An official website of the United States government
Semantic segmentation algorithms, such as UNet, that rely on convolutional neural network (CNN)-based architectures, due to their ability to capture local textures and spatial context, have shown promise for anthropogenic geomorphic feature extraction when using land surface parameters (LSPs) derived from digital terrain models (DTMs) as input predictor variables. However, the operationalization of these supervised classification methods is limited by a lack of large volumes of quality training data. This study explores the use of transfer learning, where information learned from another, and often much larger, dataset is used to potentially reduce the need for a large, problem-specific training dataset. Two anthropogenic geomorphic feature extraction problems are explored: the extraction of agricultural terraces and the mapping of surface coal mine reclamation-related valley fill faces. Light detection and ranging (LiDAR)-derived DTMs were used to generate LSPs. We developed custom transfer parameters by attempting to predict geomorphon-based landforms using a large dataset of digital terrain data provided by the United States Geological Survey’s 3D Elevation Program (3DEP). We also explored the use of pre-trained ImageNet parameters and initializing models using parameters learned from the other mapping task investigated. The geomorphon-based transfer learning resulted in the poorest performance while the ImageNet-based parameters generally improved performance in comparison to a random parameter initialization, even when the encoder was frozen or not trained. Transfer learning between the different geomorphic datasets offered minimal benefits. We suggest that pre-trained models developed using large, image-based datasets may be of value for anthropogenic geomorphic feature extraction from LSPs even given the data and task disparities. More specifically, ImageNet-based parameters should be considered as an initialization state for the encoder component of semantic segmentation architectures applied to anthropogenic geomorphic feature extraction even when using non-RGB image-based predictor variables, such as LSPs. The value of transfer learning between the different geomorphic mapping tasks may have been limited due to smaller sample sizes, which highlights the need for continued research in using unsupervised and semi-supervised learning methods, especially given the large volume of digital terrain data available, despite the lack of associated labels.
more »
« less
Effects of Neural Network Architecture on Topography Estimation From Satellite Imagery for Multi-Terrain Autonomous Vehicle Path Planning and Control
Global warming is one of the world’s most pressing issues. The study of its effects on the polar ice caps and other arctic environments, however, can be hindered by the often dangerous and difficult to navigate terrain found there. Multi-terrain autonomous vehicles can assist researchers by providing a mobile platform on which to collect data in these harsh environments while avoiding any risk to human life and speeding up the research process. The mechanical design and ultimate efficacy of these autonomous robotic vehicles depends largely on the specific missions they are deployed for, but terrain conditions can vary wildly geographically as well as seasonally, making mission planning for these unmanned vehicles more difficult. This paper proposes the use of various UNet-based neural network architectures to generate digital elevation maps from satellite images, and explores and compares their efficacy on a single set of training and validation datasets generated from satellite imagery. These digital elevation maps generated by the model could be used by researchers not only to track the change in arctic topography over time, but to quickly provide autonomous exploratory research rovers with the topographical information necessary to decide on optimal paths during the mission. This paper analyzes different model architectures and training schemes: a traditional UNet, a traditional UNet with data augmentation, a UNet with a single active skip-layer vision transformer (ViT), and a UNet with multiple active skip-layer ViT. Each model was trained on a dataset of satellite images and corresponding digital elevation maps of Ellesmere Island, Canada. Utilizing ViTs did not demonstrate a significant improvement in UNet performance, though this could change with longer training. This paper proposes opportunities to improve performance for these neural networks, as well as next steps for further research, including improving the diversity of images in the dataset, generating a testing dataset from a completely different geographic location, and allowing the models more time to train.
more »
« less
- Award ID(s):
- 2116216
- PAR ID:
- 10451173
- Date Published:
- Journal Name:
- 2023 IEEE International Conference on Mobility, Operations, Services and Technologies (MOST)
- Page Range / eLocation ID:
- 124 to 131
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Risk assessment of infrastructure exposed to ice-rich permafrost hazards is essential for climate change adaptation in the Arctic. As this process requires up-to-date, comprehensive, high-resolution maps of human-built infrastructure, gaps in such geospatial information and knowledge of the applications required to produce it must be addressed. Therefore, this study highlights the ongoing development of a deep learning approach to efficiently map the Arctic built environment by detecting nine different types of structures (detached houses, row houses, multi-story blocks, non-residential buildings, roads, runways, gravel pads, pipelines, and storage tanks) from recently-acquired Maxar commercial satellite imagery (<1 m resolution). We conducted a multi-objective comparison, focusing on generalization performance and computational cost, of nine different semantic segmentation architectures. K-fold cross validation was used to estimate the average F1-score of each architecture and the Friedman Aligned Ranks test with the Bergmann-Hommel posthoc procedure was applied to test for significant differences in generalization performance. ResNet-50-UNet++ performs significantly better than five out of the other eight candidate architectures; no significant difference was found in the pairwise comparisons of ResNet-50-UNet++ to ResNet-50-MANet, ResNet-101-MANet, and ResNet-101-UNet++. We then conducted a high-performance computing scaling experiment to compare the number of service units and runtime required for model inferencing on a hypothetical pan- Arctic scale dataset. We found that the ResNet-50-UNet++ model could save up to ~ 54% on service unit expenditure, or ~ 18% on runtime, when considering operational deployment of our mapping approach. Our results suggest that ResNet-50-UNet++ could be the most suitable architecture (out of the nine that were examined) for deep learning-enabled Arctic infrastructure mapping efforts. Overall, our findings regarding the differences between the examined CNN architectures and our methodological framework for multi-objective architecture comparison can provide a foundation that may propel future pan-Arctic GeoAI mapping efforts of infrastructure.more » « less
