OCR识别系统通常采用扫描仪获取表格文档的原始图像，由计算机进行版面分析和字符识别。
The scanner is commonly used in OCR recognition system to acquire document original image, and the image transmit to computer for document layout analysis and OCR recognition.
在分析现有倾斜检测算法基础上，提出一种改进的文档图像倾斜检测算法。
Based on the analysis of existing algorithms, an improved skew detection algorithm for document is proposed.
如果文档图像中同时存在多个条码，进一步使用了基于形态学的版面分析技术，实现多个条码区域的自动分离。
When many barcodes occur in the same image, pattern analysis based on morphology is used further to divide every part.
应用推荐