Resolving Ambiguities in Toponym Recognition in Cartographic Maps

Gelbukh, Alexander; Levachkine, Serguei; Han, Sangyong

doi:10.1007/978-3-540-25977-0_7

Cited by 10 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The extracted text is finally recognized using an artificial neural network in conjunction with a gazetteer. Similar recognition systems have been reported in (Levachkine et al, 2002) and (Gelbukh et al, 2003). In the next section we discuss our proposed approach that aims to minimize the required amount user involvement in the extraction of the text content and overcome the challenges encountered in character recognition for topographic maps.…”

Section: Document Image Analysis and Map Conversion Systems: A Brief Ovmentioning

confidence: 75%

Character Degradation Model and HMM Word Recognition System for Text Extracted from Maps

Pezeshk¹,

Tutwiler²

2011

Recent Advances in Document Recognition and Understanding

View full text Add to dashboard Cite

Section: Document Image Analysis and Map Conversion Systems: A Brief Ovmentioning

confidence: 75%

Character Degradation Model and HMM Word Recognition System for Text Extracted from Maps

Pezeshk¹,

Tutwiler²

2011

Recent Advances in Document Recognition and Understanding

View full text Add to dashboard Cite

“…However, few have integrated geographical background knowledge with the recognition process. Gelbukh et al [9] propose probability models for point, linear, and regional features, assuming the georectification (approximate correspondence between geographical coordinates and image pixels) is given. Using these models, they heuristically evaluate the likelihood of possible toponyms corresponding to a single OCR output and similar strings with small edit distance.…”

Section: A Related Workmentioning

confidence: 99%

“…First, whereas Gelbukh et al [9] assume georectified maps, we propose a robust search method for the coordinate correspondences. Second, rather than using a single OCR output, we rank many gazetteer words by their appearance probabilities, which are then used to calculate the final predicted word.…”

Section: A Related Workmentioning

confidence: 99%

Toponym Recognition in Historical Maps by Gazetteer Alignment

Weinman

2013

2013 12th International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

Abstract-Historical map documents are increasingly digitized for widespread access, but most are only coarsely indexed with meta-data while the contents are largely unsearchable. We propose to increase searchability by automatically recognizing the place names in these digitized artifacts. Using a word recognition system that produces a noisy ranked list of initial hypotheses from a lexicon of viable toponyms, we form a joint probabilistic model for inferring the most likely latent alignment between image toponyms and a gazetteer of known place locations. After a robust generalized RANSAC algorithm identifies the global alignment, we rerank the toponym hypotheses by their posterior probability. Experiments demonstrate a significant boost in word recognition accuracy on a manually annotated set of 19th century U.S. state and regional maps.

show abstract

“…Gelbukh et al [22] [23]. All words that have been recognized, using the gazetteer, are sent to GIS, while those that were not identified, remained in the picture matrix.…”

Section: Separation and Recognition Of Touching And Overlapping Charamentioning

confidence: 99%

“…After this process is finished, it is possible to obtain the whole chain to form the word "Santa". The word is checked in a gazetteer to verify that it is a valid word for this geographical location [22] [23]. Figure 6b shows the lines one, two and three used (line four is under the symbol touching 'S', but it is so far from line three).…”

Section: V-lines (Horizontal Labels)mentioning

confidence: 99%

Text/Graphics Separation and Recognition in Raster-Scanned Color Cartographic Maps

Velázquez¹,

Levachkine

2004

Graphics Recognition. Recent Advances and Perspectives

Self Cite

View full text Add to dashboard Cite

Abstract.A method to separate and recognize the touching/overlapping alphanumeric characters is proposed. The characters are processed in raster-scanned color cartographic maps. The map is segmented first to extract all text strings including those that are touching other symbols, strokes and characters. Second, OCR-based recognition with Artificial Neural Networks (ANN) is applied to define the coordinates, size and orientation of alphanumeric character strings in each case presented in the map. Third, four straight lines or a number of "curves" computed as a function of primarily recognized by ANN characters are extrapolated to separate those symbols that are attached. Finally, the separated characters input into ANN again to be finally identified. Results showed high method's rendering in the context of raster-to-vector conversion of color cartographic images.

show abstract

Resolving Ambiguities in Toponym Recognition in Cartographic Maps

Cited by 10 publications

References 8 publications

Character Degradation Model and HMM Word Recognition System for Text Extracted from Maps

Character Degradation Model and HMM Word Recognition System for Text Extracted from Maps

Toponym Recognition in Historical Maps by Gazetteer Alignment

Text/Graphics Separation and Recognition in Raster-Scanned Color Cartographic Maps

Contact Info

Product

Resources

About