Abstract:Aiming at the limitation of function and price of cursor reader in the current market, a new method of answer cards recognition based on digital image processing is proposed. Firstly, the answer card is used for image pre-processing and graphic correction and the YCbCr color space algorithm is used to split out the candidate information area and answer information area. Then, the character matching and vertical synchronization-strips are used to locate their specific coordinates.Finally, the adaptive threshold algorithm is used to binarize the answer card and calculate the each filling area in binary image. Comparing the area with the set threshold, if it is bigger than set threshold, it will be regarded as having been filled. Experiments show that the method has higher recognition speed and accuracy. What’s more, it is not sensitive to the tilt of answer cards when scanning, and it has a better recognition effect on the size of the filling area, the unbalanced paintings, and the slight stains on answer cards, etc.
[1]周万珍,郑广,王建霞, 等. 数字图像处理技术在客观题阅卷中的应用[J].数学的实践与认识, 2006, 36(8):243-247.
[2]郑广,秦敏,付东,等.基于图像识别的客观题阅卷研究[J].仪器仪表学报,2006, 27(z1):783-784,791.
[3]李云峰,贾晨辉. 基于灰度投影的快速纸币图像几何校正[J]. 计算机应用与软件, 2009, 26(6):253-255.
[4]Wakahara T, Kimura Y, Tomono A. Affine-invariant recognition of gray-scale characters using global affine transformation correlation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(4):384-395.
[5]Lo R C, Tsai W H. Perspective-transformation-invariant generalized hough transform for perspective planar shape detection and matching[J]. Pattern Recognition, 1997, 30(3):383-396.
[6]Brancati N, Pietro G D, Frucci M, et al. Human skin detection through correlation rules between the YCb and YCr subspaces based on dynamic color clustering[J]. Computer Vision & Image Understanding, 2017, 155(6):33-42.
[7]Mo S Q, Liu Z G, Zhang J, et al. An unsupervised threshold selection method for edge detection based on image self-information[J]. Journal of Optoelectronics Laser,2011, 22(8):1246-1250.
[8]吴凤和. 基于计算机视觉测量技术的图像轮廓提取方法研究[J]. 计量学报, 2007, 28(1):18-22.
[9]Maik V, Paik J, Cho D, et al. Color Correction for Multiple Light Sources Using Profile Correction System[J]. Journal of Imaging Science & Technology, 2009, 53(5):505011-505018.
[10]Dou Y, Huang Y, Li Q, et al. A fast template matching-based algorithm for railway bolts detection[J]. International Journal of Machine Learning & Cybernetics, 2014, 5(6):835-844.