Questionnaires are widely used for investigation and statistical analysis. However, paper-based questionnaires require great human resources to count the statistical results or enter data into database, which are time consuming. A system capable of recognizing results of questionnaires will be very useful in many aspects. In this paper, we develop a fast and robust digital recognition system for questionnaire on mobile device. The system trains a questionnaire classifier and detects the questionnaire at first. Then it calibrates the detected questionnaire through matrix transform, and digitizes the options of the questionnaire with the improved binarization method. Finally, it determines the chosen options by duty ration of each option and outputs the content of the selected options for statistical analysis and data entry. The system is written in C++ and Java language with libraries of OpenGL and OpenCV on Android platform. In our experiments, the system has a high speed for identification and a high accuracy for recognition in the complicated background.
[1]
N. V. Chandrasekara,et al.
Automated response recognition system for questionnaires
,
2013,
2013 8th International Conference on Computer Science & Education.
[2]
Que Li.
Technology of Selective Questionnaire Recognition based on Neural Network
,
2012
.
[3]
Øivind Due Trier,et al.
Evaluation of Binarization Methods for Document Images
,
1995,
IEEE Trans. Pattern Anal. Mach. Intell..
[4]
Kavita Ahuja,et al.
Object Recognition by Template Matching Using Correlations and Phase Angle Method
,
2013
.
[5]
Corey Greenlaw,et al.
A Comparison of Web-Based and Paper-Based Survey Methods
,
2009,
Evaluation review.