赵金龙 宁媛

摘 要： 目前答题卡识别主要分为人工阅卷和机器阅卷两种，但是人工阅卷存在效率低、劳动强度大和准确率不高等问题;而传统的机器阅卷主要是光标阅卷机，其对答题卡的要求较高，存在通用性较差、使用成本高和使用率低等缺点。针对这一问题，设计一款基于数字图像处理技术的自动识别答题卡系统，该系统通过Matlab GUI编程，操作简单方便，在普通计算机上就可以实现答题卡自动识别功能，既减轻了教师的阅卷压力，又克服了光标阅卷机必须专卡专用的缺点。结果表明该系统对答题卡具有识别精度高、鲁棒性强和成本低廉等优点，具有很高的使用价值。

关键词： 答题卡; 网上阅卷; 图像处理; Matlab GUI; Hough变换

文章编号： 2095-2163（2021）03-0190-06 中图分类号：TP391 文献标志码：A

【Abstract】At present， answer sheet recognition is mainly divided into manual scoring and machine scoring. However， manual scoring has problems such as low efficiency， high labor intensity and low accuracy. The traditional machine scoring is mainly a cursor scoring machine， which has higher requirements for answer sheets. There are disadvantages such as poor versatility， high use cost and low use rate. In response to this problem， an automatic answer sheet recognition system based on digital image processing technology is designed. The system is programmed by Matlab GUI， and the operation is simple and convenient. The automatic answer sheet recognition function can be realized on an ordinary computer， which not only reduces the pressure on the teacher to mark the exam， but also overcomes the shortcoming that the cursor marking machine must be dedicated to a special card. The results show that the system has the advantages of high recognition accuracy， robustness and low cost for answering question sheets， and has high use value.

【Key words】 answer sheet; online marking; image processing; Matlab GUI;  Hough transform

0 引 言

時下，随着各类考试数量的不断增多，试后的阅卷工作量也有明显增加。传统的人工阅卷弊端较为突出，不仅效率低下，还会耗费阅卷人大量的时间和精力，同时繁重的阅卷工作也很容易给阅卷老师造成大脑疲劳和视觉疲劳，由此就会带来阅卷错误。当前，互联网科技及新兴技术的蓬勃发展则使得传统的人工阅卷已经不能满足现代化需求，网上阅卷系统应运而生。而传统的光标阅卷机（Optical Mark Reader，OMR）虽然可以克服人工阅卷的弊端，但是其购买和维护成本均颇为可观，并且对答题卡有很高的要求，通用性较差，难以在高校内部中小型考试中普及使用。

近年来，随着数字图像处理技术和计……