Category: Data Visualization

[Note]Chapter6 Geographic information visualization

Map projection

  • equal angle
    • Mercator Projection
  • equal area
    • Albers Projection
  • equal distance
    • Azimuthal Projection

Point data visualization

Object

Discrete points of geographic space including longitude and latitude coordinate, no size property.

Vis approaches

  • Mark on the map directly
  • Indicate numeric properties by color and size
  • Apply different icons and symbols to indicate different data types
    • symbols should be easily perceived
    • the number of symbols shouldn’t be too large
    • use legend to explain the meaning of symbol
  • visualize vectorial data by icons like arrow

Problems

  • many overlaps when quantities of data.
    • Specially, overlaps occur in intensive place, whereas, blank occurs in sparse place.
  • unreasonable of use of screen space
    • a great deal of information is concentrated in small area, however, other areas are completely empty

Solutions

  • split map to various blocks. Visualize the statistic data of each block.
    • showing data in discrete way, one method is to split map area into grids. Visualization can be 3D histogram with drawback of covering each other and can use color without drawback of covering.
    • showing data in continuous way by heatmap. Compared with discrete block, the granularity is finer.
  • Apply appropriate layout algorithm and utilize the rendering and integrating method to show each data object. This can express more details compared with the above method.

line data visualization

Object

Line data with the property of length, is the data to connect two or more points.

Vis approaches

  • basic visualization methods such as color, line type, width and mark.
  • To avoid visual confusion, cluster a large quantity of data to bunches.
    • this kind of line binding technology is called flow map. There are some algorithm to improve vis performance.

Regional data

Object

Regional data contains more information than point data and line data. It can express regional properties like population intensity and per-capita data.

Approaches

  • Choropleth map
    • Color matching suggestion tool: ColorBrewer
    • Problem: data distribution and geographic region are asymmetric. Normally large data set is concentrated in limited population intensive region.
  • Cartogram map
    • Method: reshape each region according to the property of geographic region.
    • Problem: hard to guarantee the relative position of different region.
    • Solution: continuous Cartogram. This algorithm can keep the relative position of neighboring regions.
  • Map of regular shape
    • Rectangle or circle
  • Multi relational map
    • different geographic regions can have associated relation.
      • Approach: Bubble Sets. Problem: visual confusion when in quantity. Solution: LinSets which connecting a set of points by one curve.

spatio-temporal graphic data

  • frequently used approach: sequential animation.

Complex geographic data vis analysis

Map can carry various types of complex information, which can be formed as multi semantic map.

  • Approach: multiple views to help analyze multiple variables.

Challenges

  • map labeling
  • map generalization which has the aim to filter and draw map information according to the showing size.
  • online map – limited bandwidth vs real-time interaction

 

Reference

陈为 沈则潜 陶煜波. (2013). 数据可视化. 电子工业出版社. (第六章)

Advertisements