Multi-Scale Wavelet Texture-Based Script Identification Method
Zeng Li · Chinese Journal of Computers · 2000
There are variety languages or scripts need to identify and process in an international environment. This paper proposes a script identification technique based on multi scale non redundant wavelet texture analysis. The authors choose six language (Chinese, English, Japanese, Korean, Russian and Indian) script samples with different formats and fonts to demonstrate the potential of the technique. Computational complexity analysis and experiments show that multi scale non redundant wavelet texture based script identification method has less computational complexity and more stability for the variety of format or font.