基于双边滤波和Retinex算法的矿井图像增强方法

Image enhancement method of mine based on bilateral filtering and Retinex algorithm

  • 摘要: 针对常用的双边滤波算法易造成图像细节丢失及Retinex算法在光照变化剧烈时易出现光晕伪影现象等缺点,提出了一种基于双边滤波和多尺度Retinex算法的图像增强方法。 该方法首先对图像进行小波分解,获得图像高频和低频系数;然后采用多尺度Retinex算法和双边滤波结合的方案对图像低频系数进行处理,采用软阈值滤波算法对图像高频系数进行处理;最后通过离散小波反变换得到增强后的空域图像,并对其局部对比度进行自适应增强处理。实验结果表明,该方法可以有效改善图像颜色失真情况,细节丰富,对比度强,为图像后续的特征提取奠定了基础。

     

    Abstract: In view of defects that common bilateral filtering algorithm is easy to loss image details and Retinex algorithm is easy to appear phenomenon of halo artifacts when light and darkness changes violently, an image enhancement method based on bilateral filtering and multi-scale Retinex algorithm was proposed. Firstly, the image is decomposed to obtain high frequency and low frequency coefficients; then, the low frequency coefficients of the image are processed by the multi scale Rentinex algorithm and the bilateral filtering scheme, and the high frequency coefficients of the image are processed by soft threshold filtering algorithm; finally, the enhanced spatial domain image is obtained by discrete wavelet inverse transform, and the local contrast is enhanced by adaptive enhancement. The experimental results show that the method can effectively improve the image color distortion with rich details and strong contrast, which lays the foundation for the subsequent features extraction.

     

/

返回文章
返回