Distinction between handwritten and machine-printed text based on the bag of visual words model

作者：

Highlights：

• Separating handwritten from machine printed text using the BoVW model.

• Exploring various local features and weighting types.

• A decision scheme which relies upon the combination of binary SVM classifiers.

• Three distinct databases are provided for future evaluations.

• Each database contains different machine/handwritten separation scenarios.

摘要

Highlights•Separating handwritten from machine printed text using the BoVW model.•Exploring various local features and weighting types.•A decision scheme which relies upon the combination of binary SVM classifiers.•Three distinct databases are provided for future evaluations.•Each database contains different machine/handwritten separation scenarios.

论文关键词：Bag of visual words,Local features,Support vector machines,Page layout

论文评审过程：Available online 20 September 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.09.005