Enhancement algorithm of X-Ray high dynamic range image

Abstract

The invention provides an enhancement algorithm of an X-Ray high dynamic range image. The enhancement algorithm comprises the following steps: representing an original image S as a product of an illumination image L and a reflectivity image R; obtaining a logarithmic processing image of the original image S; normalizing the obtained logarithmic processing image; obtaining an image baselog of mean filtering of the smooth region by using a guide filtering algorithm; mapping the image base_log to an image value domain of the logarithmic processing image to obtain a mapped image baseys_log, and subtracting the mapped image baseys_log from the logarithmic processing image to obtain an image det ail_log; respectively distributing coefficients for the image base_log and the image det ail_log; adding the image base_log and the image det ail_log which are distributed with the coefficients to obtain an image r_log, and mapping the r_log to an image value domain of the logarithmic processing image; and obtaining the enhanced result image. The problem that when X-Ray image detection is utilized, the traditional image enhancement algorithm causes loss of original image data or synchronously increases image noise when meeting a detection scene in which the detection features are not greatly different from substances around the features, is solved, the original image data are reserved, and local features can also be enhanced.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an enhancement algorithm for X-Ray high dynamic range images. Background technique [0002] The imaging principle of X-Ray is to emit X-rays from the ray source. Because the density of the penetrated object itself is different, the amount of X-rays absorbed is also different. Objects with high density absorb more X-rays, and the X-rays that pass through Less; objects with low density absorb less X-rays and transmit more, so X-Ray can use this phenomenon of differential absorption to present the image of the object on the imager. [0003] When actually using X-Ray images for detection, you will encounter objects with very small local density differences, such as detecting small bubbles in glue sticks. For the X-Ray images, see figure 1 ; or an object with a large local density difference, such as the innermost positive electrode and the outermost positive electrode of a ...

AI Technical Summary

Problems solved by technology

[0004] The invention discloses an enhancement algorithm for X-Ray high dynamic range images, which solves the problem that when using X-Ray image detection, when encountering a detection scene where the detection feature is not much different from the surrounding substances of the feature, the original image data caused by the traditional image enhancement algorithm The problem of a large number of loss or synchronous increase of image noise can not only retain the original image data but also enhance local features at the same time

Images