Cloud Pixel Tracking for Multi-Temporal Images of Himawari-8 Satellite

Clouds present the feature of weather changes, and they are also very important for weather prediction.Cloud phase, cloud type, and the changes of cloud top properties (temperature, height and pressure) allprovide information for cloud development and rain fall prediction. To analyze the changes of cloudproperties, we must track the movement, appearance and disappearance of clouds in pixel level.The geosynchronous weather satellites orbit with the rotation of the Earth. They can providesemi-sphere image with high temporal resolution. The MTSAT-2 provides an image every 30 minutes, and itmight delay to an hour if transmission interference occurs. Some small convection clouds may develop anddisappear within 30 minutes, so MTSAT-2 cannot track them. If the clouds move fast, it is also difficult totrack them. In 2014, Japan Meteorological Agency launched Himawari-8 to Earth’s orbit. This satellitebegan to send signal in July, 2015. It not only increases the number of channels to 16, but also increases thetemporal resolution to 10 minutes. It also has the capability to monitor a small area every 2.5 minutes. In thisproposal, we will adopt Particle Swarm Optimization (PSO), Self-Organization Map Neural Network(SOM-NN) to track the clouds. They view cloud as one group and cloud pixels as particles or neurons totrack the movement of a cloud and the relative relationship between cloud pixels. And further analyze thechange of cloud top properties as part of weather prediction coefficients.