Proceedings ArticleDOI
Document Image Indexing Using Edit Distance Based Hashing
Ehtesham Hassan,Santanu Chaudhury,M. Gopal +2 more
- pp 1200-1204
Reads0
Chats0
TLDR
A novel word image based document indexing scheme by combination of string matching and hashing is presented for two document image collections belonging to Devanagari and Bengali script.Abstract:
We present a novel word image based document indexing scheme by combination of string matching and hashing The word image representation is defined by string codes obtained by unsupervised learning over graphical primitives The indexing framework is defined by distance based hashing function which does the object projection to hash space by preserving their distances We have used edit distance based string matching for defining the hashing function and for approximate nearest neighbor based retrieval The application of the proposed indexing framework is presented for two document image collections belonging to Devanagari and Bengali scriptread more
Citations
More filters
References
More filters
Proceedings ArticleDOI
Indexing of handwritten document images
TL;DR: A method for fast localization of query words in handwritten images by an adaptation of the principle of geometric hashing is presented that uses consecutive features along curves to produce small-sized image hash tables that also enable fast indexing.
Proceedings ArticleDOI
Content-oriented categorization of document images
TL;DR: Using a vector space classifier with a scanned document image database, it is shown that the word shape token-based approach is quite adequate for content-oriented categorization in terms of accuracy compared with conventional OCR-based approaches.
Proceedings ArticleDOI
An Efficient Similarity Searching Scheme in Massive Databases
TL;DR: A new LSH-based similarity searching scheme, namely SMLSH is proposed, which intelligently combines a consistent hash function and min-wise independent permutations into LSH and effectively classifies information according to the similarity with reduced memory space requirement and in a very efficient manner.
Proceedings ArticleDOI
A Scalable Content-based Image Retrieval Scheme Using Locality-sensitive Hashing
Wang Weihong,Wang Song +1 more
TL;DR: This paper proposes a scalable content-based image retrieval scheme using locality-sensitive hashing (LSH), and conducts extensive evaluations on a large image test-bed of a half million images, which is promising for building web-scale CBIR systems.
Proceedings ArticleDOI
Use of MKL as symbol classifier for Gujarati character recognition
TL;DR: The MKL based classification is proposed, where the MKL is used for learning optimal combination of different features for classification and the comparison results in 1-Vs-1 framework and using KNN classifier are presented.