This page has only limited features, please log in for full access.
This paper introduces a new GeoAI solution to support automated mapping of global craters on the Mars surface. Traditional crater detection algorithms suffer from the limitation of working only in a semiautomated or multi-stage manner, and most were developed to handle a specific dataset in a small subarea of Mars’ surface, hindering their transferability for global crater detection. As an alternative, we propose a GeoAI solution based on deep learning to tackle this problem effectively. Three innovative features are integrated into our object detection pipeline: (1) a feature pyramid network is leveraged to generate feature maps with rich semantics across multiple object scales; (2) prior geospatial knowledge based on the Hough transform is integrated to enable more accurate localization of potential craters; and (3) a scale-aware classifier is adopted to increase the prediction accuracy of both large and small crater instances. The results show that the proposed strategies bring a significant increase in crater detection performance than the popular Faster R-CNN model. The integration of geospatial domain knowledge into the data-driven analytics moves GeoAI research up to the next level to enable knowledge-driven GeoAI. This research can be applied to a wide variety of object detection and image analysis tasks.
Chia-Yu Hsu; Wenwen Li; Sizhe Wang. Knowledge-Driven GeoAI: Integrating Spatial Knowledge into Multi-Scale Deep Learning for Mars Crater Detection. Remote Sensing 2021, 13, 2116 .
AMA StyleChia-Yu Hsu, Wenwen Li, Sizhe Wang. Knowledge-Driven GeoAI: Integrating Spatial Knowledge into Multi-Scale Deep Learning for Mars Crater Detection. Remote Sensing. 2021; 13 (11):2116.
Chicago/Turabian StyleChia-Yu Hsu; Wenwen Li; Sizhe Wang. 2021. "Knowledge-Driven GeoAI: Integrating Spatial Knowledge into Multi-Scale Deep Learning for Mars Crater Detection." Remote Sensing 13, no. 11: 2116.
Understanding atmospheric phenomena involves analysis of large-scale spatiotemporal multivariate data. The complexity and heterogeneity of such data pose a significant challenge in discovering and understanding the association between multiple climate variables. To tackle this challenge, we present an interactive heuristic visualization system that supports climate scientists and the public in their exploration and analysis of atmospheric phenomena of interest. Three techniques are introduced: (1) web-based spatiotemporal climate data visualization; (2) multiview and multivariate scientific data analysis; and (3) data mining-enabled visual analytics. The Arctic System Reanalysis (ASR) data are used to demonstrate and validate the effectiveness and usefulness of our method through a case study of “The Great Arctic Cyclone of 2012”. The results show that different variables have strong associations near the polar cyclone area. This work also provides techniques for identifying multivariate correlation and for better understanding the driving factors of climate phenomena.
Feng Wang; Wenwen Li; Sizhe Wang; Chris R. Johnson. Association Rules-Based Multivariate Analysis and Visualization of Spatiotemporal Climate Data. ISPRS International Journal of Geo-Information 2018, 7, 266 .
AMA StyleFeng Wang, Wenwen Li, Sizhe Wang, Chris R. Johnson. Association Rules-Based Multivariate Analysis and Visualization of Spatiotemporal Climate Data. ISPRS International Journal of Geo-Information. 2018; 7 (7):266.
Chicago/Turabian StyleFeng Wang; Wenwen Li; Sizhe Wang; Chris R. Johnson. 2018. "Association Rules-Based Multivariate Analysis and Visualization of Spatiotemporal Climate Data." ISPRS International Journal of Geo-Information 7, no. 7: 266.
The world is undergoing rapid changes in its climate, environment, and ecosystems due to increasing population growth, urbanization, and industrialization. Numerical simulation is becoming an important vehicle to enhance the understanding of these changes and their impacts, with regional and global simulation models producing vast amounts of data. Comprehending these multidimensional data and fostering collaborative scientific discovery requires the development of new visualization techniques. In this paper, we present a cyberinfrastructure solution—PolarGlobe—that enables comprehensive analysis and collaboration. PolarGlobe is implemented upon an emerging web graphics library, WebGL, and an open source virtual globe system Cesium, which has the ability to map spatial data onto a virtual Earth. We have also integrated volume rendering techniques, value and spatial filters, and vertical profile visualization to improve rendered images and support a comprehensive exploration of multi-dimensional spatial data. In this study, the climate simulation dataset produced by the extended polar version of the well-known Weather Research and Forecasting Model (WRF) is used to test the proposed techniques. PolarGlobe is also easily extendable to enable data visualization for other Earth Science domains, such as oceanography, weather, or geology.
Sizhe Wang; Wenwen Li; Feng Wang. Web-Scale Multidimensional Visualization of Big Spatial Data to Support Earth Sciences—A Case Study with Visualizing Climate Simulation Data. Informatics 2017, 4, 17 .
AMA StyleSizhe Wang, Wenwen Li, Feng Wang. Web-Scale Multidimensional Visualization of Big Spatial Data to Support Earth Sciences—A Case Study with Visualizing Climate Simulation Data. Informatics. 2017; 4 (3):17.
Chicago/Turabian StyleSizhe Wang; Wenwen Li; Feng Wang. 2017. "Web-Scale Multidimensional Visualization of Big Spatial Data to Support Earth Sciences—A Case Study with Visualizing Climate Simulation Data." Informatics 4, no. 3: 17.
Arctic cyclone activity has a significant association with Arctic warming and Arctic ice decline. Cyclones in the North Pole are more complex and less developed than those in tropical regions. Identifying polar cyclones proves to be a task of greater complexity. To tackle this challenge, a new method which utilizes pressure level data and velocity field is proposed to improve the identification accuracy. In addition, the dynamic, simulative cyclone visualized with a 4D (four-dimensional) wind field further validated the identification result. A knowledge-driven system is eventually constructed for visualizing and analyzing an atmospheric phenomenon (cyclone) in the North Pole. The cyclone is simulated with WebGL on in a web environment using particle tracing. To achieve interactive frame rates, the graphics processing unit (GPU) is used to accelerate the process of particle advection. It is concluded with the experimental results that: (1) the cyclone identification accuracy of the proposed method is 95.6% when compared with the NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research) reanalysis data; (2) the integrated knowledge-driven visualization system allows for streaming and rendering of millions of particles with an interactive frame rate to support knowledge discovery in the complex climate system of the Arctic region.
Feng Wang; Wenwen Li; Sizhe Wang. Polar Cyclone Identification from 4D Climate Data in a Knowledge-Driven Visualization System. Climate 2016, 4, 43 .
AMA StyleFeng Wang, Wenwen Li, Sizhe Wang. Polar Cyclone Identification from 4D Climate Data in a Knowledge-Driven Visualization System. Climate. 2016; 4 (3):43.
Chicago/Turabian StyleFeng Wang; Wenwen Li; Sizhe Wang. 2016. "Polar Cyclone Identification from 4D Climate Data in a Knowledge-Driven Visualization System." Climate 4, no. 3: 43.