A template-based method for identifying input regions in survey forms

Authors:
P. H. Chien;G. C. Lee
Affiliations:
Department of Computer Science and Information Engineering, National Taiwan Normal University, Taipei, Taiwan;Department of Computer Science and Information Engineering, National Taiwan Normal University, Taipei, Taiwan
Venue:
Pattern Recognition and Image Analysis
Year:
2011

Citing 8
Cited 0

Intelligent forms processing system

Machine Vision and Applications - Special issue: document image analysis techniques
Extraction of data from preprinted forms

Machine Vision and Applications - Special issue: document image analysis techniques
A Generic System for Form Dropout

IEEE Transactions on Pattern Analysis and Machine Intelligence
Description and recognition of form and automated form data entry

ICDAR '95 Proceedings of the Third International Conference on Document Analysis and Recognition (Volume 2) - Volume 2
Use of colour for hand-filled form analysis and recognition

Pattern Analysis & Applications
Universal Data Capture Technology from Semi-structured Form

ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Hough transform based fast skew detection and accurate skew correction methods

Pattern Recognition
A Model-Based Field Frame Detection for Handwritten Filled-in Forms

DAS '08 Proceedings of the 2008 The Eighth IAPR International Workshop on Document Analysis Systems

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper presents a template-based approach for survey form analysis. User input areas on the survey forms are identified. Regions of interest (ROI) include checkboxes, radio buttons, underlined and parenthesized handwritten regions. Scale invariant vertical and horizontal projection profile patterns for each of those four types of ROI are defined as matching templates. The templates are matched against projection profiles of connected components in the given survey form. Experiments with 158 different forms at 100-dpi scan resolution resulted in recognition rate of over 99%.