期刊文献+

工业字符识别中实用的预处理技术 被引量:2

Practical Pretreatment Technology in Industry Character Recognition
下载PDF
导出
摘要 为了提高工业字符的识别率,提出了一种快速而实用的预处理技术。结合掩模边界、仿射变换和线性插值对用户选择区域内的字符进行旋转校正;利用投影方法对字符进行分行,为克服点阵字体误分行的情况,提出了波形膨胀的方法;针对黏连字符的情况,根据预定义字符宽度,确定黏连字符个数,并求取预分割范围,再利用投影方法求取最佳分割位置。实际测试表明,经过该方法处理后的字符,能达到99%以上的识别率。目前该方法已经应用到嵌入式机器视觉工业现场。 The paper proposes a practical preprocessing technology to improve the recognition rate of the industry characters.The revolving adjustment is made to the data of ROI,combining the mask boundary,the affine transformation and the linear interpolation.The paper also carrys on the branch using the projection method to the character and propose the wave expansion method to overcome the situation of the branch bitmap fonts error.According to the pre-definition character width,the number of the adhesion characters is determuned.Besides,the pre-division scope and the best division position using projection method are found out.The actual test is made to prove that: the recognition rate of the characters can achieve 99% after this method processing.This method has been applied in the embedded machine vision industrial field.
出处 《江南大学学报(自然科学版)》 CAS 2011年第1期16-20,共5页 Joural of Jiangnan University (Natural Science Edition) 
基金 国家自然科学基金项目(60804013)
关键词 字符识别 仿射变换 波形膨胀 字符分割 character recognition affine transformation profile inflation character division
  • 相关文献

参考文献6

二级参考文献34

共引文献21

同被引文献21

  • 1潘梅森,郭国强.基于图像矩的车牌号码倾斜校正[J].计算机辅助设计与图形学学报,2007,19(8):1041-1045. 被引量:17
  • 2Hinton G E, Osindero S, The Y. A fast learning algorithm for deep belief nets [J]. Neural Computation(S0899-7667), 2006, 18: 1527-1554.
  • 3V1NOD NAIR, GEOFFREY E HINTON. Rectified Linear Units Improve Restricted Boltzmann Machines [C]// Appearing in Proceedings of the 27th International Conference on Machine Learning, Haifa, Israel, 2010: 807-814.
  • 4MATTHEW D ZEILER, ROB FERGUS. Stochastic Pooling for Regularization of Deep Convolutional Neural Networks [C]// International Conference on Learning Representations, Scottsdale, Arizona, USA, May 02-04, 2013:1301-1310.
  • 5Sukittanon S, Surendran A C, Platt J C, et al. Convolutional networks for speech detection [C]// Proceedings of the Eighth International Conference on Spoken Language Processing, Jeju Island: ISCA, 2001: 1077-1080.
  • 6Yin-Nong Chen, Chin-Chuan Han. A CNN-based face detector with a simple feature map and a coarse-to-fine classifier [J]. Transactions on Pattern Analysis and Machine Intelligence(S0162-8828), 2009, 31(3): 1-12.
  • 7Genevieve B Orr, Klaus-Robert Muller. Neural Networks: Tricks of the Trade [M]. Berlin/Heidelberg: Springer Verlag, 1998: 9-50.
  • 8Chang C I, et al. Survey and comparative analysis of entropy and relative entropy thresholding techniques[J]. IEEE Transac- tions on Vision, Image and Signal Processing, 2006, 153 (6) : 837-850.
  • 9Singh S K, et al. Fuzzy edge detection based on maximum entro- py thresholding[J]. IETE Journal of Research, 2011, 57 (4) : 325-330.
  • 10Sahoo P K, et al. A thresholding method based on two-dimen- sional Renyi's entropy[J]. Pattern Recognition, 2004, 37 (6) : 1149-1161.

引证文献2

二级引证文献20

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部