This study addresses how to generate high spatial resolution image data from a low spatial resolution imaging source. An edge-restricted spatial interpolation algorithm is developed to increase the image resolution and at the same time to enhance the sharp edges and details from the original imaging source. The algorithm is based on a cubic spline-under-tension interpolation kernel. The weights of the interpolation kernel can be adjusted adaptively according to the edge information in the neighborhood of the interpolated pixels. The algorithm can be applied to a relatively low spatial resolution image source, such as video, to generate high-resolution image data for high-quality printing devices.