中文摘要：

为提高文档图像在对比度低、光照不均、大块噪声等复杂图像背景下二值化效果,提出一种背景估计与边缘检测相结合的图像二值化方法。该方法先利用Sauvola算法有效地估计图像的背景,并在此基础上,结合改进的Canny算法获取边缘变化信息,利用基于局部阈值的策略进行二值化。实验结果表明,该算法取得了较好的二值化效果,在开放的DIBCO 2011数据集上测试,其性能与Otsu、Niblack、Sauvola经典方法相比有大幅提高,其F值比ICDAR2011二值化竞赛中第一名的算法略有提高。

英文摘要：

To enhance the binarisation effect of image documents in complex image background of low contrast,uneven illumination and bulk noise,etc.,we present an image binarisation method combining the background estimation and edge detection. Firstly,Sauvola＇s algorithm is used in the method to estimate image background effectively. On this basis,the improved Canny edge detection algorithm is combined to get edge variation information. At last the local threshold-based policy is employed for binarisation. Experimental results demonstrate that the algorithm achieves better binarisation effect,test is carried out on open Document Image Binarisation Contest（DIBCO）2011 dataset,the performance of the proposed method has promising improvement than classic methods of Otsu＇s,Niblack＇s and Sauvola＇s,its F value is slightly higher than that of the No. 1 algorithm in ICDAR2011 binarisation contest.