WAVELET BASED ALGORITHM FOR THE IDENTIFICATION OF RAINMASS FROM RADAR IMAGES by Rajdeep Kaur
Abstract
Water is a very important part of human life. Economic scenario of a country highly depends on the rainfall. Rainfall estimation has become very necessary due to day by day increasing global heat level. An enormous amount of rainfall is a serious threat to both life and property because it causes a flood. Therefore, it is important to find where rainfall has occurred, the amount of rainfall and to forecast rainfall. Warnings of the flood can be generated before rainfall occurrence if heavy rainfall is forecasted accurately. Precise prediction of rainfall can be achieved by using accurate rainfall estimation technique. Basically, there are three estimation techniques namely: rain gauge, weather radar and numerical weather prediction model. Among all the three, weather radar provides more accurate information about the rainfall and therefore is widely used in rainfall prediction models. Weather radar measures reflectivity. The reflectivity of weather radar is directly related to rainfall rate. Moreover, weather radar performs short term rainfall prediction, based on the current weather situation. Therefore, a new algorithm using Met Eireann rainfall radar images has been proposed for the identification of rainmass from radar images. Physical and morphological feature of each identified rainmass is extracted. The next movement of rainmass is predicted based on the features of preceding rainmasses identified from the radar images.
