Multiscale segmentation of unstructured document pages using soft decision integration

被引:54
作者
Etemad, K [1 ]
Doermann, D [1 ]
Chellappa, R [1 ]
机构
[1] UNIV MARYLAND, INST ADV COMP STUDIES, COLLEGE PK, MD 20742 USA
关键词
document processing; multiscale analysis; context dependent classification; soft decision integration; wavelet packets; neural networks; CLASSIFICATION;
D O I
10.1109/34.566817
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present an algorithm for layout-independent document page segmentation based on document texture using multiscale feature vectors and fuzzy local decision information. Multiscale feature vectors are classified locally using a neural network to allow soft/fuzzy multi-class membership assignments. Segmentation is performed by integrating soft local decision vectors to reduce their ''ambiguities.''
引用
收藏
页码:92 / 96
页数:5
相关论文
共 16 条