A new algorithm for DEM data compression base on feature points

To take full advantage of topography characteristic and improve compression ratio when compressing digital elevation model(DEM), a new compression method based on BP neural network is proposed. An extraction algorithm of feature points including terrain ridge line and valleys is given firstly. Then the BP neural network is trained to implement DEM compression by using the extracted feature points. The experimental results demonstrate the effectiveness of presented method, which can enhance DEM compression effect.