Offline cursive script word recognition has received increasingattention during the last years. Impressive progress has been achievedin reading isolated single characters during the last decade. Cursivescript recognition still lacks a good recognition rate. Since there is ahigh variability in unconstrainted handwritten script words, the domainis much more difficult than single character recognition. To achieveacceptable results, the context has to be restricted by a given lexiconof all possible words. The only accessible information is the binaryimage of the cursive script word. Since handling of raster data iscumbersome, connectivity analysis is applied as a first processing step.Thereafter it is necessary to reduce the variability as much as possiblewithout losing relevant information. Therefore, some normalization stepsangle, rotation stroke width, and size. The normalization techniques ofthe authors' system and the subsequent feature extraction are presented.The proposed algorithms are every efficient because they are based onthe contour information provided by connectivity analysis
展开▼
