A robust system that could detect early onset of forest fires from the ground level needs to be built. In this paper, a Voronoi Segmentation based system is introduced to preprocess the aerial images of forest regions for efficient image processing that will be utilized for the placement mechanism of the said Forest Fire Early Detection and Alert System(FFEDAS). By using the segmentation, a significant change in the processing times of the images has been observed, demonstrating the efficacy of this experiment.