Research Article
Research on Recognition Method of Test Answer Sheet Based on Machine Vision
@INPROCEEDINGS{10.1007/978-3-030-72795-6_56, author={Ping Cui and Dan Li and Kailiang Zhang and Likai Wang and Weiwei Liu}, title={Research on Recognition Method of Test Answer Sheet Based on Machine Vision}, proceedings={Simulation Tools and Techniques. 12th EAI International Conference, SIMUtools 2020, Guiyang, China, August 28-29, 2020, Proceedings, Part II}, proceedings_a={SIMUTOOLS PART 2}, year={2021}, month={4}, keywords={Answer sheet recognition Automatic marking Curvelet algorithm Hough transform}, doi={10.1007/978-3-030-72795-6_56} }- Ping Cui
Dan Li
Kailiang Zhang
Likai Wang
Weiwei Liu
Year: 2021
Research on Recognition Method of Test Answer Sheet Based on Machine Vision
SIMUTOOLS PART 2
Springer
DOI: 10.1007/978-3-030-72795-6_56
Abstract
When using the cursor reading technology to mark the answer card, the cursor machine can only be used for special card, which is expensive and difficult to popularize. A new method of answer sheet recognition based on machine vision and image processing was proposed. Firstly, the improved curvelet algorithm was used to preprocess the image to solve the problem of low resolution and high noise caused by different acquisition methods. Secondly, Hough transform was used to detect lines and correct deformation of binary image. Finally, the answer area was segmented, and the vertical and horizontal projections were used to detect the question and option interval, generate grid lines, mark the center of rectangle and judge the option results. Experiments show that this method is accurate, efficient and robust to low resolution, tilt and noise.
