This paper presents a scheme for blockboard nondestructive detection. X-ray nondestructive testing system has been used to obtain blockboard x-ray images. In order to recognize defects of blockboard effectively through x-ray images, an image processing method based on wavelet transform and gray-scale mathematical morphology is proposed. Firstly, image noises are reduced by wavelet soft threshold de-noising, and then a new modified mathematical morphology edge detection operator is used to detect edges of the de-noised image. The experiment results show that the method performs well in noise-suppression and edge detection. The proposed nondestructive testing method of blockboard is practical and easy to implement.