X-ray image enhancement based on gradient domain and illumination image estimation for complex castings

Abstract When X-ray is used to detect a workpiece with a complex structure, the details will be lost due to overexposure or underexposure in some areas which is uneven thickness. To solve this problem, based on gradient field compression, we propose an algorithm to enhance the X-ray image through illumination image estimation enhancement. In gradient domain, we design a new attenuation function through scale the attenuation factor in each layer of the Laplace pyramid, then make the attenuation function a gamma transform to make the information of the workpiece in the compressed image more complete, and then, we use an illumination image estimation method(Lime) to remove the invalid information of the compressed image with the background. Finally, the details of the workpiece are refined by USM sharpening and contrast-limited adaptive histogram equalization (CLAHE). The experimental results show that the proposed method has an obvious enhancement effect on various complex workpieces, and the method is very efficient.