Questionnaire survey is one of the important research tools in conducting quantitative information collections.Sometimes the collected questionnaires amount to as many as thousands.Therefore,the data input may become one of the key steps during research.The authors have developed an automatic questionnaire input system by computer vision,which is capable of continuously capturing,digitizing and processing the CCD images to accomplish the task of questionnaire data input.This study discusses one of the major aberrations encountered in the system when using a camera to measure the geometric positions on a questionnaire.The radial distortion through a camera lenses is also known as the “fisheye effect” which always appears on the image inevitably.In the calibration procedure a nonlinear polynomial function is often used to account for this aberration.The authors address this calibration issue and provide a solution with a geometric closed form.