摘要: There are three basic segmentation schemes for compound image compression: object-based, layer-based, and block-based. This paper discusses the relative advantages of each scheme architecture, studies use fast classification techniques a that can be used together with chosen compression architecture. Particularly, we consider working on approximate object boundaries, which reaches localization precision segmentation, but in exchange allows faster, one-pass low memory requirements, map is better matched to existing methods. We show numerical results obtained printer application environment, where rigorous standards visual quality have satisfied.
sci-hub.se 参考文章(6)
Hei Tao Fung, Kevin J. Parker, Color, complex document segmentation and compression electronic imaging. ,vol. 3027, pp. 180- 191 ,(1997) , 10.1117/12.270071
Rene Sennhauser, Krystyna W. Ohnesorge, Document image compression using document analysis and block-class-specific data compression methods Image and Video Compression. ,vol. 2186, pp. 146- 155 ,(1994) , 10.1117/12.173914
Theo Pavlidis, Jiangying Zhou, Page segmentation and classification CVGIP: Graphical Models and Image Processing. ,vol. 54, pp. 484- 496 ,(1992) , 10.1016/1049-9652(92)90068-9
L eon Bottou, Patrick Haffner, Paul G Howard, Patrice Simard, Yoshua Bengio, Yann LeCun, NJ Lincroft, High quality document image compression with “DjVu” Journal of Electronic Imaging. ,vol. 7, pp. 410- 425 ,(1998) , 10.1117/1.482609
Ricardo L. de Queiroz, Fast segmentation of the JPEG-compressed documents Journal of Electronic Imaging. ,vol. 7, pp. 367- 377 ,(1998) , 10.1117/1.482607
A. Benedetti, N. Scarabottolo, Towards a dedicated compression pipeline for document image archiving Proceedings Workshop on Document Image Analysis (DIA'97). pp. 40- 43 ,(1997) , 10.1109/DIA.1997.627090