Based on the discrete wavelet transform, an algorithm is proposed to enhance the tiny and low-contrast defects in the ultrasonic testing images of the laser deposited TA15 alloys, in which the Penalized soft threshold is used to denoise the high-frequency detail information of the defects and the Gamma transform is used to enhance the contrast of the low-frequency information of the defects. Both the information entropy and the contrast are taken into account to test multiple groups of samples. The research results show that, as for the laser-deposited parts with a unique rapid solidification structure, the proposed algorithm has a better ability of weak defect enhancement compared with the Gaussian filtering and the fuzzy set enhancement theories.