一种新的表格字符分割算法的研究与实现

Research and Implementation of A New Kind of Character with Form Segmentation Algorithm

针对表格图像中,表格线及可能存在的表格与字符粘连交叠而导致字符分割困难的问题,提出一种新的表格字符分割方法。该算法基于二阶高斯微分算子,对图像进行两方向的滤波,提取水平和垂直表格框线;确定表格字符区域的位置后,提出相交算法恢复粘连交叠区域,分割出字符区域;最后垂直投影算法分割出单个字符。实验结果表明,该算法稳定性高,效果好,与已有方法相比,提高了表格字符分割效率,具有较高的实际应用价值。

According to the form image, the problem of character segmentation difficult caused by form line and possible folds of form and congtutination, this paper puts forward a new form character segmentation method. The algorithm based on the second-order gaussian differential operator, the images are filtered by two direction , horizontal and vertical form flame line are extracted ; after determining the position of the fore1 character area, this paper puts forward intersection algorithm recovery folds of form and conghitination, segment the character region ; At last, the vertic~ projection algorithm split out a single character. The experimental results show that the algorithm stability is high, the effect is good, compared with the methods available, effectively improve the efficiency of the form character segmentation and high practical value.