Abstract | ||
---|---|---|
We propose an approach to restore severely degraded document images using a probabilistic context model. Un- like traditional approaches that use previously learned prior models to restore an image, we are able to learn the text model from the degraded document itself, making the approach independent of script, font, style, etc. We model the contextual relationship using an MRF. The ability to work with larger patch sizes allows us to deal with severe degradations including cuts, blobs, merges and vandalized documents. Our approach can also integrate document restoration and super-resolution into a single framework, thus directly generating high quality images from degraded documents. Experimental results show significant improve- ment in image quality on document images collected from various sources including magazines and books, and com- prehensivelydemonstratethe robustness and adaptabilityof the approach. It works well with document collections such as books, even with severe degradations, and hence is ide- ally suited for repositories such as digital libraries. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/CVPRW.2009.5206601 | CVPR |
Keywords | Field | DocType |
Markov processes,document image processing,image resolution,image restoration,probability,random processes,text analysis,MRF,Markov random field,contextual restoration,degraded document image restoration,digital library,probabilistic context model,super-resolution image,text model | Computer vision,Computer science,Markov random field,Font,Image quality,Robustness (computer science),Context model,Artificial intelligence,Probabilistic logic,Digital library,Image restoration | Conference |
ISSN | Citations | PageRank |
1063-6919 | 26 | 1.09 |
References | Authors | |
16 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jyotirmoy Banerjee | 1 | 55 | 4.19 |
Anoop M. Namboodiri | 2 | 255 | 26.36 |
C. V. Jawahar | 3 | 1700 | 148.58 |