Visualization of geo-spatial point sets via global shape transformation and local pixel placement

Area of Science:

• Information Visualization
• Geographic Information Systems (GIS)
• Data Science

Background:

• Geo-spatial data visualization is crucial for pattern discovery.
• Existing methods like PixelMaps reveal fine structures but lack geographic context.
• Global shape functions offer recognizable maps but struggle with point sets.

Purpose of the Study:

• To present a novel framework for enhanced geo-spatial data exploration.
• To combine global shape functions with local placement functions.
• To improve the visualization of dense geo-spatial datasets.

Main Methods:

• Developed a framework integrating user-specified global shape and local placement functions.
• Combined cartogram-based layout (global shape) with PixelMaps (local placement).
• Applied the combined approach to dense geo-spatial datasets.

Main Results:

• The integrated framework offers benefits from both global and local spatial transformations.
• Improved exploration of dense geo-spatial data through enhanced pattern visibility.
• Successfully related visualized structures to basic geographic features.

Conclusions:

• The proposed framework effectively combines global and local spatial transformations for geo-spatial data.
• This hybrid approach enhances the interpretability and discoverability of patterns in dense datasets.
• Offers a valuable tool for researchers and analysts working with geo-spatial information.