Tiled Image Container for Web Compatible Compound Image Compression

Encoding images containing mixed content in web compatible formats like PNG, GIF and JPEG poses challenges in maintaining the quality along with good compression. In this paper, we describe a method for improving the compression of web compatible compound images while retaining high quality. The proposed method is based on the principle that the compression ratio of a given image depends upon both the image content and the compression algorithm used. Therefore, encoding various tiled regions in a given image with the algorithm appropriate to the content in the region improves the overall compression of the image. Further optimizations like sharing of indexed colors between the various tiles in the same image and removal of duplicate tiles improve the compression. This method would support both lossy and loss less compression. The resulting encoded data can be rendered by a web browser supporting CSS background-cropping or HTML5 canvas APIs.