Forward-Adaptive Method for Context-Based Compression of Large Binary Images

• Main Authors: Eugene Ageenko, Pasi Fränti, For Correspondence
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Published: 1999
• Subjects: image compression; document imaging; context modeling; JBIG; QMcoder; The Pointers
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.22.7753; http://cs.joensuu.fi/pages/ageenko/public/journal/ea_spe99.pdf

INTRODUCTION Spatial access stands for direct access to image fragments in the compressed file. It is a highly desired property of document imaging applications dealing with spatial data [1,2]. Applications such as Engineering Document Management (EDM) [3] and Geographic Information System (GIS) [4] use large format images and therefore benefit from spatial access also. Typical viewing devices have a smaller size and resolution than the raster image and thus, only a small fragment of the entire image may be viewed at a time. Spatial access enables the user to retrieve and decompress only the desired image fragment and eliminate unnecessary delays caused by the retrieval and decompression of the entire image. Spatial access can be supported by tiling, i.e. partitioning the image into fixed size rectangular blocks, which are denoted here as clusters. A cluster index table is constructed from the pointers indicating