基于改进SURF特征与模糊推理的复杂图片中的文字识别

doi:10.3969/j.issn.1006-2475.2019.04.013

计算机与现代化

基于改进SURF特征与模糊推理的复杂图片中的文字识别

（陕西科技大学电气与信息工程学院，陕西西安710021）

收稿日期:2018-10-30 出版日期:2019-04-26 发布日期:2019-04-30
作者简介:陶筱娇(1989-)，女（布依族），贵州都匀人，助理工程师，硕士，研究方向：模式识别，图像处理,E-mail: taoxiaojiao11@163.com；卢锦（1984-），女，陕西西安人，讲师，博士，研究方向：模式识别。
基金资助:
陕西省教育厅科研计划项目（17JK0084）

Chinese Character Recognition in Complex Images Based on #br# Improved SURF Descriptor Features and Fuzzy Reasoning

(College of Electrical & Information Engineering, Shaanxi University of Science & Technology, Xi’an 710021, China)

Received:2018-10-30 Online:2019-04-26 Published:2019-04-30

摘要/Abstract

摘要： 鉴于现有文字匹配算法在位置、方向和亮度变化上缺乏足够的鲁棒性，根据汉字结构的特殊性，本文采用改进的SURF算法——SSURF来提取文字特征。首先，计算所有训练样本的SSURF描述符，并将同一类别样本的描述符互相匹配，然后计算匹配次数超过1/2的关键点的匹配率，最后用训练样本SSURF描述符的均值和SSURF描述符与均值的最大欧氏距离来建立类数据库。在识别过程中，计算待识别文字的所有关键点，并将关键点的最大模糊匹配度作为该点的模糊匹配度，最后基于模糊推理实现文字识别。实验结果表明，本文算法识别性能更好。

关键词: 汉字识别, SURF算法, 形状信息, 模糊推理, SSURF算法

Abstract: In considering the robustness of Chinese characters matching to the variations of orientation, position and brightness, this paper describes a new method based on improved Speeded Up Robust Features named SSURF to extract characters features. Firstly, the SSURF descriptors of the same class samples are matched. Then the matching rate of key points whose matching times exceed 1/2 is calculated. Finally, the mean value of SSURF descriptors of training samples and the maximum Euclidean distance between SSURF descriptors and mean values are used to establish class database.Experimental results demonstrate that the proposed method yields a better performance.

Key words: Chinese character recognition, SURF, shape information, fuzzy reasoning, SSURF

中图分类号:

TP391.4

陶筱娇，卢锦. 基于改进SURF特征与模糊推理的复杂图片中的文字识别[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2019.04.013.

TAO Xiao-jiao, LU Jin . Chinese Character Recognition in Complex Images Based on #br# Improved SURF Descriptor Features and Fuzzy Reasoning[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2019.04.013.

参考文献［23］

［1］	NAGY G. Twenty years of document image analysis in PAMI［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000,22(1):38-62.
［2］	LIU Y， KASVAND T. A new approach to machine recognition of Chinese characters［C］// Proceedings of the 7th International Conference on Pattern Recognition. 1984:381-384.
［3］	金莲芳,覃团发,王海霞,等. 基于结构量化直方图的图像检索方法［J］. 中国图象图形学报, 2006,11(2):180-185.
［4］	丁世举,顾乃杰,黄章进,等. 一种基于文字识别和页面布局的APP控件识别方法［J］. 计算机工程， 2018,44(5):220-226.
［5］	罗晓清,王鹏飞. 基于尺度不变特征变换的苹果图像融合［J］. 计算机与现代化, 2017(10):5-9.
［6］	兰佩,方超. 基于全局与局部特征融合的人脸识别方法［J］. 计算机与现代化， 2014(3)：109-113.
［7］	金贞,齐开悦,陈凯. 一种适用于复杂图像的汉字特征［J］. 信息安全与通信保密, 2010(2):62-64.
［8］	LOWE D G. Distinctive image features from scale-invariant keypoints［J］. International Journal of Computer Vision, 2004,60(2):91-110.
［9］	BAY H, ESS A, TUYTELAARS T， et al. Speeded-up robust features(SURF)［C］// Computer Vision and Image Understanding, 2008,110(3):346-359.
［10］	ZHOU Y, ZHENG Q, CHEN K. Chinese Character Recognition Using Local Feature［EB/OL］. ［2009-07-30］. http://cbir.sjtu.edu.cn/kchen/.
［11］	SHI C, WANG C, XIAO B, et al. Scene text recognition using part-based tree-structured character detection［C］// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR). 2013:2961-2968.
［12］	ANZID H, GOIC G L, BEKKARI A, et al. Multimodal images classification using Dense SURF, spectral information and support vector machine［J］. Procedia Computer Science, 2019,148:107-115.
［13］	FISCHLER M A, BOLLES R C. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography［J］. Communications of ACM, 1981,24(6):381-395.
［14］	王生生,郭湑,张家晨,等. 基于全局与局部形状特征融合的形状识别算法［J］. 吉林大学学报(工学版), 2016,45(5):1627-1632.
［15］	CANNY J. A computational approach to edge detection［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986,8(6):679-698.
［16］	董瑶,李伟超,刘今越,等. 用于航空发动机叶片焊接修复的改进Canny算法［J］. 焊接学报, 2018,39(1):37-40.
［17］	王婷婷,何科峰,程然. 具有形状信息的多个群目标跟踪算法［J］. 雷达科学与技术, 2017,15(5):531-536.
［18］	BELONGIE S, MALIK J, PUZICHA J. Shape matching and object recognition using shape contexts［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24(4):509-522.
［19］	NEWELL A J, GRIFFIN L D. Multiscale histogram of oriented gradient descriptors for robust character recognition［C］// Proceedings of IEEE the 11th International Conference on Document Analysis and Recognition. 2011:1085-1089.
［20］	WANG K, BELONGIE S. Word spotting in the wild［C］// Proceedings of the European Conference on Computer Vision. 2010:591-604.
［21］	YAN J K, ZHENG H, XI J M. Logo recognition in low quality document images［C］// Proceedings of International Conference on Intelligent Information Technology. 2002:185-189.
［22］	YAN J K, ZHANG Z, ZHENG H. Application of deformable grid in image recognition［J］. Journal of Electronics & Information Technology, 2004,26(8):1183-1189.
［23］	王正杰,杨伟丽,王喆,等. 4种分类算法参数选择及分类特点研究［J］. 计算机与现代化， 2018(2):54-60.

[1]	杨慧婕，刘微，黄先莉，刘守印. 基于模糊C均值改进算法和ANFIS的蓄电池SOC预测[J]. 计算机与现代化, 2017, 0(12): 111-116.
[2]	刘江南1，谢玮1，马家辰1,2. 基于ANFIS的机械手运动学逆解的求取[J]. 计算机与现代化, 2015, 0(10): 112-117.

基于改进SURF特征与模糊推理的复杂图片中的文字识别

Chinese Character Recognition in Complex Images Based on #br# Improved SURF Descriptor Features and Fuzzy Reasoning

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献［23］

相关文章 2

编辑推荐

Metrics

本文评价

基于改进SURF特征与模糊推理的复杂图片中的文字识别

Chinese Character Recognition in Complex Images Based on #br# Improved SURF Descriptor Features and Fuzzy Reasoning

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献 ［23］

相关文章 2

编辑推荐

Metrics

本文评价

参考文献［23］