工程制图习题答题区域分割算法研究
Research on Region Segmentation Algorithm of Engineering Drawing Exercises Based on Hough Transform
DOI: 10.12677/CSA.2021.114104, PDF, 下载: 466  浏览: 569  国家科技经费支持
作者: 董志宏, 罗立宏:广东工业大学,广东 广州
关键词: 工程制图习题图像处理图像分割Hough变换投影法Engineering Drawing Exercises Image Processing Image Segmentation Hough TransformProjection Method
摘要: 针对工程制图习题册数字化阅卷时不能快速从中分割出学生答题区域的问题,提出了一种基于投影法的答题区域分割算法。首先对习题图像进行灰度转换、图像二值化等预处理操作,然后使用Canny算子得到边缘图像,进行Hough直线变换,筛选出符合条件的直线,计算出倾斜角完成图像的倾斜校正;之后找出边框的矩形轮廓,结合水平垂直投影法得到的投影特征定位出各小题的图形位置,实现对答题区域的分割。实验结果表明,该算法很好地实现了对工程制图习题的答题区域分割,具有效率高,准确率高的特点。
Abstract: In order to solve the problem that the students’ question area cannot be quickly segmented from the digital marking of engineering drawing exercise book, an algorithm of question area segmentation based on projection method is proposed. Firstly, preprocess the exercise image such as gray conversion and image binarization, then use Canny operator to get the edge image, carry out Hough line transformation, select the line that meets the conditions, calculate the tilt angle, and complete the tilt correction of the image; then find out the rectangular outline of the frame, and locate the graphic position of each sub problem combined with the projection features obtained by the horizontal and vertical projection method, realize the segmentation of the answer area. The experimental results show that the algorithm can segment the answer area of engineering drawing exercises well, and has the characteristics of high efficiency and high accuracy.
文章引用:董志宏, 罗立宏. 工程制图习题答题区域分割算法研究[J]. 计算机科学与应用, 2021, 11(4): 1008-1018. https://doi.org/10.12677/CSA.2021.114104

参考文献

[1] 赵凯莉, 李鹏, 亓立. “工程制图”课程智慧课堂建设与教学模式的探索[J]. 现代信息科技, 2020, 4(21): 195-198.
[2] 刘淇, 李瑞阳, 李雨盈, 刘佳文. 基于文字识别技术的作业自动批改系统[J]. 电子技术与软件工程, 2020(19): 31-33.
[3] 要曙丽, 王少荣, 盖孟, 王震. 一种答题卡客观题识别算法[J]. 图学学报, 2019, 40(5): 900-907.
[4] 张菲菲, 种兰祥. 基于计算机视觉的答题卡识别系统设计与实现[J]. 中国教育信息化, 2019(6): 51-54.
[5] 罗朝阳, 张鹏超, 姚晋晋, 王彦, 赵瑞杰. 基于Hough变换的答题卡识别[J]. 计算机应用与软件, 2020, 37(3): 251-256.
[6] 方慧琴, 程永强. 阅卷系统中的答题区域快速智能分割算法研究[J]. 现代电子技术, 2018, 41(20): 33-36, 41.
[7] 郝泽兴, 郭改枝. 基于改进Canny算子和Hough变换的表格图像校正方法[J]. 内蒙古师范大学学报(自然科学汉文版), 2020, 49(5): 433-441.
[8] Anagha, P.H. and Baskar, A. (2020) An Automatic Histogram Detection and Information Extraction from Document Images. International Journal of Speech Technology, 24, 77-85.
https://doi.org/10.1007/s10772-020-09756-1
[9] Bezmaternykh, P. and Nikolaev, D.P. (2020) A Docu-ment Skew Detection Method Using Fast Hough Transform. 12th International Conference on Machine Vision, Amster-dam, 16-18 November 2019, Article ID: 114330J.
https://doi.org/10.1117/12.2559069
[10] 郑行家, 钟宝江. 图像直线段检测算法综述与测评[J]. 计算机工程与应用, 2019, 55(17): 9-19.
[11] 王琦, 宋伟东, 王竞雪. 基于Hough变换的直线提取方法及改进[J]. 测绘与空间地理信息, 2019, 42(6): 214-217+221.
[12] Chettat, M., Gaceb, D. and Belhadi, S. (2020) A Fast High Precision Skew Angle Estimation of Digitized Documents. International Arab Journal of Information Technology, 17, 825-831.
https://doi.org/10.34028/iajit/17/5/16
[13] Oba, B., Wkh, A. and Dm, B. (2020) Using Skeleton and Hough Transform Variant to Correct Skew in Historical Documents. Mathematics and Computers in Simulation, 167, 389-403.
https://doi.org/10.1016/j.matcom.2019.05.009
[14] Ramanan, M. (2019) A Hybrid Approach for Skew Detection and Correction in the Multi-Script Scanned Document. Asian Journal of Research in Computer Science, 4, 1-8.
https://doi.org/10.9734/ajrcos/2019/v4i230112
[15] Martorell, O., Buades, A. and Lisani, J.L. (2021) Multiscale Detection of Circles, Ellipses and Line Segments, Robust to Noise and Blur. IEEE Access, 9, 25554-25578.
https://doi.org/10.1109/ACCESS.2021.3056795
[16] 周再文, 王建, 朱恰, 刘星雨, 马紫雯, 高贤君. 基于最小外接矩形的遥感影像建筑物轮廓优化方法[J]. 北京测绘, 2021, 35(1): 1-6.
[17] Babczyński, T. and Ptak, R. (2020) Line Segmentation of Handwritten Text Using Histograms and Tensor Voting. International Journal of Applied Mathematics and Computer Science, 27, 195-206.
[18] 王莉丽, 陈晔, 刘玲. 基于投影轮廓分析的文本图像版面分割算法研究[J]. 数字技术与应用, 2017(3): 164-165.