Inspection of oil and gas and chemical materials pipelines is one of the most important problems in the material transfer industry. Lack of corrosion detection in these pipes causes irreparable human and financial accidents. Due to the under loading of these pipes, kind of tests are needed that can be achieved to location of corrosion without any damage to the tube. One of these destructive tests is industrial radiography, which is used in this case. In this method, the sensitive plate is closed around the tube and transmitted X-rays or gamma-ray and the location of corrosion are recorded the sensitive plate. Due to the scattering, the resulted images need to be processed in order to clear the areas of defect and corrosion regions. Interpreter and radiographers need images with better quality and contrast to detect areas of defects. Image processing algorithms can help. The complex wavelet algorithm, using image decomposition into various phase and domain components and removing some of these components, can be effective in improving image contrast. In this research, the complex wavelet method was used to remove the background and clarity of images. The results show that this method has acceptable performance in improving the contrast of radiographic images in corrosion areas of ammonia tubes. And the evaluation of industrial radiography also confirms the improvement of images with a high percentage.