-
Taken together, lakes and drained lake basins may cover up to 80% of the lowland landscapes in permafrost regions of the Arctic. Lake formation, growth, and drainage in lowland permafrost regions create a terrestrial and aquatic landscape mosaic of importance to geomorphic and hydrologic processes, tundra vegetation communities, permafrost and ground-ice characteristics, biogeochemical cycling, wildlife habitat, and human land-use activities. Our project focuses on quantifying the role of thermokarst lake expansion, drainage, and drained lake basin evolution in the Arctic System. We did this through a combination of field studies, environmental sensor networks, remote sensing, and modeling. This dataset consists of an orthomosaic and digital surface model (DSM) derived from drone surveys on 23 July 2022 at Derksen and Schmutz Basins on the Arctic Coastal Plain of northern Alaska. 6,158 digital images were acquired from a DJI Phantom 4 Real-Time Kinematic (DJI P4RTK) quadcopter with a DJI D-RTK 2 Mobile Base Station. The mapped area was around 580 hectares (ha). The drone system was flown at 100 meters (m) above ground level (agl) and flight speeds varied from 7–8 meters/second (m/s). The orientation of the camera was set to 90 degrees (i.e. looking straight down). The along-track overlap and across-track overlap of the mission were set at 80% and 70%, respectively. All images were processed in the software Pix4D Mapper (v. 4.8.4) using the standard 3D Maps workflow and the accurate geolocation and orientation calibration method to produce the orthophoto mosaic and digital surface model at spatial resolutions of 5 and 10 centimeters (cm), respectively. Elevation information derived over waterbodies is noisy and does not represent the surface elevation of the feature. A Leica Viva differential global positioning system (GPS) provided ground control for the mission and the data were post-processed to WGS84 UTM Zone 5 North in Ellipsoid Heights (meters).more » « less
-
Taken together, lakes and drained lake basins may cover up to 80% of the lowland landscapes in permafrost regions of the Arctic. Lake formation, growth, and drainage in lowland permafrost regions create a terrestrial and aquatic landscape mosaic of importance to geomorphic and hydrologic processes, tundra vegetation communities, permafrost and ground-ice characteristics, biogeochemical cycling, wildlife habitat, and human land-use activities. Our project focuses on quantifying the role of thermokarst lake expansion, drainage, and drained lake basin evolution in the Arctic System. We did this through a combination of field studies, environmental sensor networks, remote sensing, and modeling. This dataset consists of an orthomosaic and digital surface model (DSM) derived from drone surveys on 20 July 2022 at Novo Basin on the Arctic Coastal Plain of northern Alaska. 332 digital images were acquired from a DJI Phantom 4 Real-Time Kinematic (DJI P4RTK) quadcopter with a DJI D-RTK 2 Mobile Base Station. The mapped area was around 43 hectares (ha). The drone system was flown at 100 meters (m) above ground level (agl) and flight speeds varied from 7–8 meters/second (m/s). The orientation of the camera was set to 90 degrees (i.e. looking straight down). The along-track overlap and across-track overlap of the mission were set at 80% and 70%, respectively. All images were processed in the software Pix4D Mapper (v. 4.8.4) using the standard 3D Maps workflow and the accurate geolocation and orientation calibration method to produce the orthophoto mosaic and digital surface model at spatial resolutions of 5 and 10 centimeters (cm), respectively. Elevation information derived over waterbodies is noisy and does not represent the surface elevation of the feature. A Leica Viva differential global positioning system (GPS) provided ground control for the mission and the data were post-processed to WGS84 UTM Zone 5 North in Ellipsoid Heights (meters).more » « less
-
Emergence of beavers as ecosystem engineers in the New Arctic project focuses on establishing field sites at tundra beaver ponds to study the implications of beaver engineering on hydrology and permafrost. Drones are being used to collect baseline data and track beaver dam building and pond evolution over time. This dataset consists of an orthomosaic and digital surface model (DSM) derived from drone surveys on 04 August 2021 at the Upper Nome River, MP31, site on the Seward Peninsula, Alaska. 248 digital images were acquired from a DJI Phantom 4 Real-Time Kinematic (DJI P4RTK) quadcopter with a DJI D-RTK 2 Mobile Base Station. The mapped area was around 44 hectares (ha). The drone system was flown at 120 meters (m) above ground level (agl) and flight speeds varied from 8-9 meters/second (m/s). The orientation of the camera was set to 90 degrees (i.e. looking straight down). The along-track overlap and across-track overlap of the mission were set at 80% and 70%, respectively. All images were processed in the software Pix4D Mapper (v. 4.7.5) using the standard 3D Maps workflow and the accurate geolocation and orientation calibration method to produce the orthophoto mosaic and digital surface model at spatial resolutions of 5 and 10 centimeters (cm), respectively. Elevation information derived over waterbodies is noisy and does not represent the surface elevation of the feature. A Leica Viva differential global positioning system (GPS) provided ground control for the mission and the data were post-processed to WGS84 UTM Zone 3 North in Ellipsoid Heights (meters).more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/MOST57249.2023.00021
