Convolutional Neural Networks for Document Image Classification

被引：92

作者：

Kang, Le ^{[1
]}

Kumar, Jayant ^{[1
]}

Ye, Peng ^{[1
]}

Li, Yi ^{[2
,3
]}

Doermann, David ^{[1
]}

机构：

[1] Univ Maryland, College Pk, MD 20742 USA

[2] NICTA, Des Plaines, IL USA

[3] Australian Natl Univ, Canberra, ACT, Australia

来源：

2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR) | 2014年

基金：

美国国家科学基金会;

关键词：

D O I：

10.1109/ICPR.2014.546

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents a Convolutional Neural Network (CNN) for document image classification. In particular, document image classes are defined by the structural similarity. Previous approaches rely on hand-crafted features for capturing structural information. In contrast, we propose to learn features from raw image pixels using CNN. The use of CNN is motivated by the the hierarchical nature of document layout. Equipped with rectified linear units and trained with dropout, our CNN performs well even when document layouts present large inner-class variations. Experiments on public challenging datasets demonstrate the effectiveness of the proposed approach.

引用

页码：3168 / 3172

页数：5