NON-RIGID REGISTRATION OF CLOUD MOVEMENT
暂无分享,去创建一个
Our goal is to optimize the interplay between conventional and solar power plants. We require a short-term and highly accurate prediction of the energy output. On a coarse scale, the weather influencing the power output can be predicted from satellite imagery. To achieve a higher temporal and spatial resolution a ground-based camera is required. In a first step, we focus on determining and predicting the movement of clouds using nonrigid registration techniques. Data aquisition Images were aquired with a Canon 550D and a Walimex 8mm fisheye lens in intervals of 20 seconds. A total of 180 images (i.e. one hour) were used for evaluation. Example Image
[1] Jan Modersitzki,et al. Curvature Based Image Registration , 2004, Journal of Mathematical Imaging and Vision.
[2] Jean-Philippe Thirion,et al. Image matching as a diffusion process: an analogy with Maxwell's demons , 1998, Medical Image Anal..
[3] J. Kleissl,et al. Intra-hour forecasting with a total sky imager at the UC San Diego solar energy testbed , 2011 .