Research Of Industrial X-ray Image Enhancement Algorithm

Non-destructive testing based on X-ray images has now become a widely used detection technology.However,in traditional X-ray imaging technology,the visibility of X-ray images is usually limited due to the differences in the material of the detected object,the complexity of the structure,the superimposition of the castings' structure under investigation and scattered X-rays and the defects of the imaging equipment.It also causes the low contrast and blurry key features of the X-rays image.Even if the image contains rich information,it is difficult to be directly observed by the human eye.Therefore,it is necessary to use digital image enhancement technology to enhance industrial radiographic images,improve the brightness contrast and retain the detailed information of the original image,so as to satisfy the needs of industrial applications.In view of the characteristics of industrial X-ray images,the main work of this paper is as follows:1.An industrial image enhancement based on a simple radiographic scattering model is proposed.In conventional radiography,image visibility is often limited mainly due to the superimposition of the castings' structure under investigation and scattered X-rays.For this problem,an image enhancement method based on a simple radiography scattering model improves the visibility of a single X-ray image is proposed.Based on simple radiographic scattering model and dark channel prior,we introduce the core parts of the algorithm based on boundary constraints and context regularization in detail.Then,according to the characteristics of industrial X-ray image,a new median filter is added in the restoration process,which can effectively suppress the pulse noise component in the X-ray image.At the same time,the guided filter is applied to the transmission image before refining,and the effect of the edge part where the gray value jumps is improved.Through experimental comparison,the improved algorithm on subjective visual effect not only improves the overall brightness contrast of the image,but also makes the structure of object more visible and clearer,the image edge part is more natural,and key parts such as flaws and details can also be well highlighted.2.An industrial image enhancement based on tone mapping and a simple radiographic scattering model is proposed.Considering that most industrial X-ray images are high dynamic range images,tone mapping is introduced into the algorithm.Firstly,an adaptive local tone mapping algorithm is introduced in detail,and according to the characteristics of industrial X-ray image,by adding weighted least squares filter to the algorithm,an adaptive local tone mapping algorithm based on WLS filter is proposed as the preprocessing of the whole algorithm,and the image details are greatly enhanced;Secondly,the CLAHE algorithm is added as the postprocessing of the whole algorithm,which is effective in enhancing the brightness and contrast of the image.Finally,combined the two algorithms with the industrial X-ray image enhancement algorithm based on a simple radiographic scattering model to obtain the final enhanced images.Through experimental comparison,the algorithm not only improves the overall brightness contrast of the image in subjective visual,but also highlights more detailed information,which can make the object structure more visible.