一个文本图像二值化PDE模型
首发时间:2019-04-22
摘要:文本图像二值化在文本图像分析和模式识别中起着至关重要的作用。最近几年,基于偏微分方程(Partial Differential Equation, PDE)的文本图像二值化方法引起了众多研究者们的兴趣。基于这一方法,本文利用局部力和全局力的结合来提取文本图像的全局和局部的强度信息,提出了一个文本图像二值化模型。在数值实现时,本文采用三步分裂的方法来有效地求解对应的演化方程,即将演化方程拆分为两个线性微分方程和一个非线性微分方程。通过与四种基于PDE的二值化方法和经典的Ostu法进行定性和定量地对比,所提模型有最优的实验结果。
关键词: 文本图像二值化 偏微分方程; 三步分裂法
For information in English, please click here
A PDE model based for document image binarization
Abstract:Document image binarization is an important step in the document image analysis and pattern recognition. In recent years, document image binarization models based on partial differential equation (PDE) have attracted the interest of many researchers. This paper presents a binarization method for document images, which is a PDF model. The proposed model extracts the global and local intensity information by the global and local force, respectively, which has good performance for handling document image binarization. A three-step time-splitting scheme is used to numerically solve the evolution equation efficiently, in which the evolution equation is decomposed into two linear differential equations and a nonlinear differential equation. The experimental results show that the proposed method has the better performance compared with five state-of-the-art methods qualitatively and quantitatively.
Keywords: document image binarization partial differential equation time-splitting scheme
基金:
引用
No.****
同行评议
共计0人参与
勘误表
一个文本图像二值化PDE模型
评论
全部评论0/1000