Seasonal Adaptation of Vegetation Color in Satellite Images for Flight Simulations

Automatic vegetation identification plays an important role in many applications including remote sensing and high performance flight simulations. This paper proposes a novel method that identifies vegetative areas in satellite images and then alters vegetation color to simulate seasonal changes based on training image pairs. The proposed method first generates a vegetation map for pixels corresponding to vegetative areas, using ISODATA clustering and vegetation classification. The ISODATA algorithm determines the number of clusters automatically. We then apply morphological operations to the clustered images to smooth the boundaries between clusters and to fill holes inside clusters. Six features are then computed for each cluster and then go through a feature selection algorithm and three of them are determined to be effective for vegetation identification. Finally, we classify the resulting clusters as vegetation and non vegetation types based on the selected features using a multilayer perceptron (MLP) classifier. We tested our algorithm by using the 5-fold cross-validation method and achieved 96% classification accuracy based on the three selected features. After the vegetation areas in the satellite images are identified, the proposed method then generates seasonal color adaptation of a target input image based on a pair of training images and, which depict the same area but were captured in different seasons, using image analogies technique. The final output image has seasonal appearance that is similar to that of the training image. The vegetation map ensures that only the colors of vegetative areas in the target image are altered and it also improves the performance of the original image analogies technique. The proposed method can be used in high performance flight simulations and other applications.