Skip to main content
placeholder image

Cloud motion tracking for short-term on-site cloud coverage prediction

Conference Paper


Abstract


  • A technique for cloud motion tracking and cloud motion prediction using ground-based sky images is presented. This cloud motion prediction technique primarily targets irradiance prediction as an application in Electrical Engineering. A sequence of whole sky images is processed to determine the time taken by clouds to reach the sun position on the image. Cross-correlation technique was used to track individual clouds from one image frame to next frame. Using Harris features detection algorithm cloud features were found and the deformation vectors were produced. To find the velocity vectors of each feature points Lukas-Kanade optical flow algorithm is proposed. Using the optical flow algorithm, 3 min ahead cloud position was estimated.

UOW Authors


  •   Dissawa, D (external author)
  •   Ekanayake, M P.B. (external author)
  •   Godaliyadda, G (external author)
  •   Ekanayake, Janaka (external author)
  •   Agalgaonkar, Ashish

Publication Date


  • 2017

Citation


  • D. M.L.H. Dissawa, M. P.B. Ekanayake, G. M.R.I. Godaliyadda, J. B. Ekanayake & A. P. Agalgaonkar, "Cloud motion tracking for short-term on-site cloud coverage prediction," in 17th International Conference on Advances in ICT for Emerging Regions, ICTer 2017 - Proceedings, 2017, pp. 332-337.

Scopus Eid


  • 2-s2.0-85049473105

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers1/1578

Start Page


  • 332

End Page


  • 337

Place Of Publication


  • United States

Abstract


  • A technique for cloud motion tracking and cloud motion prediction using ground-based sky images is presented. This cloud motion prediction technique primarily targets irradiance prediction as an application in Electrical Engineering. A sequence of whole sky images is processed to determine the time taken by clouds to reach the sun position on the image. Cross-correlation technique was used to track individual clouds from one image frame to next frame. Using Harris features detection algorithm cloud features were found and the deformation vectors were produced. To find the velocity vectors of each feature points Lukas-Kanade optical flow algorithm is proposed. Using the optical flow algorithm, 3 min ahead cloud position was estimated.

UOW Authors


  •   Dissawa, D (external author)
  •   Ekanayake, M P.B. (external author)
  •   Godaliyadda, G (external author)
  •   Ekanayake, Janaka (external author)
  •   Agalgaonkar, Ashish

Publication Date


  • 2017

Citation


  • D. M.L.H. Dissawa, M. P.B. Ekanayake, G. M.R.I. Godaliyadda, J. B. Ekanayake & A. P. Agalgaonkar, "Cloud motion tracking for short-term on-site cloud coverage prediction," in 17th International Conference on Advances in ICT for Emerging Regions, ICTer 2017 - Proceedings, 2017, pp. 332-337.

Scopus Eid


  • 2-s2.0-85049473105

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers1/1578

Start Page


  • 332

End Page


  • 337

Place Of Publication


  • United States