Accurate rainfall forecasting is crucial for various applications such as agriculture, hydrology, and disaster management. Recent advancements in satellite technology have enabled the collection of high spatiotemporal resolution data, which can be used to improve rainfall forecasting models. In this study, we utilize the cloud brightness temperature (CBT) derived from INSAT 3DR satellite images to predict rainfall over the Indian region. The CBT values are calculated using a time series approach, and the random forest algorithm is employed to develop a forecasting model. The performance of the model is evaluated using a set of evaluation metrics, and the results show that the proposed methodology can accurately predict rainfall with a high degree of accuracy. The advantages of using high spatiotemporal satellite imagery and machine learning techniques for rainfall forecasting are discussed, and future research directions in this area are also explored.