This paper presents an automatic image segmentation based on the fast active contour model and storage system, which automatically segments the tumor from mammograms and based on the results, performs a progressive compression in the storage. This is performed in two subsystems called computerized tumor boundary segmentation subsystem and tumor region reserved compression subsystem. In the first subsystem, a threshold selection method is used first to remove the background from the image, then the morphological operator is used to remove the noise, the initial contour estimation is the boundary of the extracted tumor. A modified algorithm based on the greedy algorithm for active contour modeling is also presented to approach tumor boundary. Finally, the test mammograms are segmented into tumor, breast without tumor and background. In the second subsystem, Vector Quantization GHNN (Gray-based Competitive Hopfield Neural Network) is applied on the three regions with different compression rates according their importance factors so as to reserve important tumor features and simultaneously reduce the size of mammograms for storage efficiency.