一种改进的中文文档图像倾斜检测方法

图像获取设备将纸质文档转换为文档图像时，经常会使文档图像出现某种程度的倾斜，从而可能使后续的文档版面理解和OCR识别算法失败。文中提出一种基于近邻法的中文图像的倾斜角度检测方法，并采用最小二乘法减小倾斜估计的误差，从而大大优化了运算速度，增强了算法的鲁棒性，与现有方法相比，具有运算速度快，检测精度高的优势。算法在Visual C＋＋下编程加以实现，通过对检测库中100幅倾斜中文文档图像的检测证明，该方法具有精度高和适应性强的特点。

An Improved Method for Skew Detection of Chinese Document Images

When image digitalization devices translate document files into document images, the document images are always skewed to some extent, which would result in a failure of the follow - up processing. An improved skew document image detection method based on nearest - neighbor algorithm for Chinese document is presented, in which the least square method is used to reduce the error of skew detection, The new algorithm can accelerate the convergence rate and reduce the error compared with the existing algorithms, such as the nearest - neighbor algorithm. The experimental results of 100 slant Chinese document images show that the proposed method has high precision and robustness.