摘 要：盲文是视力障碍人士用来学习技能和了解世界的重要手段，盲文检测则是促进视力障碍人士和正常人士交流的关键技术。针对盲文扫描图像，利用 HOG（Histogram of Oriented Gradient）特征和 SVM（Support Vector Machine）提出了一种盲文检测方法，并在盲文扫描图像数据集 DSBI 上进行了验证，实验结果表明，所提出的方法能够有效检测盲文点信息。
关键词：视力障碍；盲文检测；HOG；SVM
DOI:10.19850/j.cnki.2096-4706.2021.15.035
基金项目：湛江市非资助科技攻关计划项 目（2021B01136）；广东省特殊儿童发展与教育重点实验室项目（TJ202011）；广东省教育厅普 通高校特色创新项目（2021KTSCX065）；岭南 师范学院创建国家教师教育创新实验区项目研究成果
中图分类号：TP391.4 文献标识码：A 文章编号：2096-4706（2021）15-0131-03
Braille Detection Method Based on SVM
LU Liqiong1，2，WU Dong1，2
(1.School of Computer Science and Intelligence Education, Lingnan Normal University, Zhanjiang 524048, China; 2.Guangdong Provincial Key Laboratory of Development and Education for Special Needs Children, Lingnan Normal University, Zhanjiang 524048, China)
Abstract: Braille is an important means for the visually impaired people to learn skills and understand the world. And braille detection is the key technology to promote the communication between visually impaired people and normal people. For braille scanned images, a braille detection method is proposed by using HOG (Histogram of Oriented Gradient) feature and SVM (Support Vector Machine), and verified on braille scanned image data set DSBI. The experimental results show that the proposed method can effectively detect braille point information.
Keywords: visually impaired; braille detection; HOG; SVM
参考文献：
[1] 中国青年网 . 有 1800 万人 , 在声音里寻找光 [EB/OL]. （2020-12-06）.https://baijiahao.baidu.com/s?id=1685329774 912727962&wfr=spider&for=pc,2020.12.06.
[2] ISAYED S，TAHBOUB R. A review of optical Braille recognition [C]//2015 2nd World Symposium on Web Applications and Networking (WSWAN). Sousse：IEEE，2015：1-6.
[3] LI R Q，LIU H，WANG X D, et al. DSBI: DoubleSided Braille Image Dataset and Algorithm Evaluation for Braille Dots Detection [C]//ICVIP 2018: Proceedings of the 2018 the 2nd International Conference on Video and Image Processing. 2018.
[4] DALAL N，TRIGGS B. Histograms of oriented gradients for human detection[C]//IEEE Computer Society Conference on Computer Vision & Pattern Recognition.San Diego：IEEE，2005：886-893.
[5] NEUBECK A，GOOL LJV. Efficient Non-Maximum Suppression [C]//International Conference on Pattern Recognition (ICPR).Hong Kong：IEEE，2006：850-855.
作者简介：卢利琼（1980—），女，汉族，湖北崇阳人，讲师，博士，主要研究方向：文本识别；吴东（1981—），男，汉族，广 东合浦人，副教授，硕士，主要研究方向：模式识别