Abstract:
Based on the analysis of Uyghur character structure features, this paper studies the characteristics of Uyghur character in wavelet transform and according to these characteristics it presents a localization method of Uyghur character in the video image based on wavelet transform. By using local time-frequency and multi-scale featured wavelet analysis method, distinguished texture features with different space resolutions and the edge components with different directions sub-image are extracted. Then by mathematical morphology method, the useless information is greatly decreased and the ultimate target text region is acquired. Experiments show that the proposed method can detect text effectively and is suited for the localization of Uyghur characters in complex background.
Key words:
video image,
character localization,
wavelet transform,
Uyghur character,
wavelet analysis
摘要: 在对维吾尔文字结构特点分析的基础上,研究维吾尔文字在小波变换中的特征,依据这些特征提出基于小波变换的视频图像中维吾尔文字定位的方法。用具有良好的时频局部和变尺度特性的小波分析方法,提取出纹理清晰、具有不同空间分辨率、不同方向的边缘子图像,用数学形态学的方法对提取出的细节图像进行进一步的形态运算,消除噪声,得到最终的文本目标区域。实验表明该方法简单有效,适合于背景较复杂的维吾尔文字定位。
关键词:
视频图像,
文字定位,
小波变换,
维吾尔文字,
小波分析
CLC Number:
HA Li-Dan-?A, KU Er-Ban-?Mai-Chi-Mu-Sha. Localization of Uyghur Character in Image Based on Wavelet Transform[J]. Computer Engineering, 2011, 37(8): 197-198.
哈力旦?A, 库尔班?买提木沙. 基于小波变换的图像中维吾尔文字定位[J]. 计算机工程, 2011, 37(8): 197-198.