Cloud Classification for Weather Information by Artificial Neural Network

Sanju Kuril, Indu Saini, B. S. Saini
2013 International Journal of Applied Physics and Mathematics  
Continuous cloud classification through satellite image is an approach which is very important in many atmospheric and Environment studies such as weather analysis and forecasting. The detection and classification of clouds of satellite image the most effective method for identifying the individual cloud type for weather analysis and forecasting. It involves two mainly Stages Feature Extraction and Classification, Feature extraction has been done by Haar DWT i.e. Haar Discrete Wavelet Transform
more » ... and the classification has been done by ANN i.e. Artificial neural network. Classification and accuracy has been improved by using artificial neural network on satellite images. The performance of the designed classifier is evaluated and which gives the overall accuracy of 94.37 % with 91.66% for Low Level Clouds, 98.33% for Middle level Clouds and 93.33% for High Level Clouds.
doi:10.7763/ijapm.2013.v3.167 fatcat:h2nacpmrljhmniecyyq6wb7724