A cloud consists of a visible mass of minute liquid droplets, frozen crystals and other particles suspended in the atmosphere. Clouds may be classified by form and by weight. The need of this research is in predicting the weather conditions, and sky conditions or cloud cover changes on earth’s surface caused by snow, storms, tides and floods and in measuring rain pressure, wind and humidity and in a better understanding of Earth's weather. Weather forecasting now relies on computer-based models. This paper proposes a novel AdaBoost framework approach for detection of cloud types. One of the divisions of clouds is based upon the range of altitude. The dataset used in this research was created for different classes of clouds which included stratus, stratocumulus, cumulus and cumulonimbus types. Our proposed approach is tested on this dataset and is also compared with other state-of-the art algorithms (Support Vector Machine (SVM), K-Nearest Neighbour (KNN), Decision Tree (DT) and Random Forest (RF)) that have shown a marvellous performance in the area of feature extraction. The results have shown a comparable performance, in terms of differentiating amongst the wide range of cloud classification and are able to recognise them with the best accuracy of 75.34%.
Volume 11 | 06-Special Issue
Pages: 372-376