Classification of Forest Fire using InceptionV3 model

The project "Classification of Forest Fires using InceptionV3 model" aims to develop deep learning models to detect forest fires using satellite images. The recommended approach makes use of the InceptionV3 model to analyse the input of satellite images and find forest fires. We have used U-Net architecture which has some limitations in detecting forest fires early and effectively. Using InceptionV3 instead of U-Net will improve the Accuracy and to make comparisons among these models. The limitation of U-Net for forest fire detection lies in its focus on detailed segmentation rather than capturing broader contextual information, which is crucial for understanding complex patterns like forest fires. This constraint can be addressed by InceptionV3, which can capture more comprehensive patterns and enhance the comprehension of forest fires in satellite pictures due to its capacity to comprehend substantial context and features at many scales. By using this model, we aim to make our project more accurate and useful for predicting and managing forest fires. We'll compare the performance of this model with the U-Net to see which one is better for finding fires in different situations. This upgrade is designed to enhance our system's accuracy and effectiveness in forest fire detection and protection of the environment.