Title :
A handwritten mechanical drawing recognition system
Author :
Mori, S. ; Maeda, M. ; Kinbara, H. ; Kunieda, T. ; Deguchi, S.
Author_Institution :
Ricoh Ltd., Yokohama, Japan
Abstract :
A handwritten machine drawing recognition system has been developed that is based on a knowledge-based vision paradigm. The system is divided into four units: presegmentation, segmentation, partial recognition, and recognition. The segmentation is tightly related to partial recognition in which a hierarchical projection method is used and each divided region is recognized by the model-based system. Thus, partially recognized and pending regions are globally recognized. The experimental results were quite good
Keywords :
computer vision; computerised pattern recognition; divided region; handwritten machine drawing recognition system; handwritten mechanical drawing recognition system; hierarchical projection method; knowledge-based vision paradigm; model-based system; partial recognition; presegmentation; recognition; segmentation; Character generation; Character recognition; Concrete; Data mining; Data structures; Flowcharts; Handwriting recognition; Indium tin oxide; Logic devices; Testing;
Conference_Titel :
Industrial Applications of Machine Intelligence and Vision, 1989., International Workshop on
Conference_Location :
Tokyo
DOI :
10.1109/MIV.1989.40562
