基于Lαβ变换和伽马校正的蒙古族家居纹样增强研究

RESEARCH ON THE ENHANCEMENT OF MONGOLIAN HOME FURNISHING PATTERN BASED ON LαβTRANSFORMATION AND GAMMA CORRECTION METHOD

为了蒙古族家居纹样的细节更加清晰,以及更好地改善纹样的视觉效果和提高识别精度,本文提出一种基于Lαβ变换和伽马校正的蒙古族家居纹样增强算法。首先将蒙古族家居的原始彩色纹样的RGB空间的原始彩色纹样转换到Lαβ色彩空间;其次利用伽马校正方法提高纹样的信息熵,增强纹样的细节;最后再将Lαβ色彩空间的纹样返回到RGB空间。结果表明,本算法与CLAHE方法比较,不仅提高了纹样的客观评价指标,丰富了图像中存在的信息,而且使纹样主观上细节更加突出、纹理更加清晰,其增强效果更好。

In order to make the details of the Mongolian home furnishing pattern clearer,and to better improve the visual effect and recognition accuracy of the patterns,a Mongolian home furnishing pattern enhancement algorithm was proposed based on the Lαβtransformation and gamma correction method.Firstly,the original color pattern of the RGB space of the Mongolian home furnishing was converted into Lαβcolor space.Then the information entropy and the details of the pattern were enhanced by using the gamma correction method.Finally,the pattern of Lαβcolor space was returned to the RGB space.The results showed that,compared with the CLAHE method,this algorithm not only improved the objective evaluation index of the pattern,enriched the information in the image,but also made the subjective details of the pattern more prominent,the texture clearer,and its enhancement effect better.