Singapore Whole sky IMaging SEGmentation Database (SWIMSEG) [1]

  • 1013 sky images 
  • 600 x 600 pixels
  • Captured by ground-based sky imager WAHRSIS (Wide Angle High Resolution Sky Imaging System)
  • Corresponding ground truth masks were created in consultation with cloud experts from the Singapore Meteorological Services

Figure 1: Sample Images from SWIMSEG

Sky Image Ground Truth
0003 sky image 0003 Ground Truth
0029 sky image 0029 Ground Truth
   0140 Ground Truth
 0189 sky image  0189 Ground Truth

Singapore Whole sky IMaging CATegories Database (SWIMCAT) [2]

  • 784 images broken into 5 categories: clear sky, veil clouds, patterned clouds, thick dark (night), and thick white (day)
  • 125 x 125 pixels
  • Images categorized by visual features in consultation with Singapore Meteorological Services
  • For this project, the night time images (thick dark) were discarded in favor of the remaing 533 involving sunlight

Figure 2: Sample Images from SWIMCAT

Sky Image SWIMCAT Classification
Clear Sky
Patterned Cloud Patterned Cloud
Thick Cloud  Thick Cloud
Veil Cloud  Veil Cloud


  1. Dev, Soumyabrata, et al. “Color-Based Segmentation of Sky/Cloud Images From Ground-Based Cameras.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 10, no. 1, 12 June 2016, pp. 231–242.
  2. S. Dev, Y. H. Lee, S. Winkler. Categorization of cloud image patches using an improved texton-based approach. Proc. IEEE International Conference on Image Processing (ICIP), Québec City, Canada, Sep. 27-30, 2015.