Writer identification using curvature-free features

详细信息    查看全文

• 作者：Sheng He ; ^{heshengxgd@gmail.com}Author Vitae ; Lambert Schomaker ^{L.Schomaker@ai.rug.nl}Author Vitae
• 关键词：Writer identification ; Curvature-free ; Run-lengths of local binary pattern ; Cloud of line distribution
• 刊名：Pattern Recognition
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：63
• 期：Complete
• 页码：451-464
• 全文大小：2493 K
• 卷排序：63

文摘

We propose two novel and curvature-free features: LBPruns and COLD features for writer identification. The LBPruns is the joint distribution of run-length and local binary pattern. The COLD is the joint distribution of the relation between orientations and lengths of line segments. The combination of the LBPruns and COLD features provides a significant improvement on the curvature-less CERUG data set.