Abstract: Radiograph testing based weld defect recognition has been widely used in nondestructive testing of the weld defect. The main steps of the recognition are as follows. Firstly, high-resolution video camera is used to capture the tested RT film, following the image denoising and enhancement preprocessing operations, then pattern recognition and other means are used to identify the main defects in the film. Image enhancement is one of the very useful image processing tools. The purpose of image enhancement is to strengthen the contrast between the defects and background, in order to facilitate the subsequent segmentation and defect pattern recognition. In this paper, we propose a weld image enhancement algorithm based on the Local Binary Pattern(LBP), and apply it into the weld defect image recognition system. Experiment result shows that, after such image enhancement, the defects in the image become clearer and could be recognized better than without enhancement.