Abstract: This paper proposes a specific method for localization and validation of Uyghur text in video frames according to the characteristics of Uyghur font. This method gets four edge maps separately in horizontal, vertical, up-right and up-left direction by using the RGB color edge detection operator, then the improved fuzzy C-means clustering algorithm is applied to detect candidate text regions according to the weighed edge map, the false text regions are removed under the heuristic rules of true text regions. The characteristic of the Uyghur text baseline is used to check whether the detected text regions are true Uyghur text regions or not. The experimental results show that the proposed method achieves good effects in the Uyghur video images both in simple background and complex background.

Key words: Uyghur, edge detection, fuzzy C means clustering, text baseline

摘要： 在基于维吾尔文特殊字体的基础上，提出了一种维吾尔文视频文字定位方法。该方法首先利用RGB彩色边缘检测算子获得水平、垂直、右上方和左上方的边缘图，然后根据加权后的边缘图提取图像的纹理特征，用改进的模糊C均值聚类算法检测出候选的文本区域，根据文本区域的启发式规则，去除虚假的文本区域，最后由维吾尔文本的基线特征判定检测出的区域是否为维吾尔文本区域。实验结果表明，这种方法在简单背景和复杂背景视频图像中均具有较好的效果。

关键词: 维吾尔文, 边缘检测, 模糊C均值聚类, 文本基线