Today, good performance of handwriting recognition system has high complexity or complex computation especially in training and classification. We have developed an offline handwriting recognition system with structural approach. Each character through the stages of pre-processing, structural feature extraction and classification process using a combination of similarity endpoint, branch, line and curve, loop, number and position of each feature obtained from the endpoint and branch. This research focuses on feature extraction stage and classification process. Classification process performed using three stages: selection of dataset, mounting features and calculation similarity. Because of acquisition process of handwriting were performed using offline method, then confounding elements becomes very high. The approach taken in this research can be improved its level accuracy of detection digit number to 89,80%, capital letters 86,60% and normal letter 84,92%.
This research try to apply the thinning stage using a river-lake algorithm. It has an advantage in the algorithm speed. The algorithm approach collects points which are assumed as the center point in vertical and horizontal set which every vertical point is connected to horizontal set through projection. The result of the skeleton extraction has single-pixel width which represents starting shape and is conducted in high speed process. In addition, feature extraction and classification using a structural approach, so that the classification process uses a combination of similarity of endpoints, branch points, lines, curves and rings (loops), the number and position of each character features that obtained through solving process of the endpoint, branches and junctions. Stages classification is done in three stages, namely the stage of selecting a dataset, matching features and similarity calculation. The approach taken in this research, can be prove that this method can be apply to recognition technique.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.