为了改善X射线成像对比度低、背景噪声明显、被测工件细节不易显示等缺点, 应用改进的Retinex可变框架模型增强算法, 对X射线图像进行图像增强, 提高了原图像的对比度和像质, 达到了细节区域的最佳视觉效果。理论分析和试验结果表明, 该算法与同态滤波、直方图均衡化等算法相比, 有效地增强了对比度、抑制了背景噪声, 图像所达到的指标比前两种算法要高。

In order to overcome the shortcomings of X-ray imaging such as low contrast, background noise obviously, the lack of the workpiece details, the present studies aimed to apply the enhanced improved Retinex algorithm to deal with the radiographic images, and to improve the contrast and the quality of the original images, and hence the best visual effects of the minutia area were achieved. The theoretical analysis and the result of the experiment indicated that this algorithm could effectively enhance contrast, control the context and noise, as compared with other algorithms, such as Homomorphism Filter and Gray Equalize. After being enhanced, the radiographic images by using this algorithm had better contrast than the above-mentioned other two algorithms.