基于卷积神经网络的自然背景字符识别

被引：3

作者：

郁松

彭志文

机构：

[1] 中南大学软件学院

来源：

计算机应用与软件 | 2017年 / 34卷 / 12期

关键词：

自然背景字符识别; 卷积神经网络; 图像分类;

D O I：

暂无

中图分类号：

TP183 [人工神经网络与计算]; TP391.41 [];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ; 080203 ;

摘要：

随着计算机视觉技术的发展,自然背景中字符的识别在图片检索、视频检索、无人车识别周围场景信息等领域都扮演了不可或缺的角色。相对于手写字符、打印字符的识别,自然背景字符的识别有着光照强度变化大、背景纹理复杂、字体样式和颜色多变等特点,这都给识别带来了巨大的挑战。主要是基于Le Net-5的网络结构设计了一种适合于识别自然背景字符的卷积神经网络,由于在这一领域以往的研究工作的基准数据集是较小的数据集(Chars74K-15),为了便于比较,实验也是基于同样的数据集。但因为卷积神经网络是在巨大数据量的驱动下才会有良好的效果,因此还提出了一种预处理方式和fine-tune相结合用于解决自然背景字符图片数据量较小的问题。

引用

页码：228 / 234

页数：7

共 12 条

[1]

Detecting text in natural scenes with stroke width transform. Epshtein B,Ofek E,Wexler Y. IEEE Conference on Computer Vision and Pattern Recognition . 2010

[2]

PhotoOCR:Reading text in uncontrolled conditions. Bissacco A,Cummins M,Netzer Y.et al. IEEE International Conference on Computer Vision . 2013

[3]

Multi-column deep neural networks for image classification. Ciresan D,Meier U,Schmidhuber J. 2012 IEEE Conference on Computer Vision and Pattern Recognition . 2012

[4]

Real-time scene text localization and recognition. L.Neumann,K.Matas. Computer Vision and Pattern Recognition . 2012

[5]

Feature Representations for Scene Text Character Recognition:A Comparative Study. Chuncai Yi,Xiaodong Yang,Yingli Tian. 12th International Conference on Document Analysis and Recognition . 2013

[6]

Scene text recognition using part-based tree-structured character detection. Shi C,Wang C,Xiao B,et al. Computer Vision and Pattern Recognition (CVPR),2013 IEEEConference on . 2013

[7]

Scene text recognition using sparse coding based features. Zhang D,Wang D H,Wang H. IEEE International Conference on Image Processing . 2014

[8]

Scene text recognition with deeper convolutional neural networks. Zhang Y. 2015 IEEE International Conference on Image Processing (ICIP) . 2015

[9]

Scene character recognition using PCANet. Chen C,Wang D H,Wang H. International Conference on Internet Multimedia Computing and Service . 2015

[10]

Character recognition in natural images. T. E. de Campos,B. R. Babu,M. Varma. VISAPP . 2009

← 1 2 →