Page Segmentation for Historical Handwritten Documents Using Fully Convolutional Networks

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 模式分析与学习团队

	Page Segmentation for Historical Handwritten Documents Using Fully Convolutional Networks
	Xu Y(徐玥)1,2 ; He WH(何文浩)1,2 ; Yin F(殷飞)2 ; Liu CL(刘成林)1,2
	2017
会议日期	2017.11.9 - 2017.11.15
会议地点	Kyoto, Japan
关键词	Page Segmentation Layout Analysis Fully Convolutional Network
页码	541~546
英文摘要	Page segmentation is a fundamental and challenging task in document image analysis due to the layout diversity. In this work, we propose a pixel-wise segmentation method for historical handwritten documents using fully convolutional network (FCN). The document image is segmented into different regions by classifying pixels into different categories: background, main text body, comments, and decorations. By supervised learning on document images with pixel-wise labels, the FCN can extract discriminative features and perform pixelwise segmentation accurately. After pixel-wise classification, postprocessing steps are taken to reduce noises, correct wrong segmentations and find out overlapping regions. Experimental results on the public dataset DIVA-HisDB containing challenging medieval manuscripts demonstrate the effectiveness and superiority of the proposed method, which yields pixel-level accuracy of above 99%.
语种	英语
内容类型	会议论文
源URL	[http://ir.ia.ac.cn/handle/173211/20019]
专题	自动化研究所_模式识别国家重点实验室_模式分析与学习团队
作者单位	1.中国科学院大学 2.中国科学院自动化研究所，模式识别国家重点实验室
推荐引用方式 GB/T 7714	Xu Y,He WH,Yin F,et al. Page Segmentation for Historical Handwritten Documents Using Fully Convolutional Networks[C]. 见:. Kyoto, Japan. 2017.11.9 - 2017.11.15.