For those who live along the coast, tropical cyclones are the greatest nightmare ever. Numerous issues with people’s lives and their properties are brought on by this chaos. Early alerts regarding the intensity of the cyclone and its projected path can prevent fatalities and protect property. A crucial role is played by meteorological and weather departments in this process; both departments rely on real-time satellite feeds. With the use of deep learning mechanisms, this prediction process can be completed faster. On the other hand, numerous systems exist to forecast the tracking direction of cyclones using statistical datasets, while numerous systems exist to estimate the intensity of cyclones using images. The people involved won’t benefit much more from it while both are examining the many criteria. In order to estimate the cyclone’s intensity and direction, a hybrid model has been implemented in the proposed approach. This ultimately aids the people living in the coastal area in making informed decisions that could perhaps save their lives. The suggested model deploys an LSTM neural network, which effectively predicts the tracking path coordinates using static data, in conjunction with a convolution neural network to assess intensity using satellite imagery. The results demonstrate that every attempt is made to implement the hybrid model in order to minimize the hybrid model’s major root mean square error probability.