基于Retinex和同态滤波的X射线电池图像增强算法

X-Ray Battery Image Enhancement Algorithm Based on Improved Retinex and Homomorphic Filtering

针对工业X射线电池图像对比度低、视觉效果不好、不同区域处理效果差异大等问题,提出一种基于多尺度Retinex和同态滤波的X射线电池图像增强算法.首先利用多尺度Retinex算法中的高斯滤波对照射分量估计,从而得到反射分量.用改进的直方图均衡化方法来处理照射分量,采用改进的巴特沃斯高通滤波器对反射分量的局部细节增强.接着,将照射分量与反射分量按比例融合,最后用改进的同态滤波对图像增强,即可得到增强后的图像.采用客观评价的方法对算法的有效性进行评价,结果表明,所提的算法图像增强效果好且可以看到更多的细节.

Industrial X-ray battery images have shortcomings such as low contrast,obvious noise effect,and poor visual effects.The paper proposes multi-scale Retinex and homomorphic filtering algorithm to enhance the X-ray battery image.Reflection component is obtained by using Gaussian filter in multi-scale Retinex algorithm to estimate illumination component.The improved histogram equalization method is used to process the illumination component to enhance the contrast of the image.The reflection component is enhanced with improved butterworth high-pass filter,the reflected component is fused in the same proportion with the illumination component and then enhanded images can be obtained by the improved homomorphic filtering.The effectiveness of the proposed algorithm is evaluated by objective methods.The results show that the proposed algorithm has a good image enhancement effect and more detailes can be seen by it.