In this book, we proposed a developed system to restore foggy degraded images using Back propagation neural network(BP-NN). The suggested method is based on mapping between foggy scene and its corresponding original scene. The weight matrix resulted from training the BP-NN represents the filter of the fog.By using more than one method and more than one feature set, we achieved the effective features that suits BP-NNmethod for foggy image restoration. Seven different approaches (seven different features) are suggested to generate proper fog filter (i.e. seven different fog filters are generated).