrain textural features of the DEM data. Moreover, the resize-convolution was utilized to replace the traditional deconvolution process to overcome the staircase in the generated data. The TTGM was trained on non-void SRTM (Shuttle Radar Topography Mission) 1-arc-second data patches in mountainous regions collected across the globe. Then, information neighboring the voids was involved in order to infer the latent encoding for the missing areas approximated to the distribution of training data. This was implemented with a loss function composed of pixel-wise, contextual, and perceptual constraints during the reconstruction process. The most appropriate fill surface generated by the TTGM was then employed to fill the voids, and Poisson blending was performed as a postprocessing step. Two models with different input sizes (64 × 64 and 128 × 128 pixels) were trained, so the proposed method can efficiently adapt to different sizes of voids. The experimental results indicate that the proposed method can obtain results with good visual perception and reconstruction accuracy, and is superior to classical interpolation methods.

Author(s):  
Hengyi Cai ◽  
Hongshen Chen ◽  
Yonghao Song ◽  
Xiaofang Zhao ◽  
Dawei Yin

Humans benefit from previous experiences when taking actions. Similarly, related examples from the training data also provide exemplary information for neural dialogue models when responding to a given input message. However, effectively fusing such exemplary information into dialogue generation is non-trivial: useful exemplars are required to be not only literally-similar, but also topic-related with the given context. Noisy exemplars impair the neural dialogue models understanding the conversation topics and even corrupt the response generation. To address the issues, we propose an exemplar guided neural dialogue generation model where exemplar responses are retrieved in terms of both the text similarity and the topic proximity through a two-stage exemplar retrieval model. In the first stage, a small subset of conversations is retrieved from a training set given a dialogue context. These candidate exemplars are then finely ranked regarding the topical proximity to choose the best-matched exemplar response. To further induce the neural dialogue generation model consulting the exemplar response and the conversation topics more faithfully, we introduce a multi-source sampling mechanism to provide the dialogue model with both local exemplary semantics and global topical guidance during decoding. Empirical evaluations on a large-scale conversation dataset show that the proposed approach significantly outperforms the state-of-the-art in terms of both the quantitative metrics and human evaluations.


Geosciences ◽  
2019 ◽  
Vol 9 (3) ◽  
pp. 117 ◽  
Author(s):  
František Chudý ◽  
Martina Slámová ◽  
Julián Tomaštík ◽  
Roberta Prokešová ◽  
Martin Mokroš

An active gully-related landslide system is located in a deep valley under forest canopy cover. Generally, point clouds from forested areas have a lack of data connectivity, and optical parameters of scanning cameras lead to different densities of point clouds. Data noise or systematic errors (missing data) make the automatic identification of landforms under tree canopy problematic or impossible. We processed, analyzed, and interpreted data from a large-scale landslide survey, which were acquired by the light detection and ranging (LiDAR) technology, remotely piloted aircraft system (RPAS), and close-range photogrammetry (CRP) using the ‘Structure-from-Motion’ (SfM) method. LAStools is a highly efficient Geographic Information System (GIS) tool for point clouds pre-processing and creating precise digital elevation models (DEMs). The main landslide body and its landforms indicating the landslide activity were detected and delineated in DEM-derivatives. Identification of micro-scale landforms in precise DEMs at large scales allow the monitoring and the assessment of these active parts of landslides that are invisible in digital terrain models at smaller scales (obtained from aerial LiDAR or from RPAS) due to insufficient data density or the presence of many data gaps.


2020 ◽  
Vol 12 (18) ◽  
pp. 3016
Author(s):  
Ignacio Borlaf-Mena ◽  
Maurizio Santoro ◽  
Ludovic Villard ◽  
Ovidiu Badea ◽  
Mihai Andrei Tanase

Spaceborne remote sensing can track ecosystems changes thanks to continuous and systematic coverage at short revisit intervals. Active remote sensing from synthetic aperture radar (SAR) sensors allows day and night imaging as they are not affected by cloud cover and solar illumination and can capture unique information about its targets. However, SAR observations are affected by the coupled effect of viewing geometry and terrain topography. The study aims to assess the impact of global digital elevation models (DEMs) on the normalization of Sentinel-1 backscattered intensity and interferometric coherence. For each DEM, we analyzed the difference between orbit tracks, the difference with results obtained with a high-resolution local DEM, and the impact on land cover classification. Tests were carried out at two sites located in mountainous regions in Romania and Spain using the SRTM (Shuttle Radar Topography Mission, 30 m), AW3D (ALOS (Advanced Land Observation Satellite) World 3D, 30 m), TanDEM-X (12.5, 30, 90 m), and Spain national ALS (aerial laser scanning) based DEM (5 m resolution). The TanDEM-X DEM was the global DEM most suitable for topographic normalization, since it provided the smallest differences between orbital tracks, up to 3.5 dB smaller than with other DEMs for peak landform, and 1.4–1.9 dB for pit and valley landforms.


2006 ◽  
Vol 42 (8) ◽  
Author(s):  
Adriano Rolim Paz ◽  
Walter Collischonn ◽  
André Luiz Lopes da Silveira

2021 ◽  
Vol 47 (4) ◽  
pp. 191-199
Author(s):  
Vadim Belenok ◽  
Yuriy Velikodsky ◽  
Oleksandr Nikolaienko ◽  
Nataliia Rul ◽  
Sergiy Kryachok ◽  
...  

The article considers the question of estimating the accuracy of interpolation methods for building digital elevation models using Soviet topographic maps. The territory of the Kursk region of the Russian Federation was used as the study area, because it is located on the Central Russian Upland and characterized by the complex structure of the vertical and horizontal dissection of the relief. Contour lines automatically obtained using a Python algorithm were used as the initial elevation data to build a digital elevation model. Digital elevation models obtained by thirteen different interpolation methods in ArcGIS and Surfer software were built and analyzed. Special attention is paid to the ANUDEM method, which allows to obtain hydrologically correct digital elevation models. Recommendations for the use of one or another method of interpolation are given. The results can be useful for professionals who use topographic maps in their work and deals with the design using digital elevation models.


Sign in / Sign up

Export Citation Format

Share Document