Abstract | ||
---|---|---|
We present a new parts-based multi-scale recognition method for graphic symbols, especially those connecting or intersecting with other elements in the context. The main idea is to decompose the symbol into the set of multi-scale local parts, some of which are not or less affected by the contextual interferences, and then recognize the symbol based on detecting and integrating individual symbol parts. An ensemble learning and classification scheme is employed, which combines three ingredients: 1) the multi-scale spatial pyramid representation of the symbol that consists of local parts for matching. 2) the random forest based classifying of symbol parts and discriminative learning of the mappings between parts and the symbol. 3) the probabilistic aggregation of individual part detections to form the symbol recognition output. The experimental results on simulation datasets show the effectiveness of the proposed method and its promising properties in handling non-segmented symbols. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1007/978-3-642-36824-0_25 | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Keywords | Field | DocType |
new parts-based multi-scale recognition,discriminative learning,symbol part,non-segmented symbol,parts-based multi-scale method,individual part detection,multi-scale spatial pyramid representation,graphic symbol,multi-scale local part,symbol recognition output,individual symbol part | Symbol recognition,Pattern recognition,Symbol,Computer science,Classification scheme,Pyramid (image processing),Artificial intelligence,Probabilistic logic,Random forest,Ensemble learning,Discriminative learning | Conference |
Volume | Issue | ISSN |
7423 LNCS | null | 16113349 |
Citations | PageRank | References |
0 | 0.34 | 20 |
Authors | ||
3 |