Satellite cloud picture extrapolation method based on pyramid optical flow method and discrete Kalman filtering

• Main Authors: DAI QIANGSHENG, HAO YUCHEN, SUN YONG, MEI XUEFENG, PAN SHULEI, CHAI YUN, HUO XUESONG, BAI JINGJING
• Format: Patent
• Language: Chinese; English
• Published: 27.02.2024
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a satellite cloud picture extrapolation method based on a pyramid optical flow method and discrete Kalman filtering. The satellite cloud picture extrapolation method comprises the following steps: firstly, acquiring satellite cloud picture data at a historical moment, and preprocessing a satellite cloud picture; secondly, satellite cloud pictures at continuous moments are selected from preprocessed satellite cloud picture results to serve as historical data, and pyramid image layering is constructed; then taking the optical flow motion field of the upper layer as the initial motion field of the image of the lower layer, iteratively calculating a new optical flow motion field, taking the calculated optical flow field as the initial state of a discrete Kalman filter, and forecasting the satellite cloud atlas at continuous moments; and finally, inputting historical cloud picture data into the trained discrete Kalman filter to obtain an extrapolated cloud picture result. According to the m