O. E. Kia, "Document image compression and analysis," Ph.D. dissertation, Depart of Electrical Engineering, University of Maryland, 1997. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Lossless Document Image Compression - Inglis (1999) (4 citations) (Correct)
....Lossy SPM can be combined with the residue encoding stage to replicate MGTIC s two level progressive encoding. SPM achieved lossless compression ratios between 10 and 20 percent better than MGTIC. Kia described a component based compression system that processed images in a compressed domain [Kia97] No implementation of Kia s system is available, and few results are given. Zhang extended the code of MGTIC by modifying the template matching, reading order, and character positions encoding [Zha97] Like Kia s method, a public implementation is not available, and although Zhang tests on a ....
....the best compression (due to the size of the input text) and this section has shown that it is not beneficial to encode space information. For space information to be beneficial, the number of the components in an image must be larger than the number of components in the UW corpus. Although Kia [Kia97] suggested that insertion of spaces can help compression, we have shown that for single pages in the absence of training information, space insertion does not give compression gains. 4.4.3 Evaluating the docstrum enhancements In Sections 4.4.1 and 4.4.2 two enhancements to the docstrum method ....
[Article contains additional citation context not shown here]
Omid Ebrahimi Kia. Document Image Compression and Analysis. PhD thesis, University of Maryland at College Park, 1997.
Integrated Segmentation and Clustering for Enhanced Compression of.. - And Self-citation (Kia) (Correct)
....and a more stable representation for processing in the compressed domain. 1 Introduction Recent advances in document image compression have capitalized on the symbol level redundancies at the symbol level to achieve high levels of compression. This is accomplished by using a symbol lookup table [1, 2, 6, 8, 9, 10] and adding residual information (if necessary) to define a context based symbolic compression [4, 9] These methods can produce high compression ratios and facilitate lossy compression and compressed domain processing because they allow symbol level access. The underlying premise for these ....
....of dominant components. This method has the desirable characteristic that a stable level of representation is achieved in component shape and size. It also improves compression by 4 . It is important to note that this method is independent of the clustering algorithm [10] and data representation [6, 9] used, and that it can benefit from, and contribute to, their operation. For example, a better clustering algorithm that can achieve a higher level of separation can provide better prototypes to be used in partial matching. Better segmentation results can increase the performance of ....
O. Kia. Document Image Compression and Analysis. PhD thesis, University of Maryland at College Park, 1997.
Duplicate Detection in Symbolically Compressed Documents - Dar-Shyang Lee And (1999) (4 citations) (Correct)
No context found.
O. E. Kia, "Document image compression and analysis," Ph.D. dissertation, Depart of Electrical Engineering, University of Maryland, 1997.
An efficient union-find algorithm for extracting the.. - Lapeyre, Strandh (2004) (Correct)
No context found.
Omid Ebrahimi Kia. Document image compression and analysis. PhD thesis, Faculty of the Graduate School of the University of Maryland at College Park, 1997.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC