Adaptive Enhancement of High Greyscale Weld Images Based on the New Pixel Self-transformation Method

Abstract

X-ray nondestructive testing technology is widely used in the weld detection of oil and gas pipeline industry. The industrial welding manufacturing field produces 12-bit and higher high grey scale images that cannot be displayed directly on an 8-bit display. In this paper, we propose an adaptive pixel self-transformation method for high greyscale weld images. The method takes into account the characteristics of weld images and human perception. Firstly, a high dynamic range tone mapping model is designed, which converts the original RAW data into optical observable data. Secondly, in order to solve the problem that the traditional pixel self-transformation method can only adapt to the input with 256 levels of greyscale and cannot process the high-bit image data. An adaptive high-bit pixel self-transformation is proposed. Finally, the method of this paper was tested on 12-bit X-ray films. The experimental results show that the designed method, which improves the recognizability of weld images and can effectively obtain weld images with outstanding details, is an efficient weld feature enhancement algorithm.