The aircraft structures are made of aluminium alloys because of its various advantages, including ease of manufacture, high tolerance and ease of maintenance. Corrosions and cracks are often found in high-strength aluminium alloys. The industrial radiographic testing method and digital radiography are two most important tools for detecting different kinds of defects in aluminium structures. However, because of greater sensitivity and dynamic range of phosphor plates in computed radiography than in film, digital radiography can produce clear and high-contrast images, but digital radiography images appear foggy. In this study, a dehazing algorithm is implemented for the digital radiography images of airplane parts to remove fog. The used dehazing algorithm is based on the dark channel prior and it is based on the statistics of outdoor haze-free images. In most of the local regions of the radiography images, some pixels very often have very low intensity in at least one colour (RGB: red, green, blue) channel which are called dark pixels. In hazy radiography images, the intensity of these dark pixels in that channel is mainly contributed by scattering. Therefore, these dark pixels can directly provide an accurate estimation of the haze transmission and combining a haze imaging model and a soft matting interpolation method can be recovered a high-quality haze free in the radiography image and produce a good depth map and the defects. The results show that the fog-removed images have better contrast and the shapes of defects are very clear. In addition, some invisible cracks in the digital images can be seen in the defogged image.