Title :
Neural-based solutions for the segmentation and recognition of difficult handwritten words from a benchmark database
Author :
Blumenstein, M. ; Verma, B.
Author_Institution :
Sch. of Inf. Technol., Griffith Univ., Brisbane, Qld., Australia
Abstract :
A new intelligent segmentation technique is proposed that may be used in conjunction with a neural classifier and a simple lexicon for the recognition of difficult handwritten words. A heuristic segmentation algorithm is initially used to over-segment each word. An artificial neural network (ANN) trained with 32,034 segmentation points is then used to verify the validity of the segmentation points found. Following segmentation, character matrices from each word are extracted, normalised and then passed through a global feature extractor, after which a second ANN trained with segmented characters is used for classification. These recognised characters are grouped into words and presented to a variable-length lexicon that utilises a string processing algorithm to compare and retrieve those words with the highest confidences. This research provides promising results for segmentation, character and word recognition
Keywords :
document image processing; feature extraction; handwritten character recognition; image classification; image segmentation; matrix algebra; neural nets; string matching; artificial neural network training; benchmark database; character matrices; character recognition; confidence; difficult handwritten words; global feature extractor; heuristic segmentation algorithm; intelligent segmentation technique; neural classifier; over-segmentation; segmented characters; string processing algorithm; variable-length lexicon; word recognition; Character recognition; Databases; Gold; Handwriting recognition; Histograms; Image segmentation; Information technology; Postal services; Read only memory; Telephony;
Conference_Titel :
Document Analysis and Recognition, 1999. ICDAR '99. Proceedings of the Fifth International Conference on
Conference_Location :
Bangalore
Print_ISBN :
0-7695-0318-7
DOI :
10.1109/ICDAR.1999.791779