The detection technology of ultrasonic guided waves has a wide application prospect in structural health monitoring. In this paper, a kind of Lamb waves defect imaging method and its optimization based on envelope amplitude are proposed for detecting aluminum sheet. The sparse PZT array is used as a transmitting and receiving device to generate the excitation signal and acquire the Lamb waves for difference between the health signal and the defect signal, which contains defect position information. An appropriate penalty function is determined introducing the factors containing the position weight vector to highlight the characteristics of the defective pixel. It transforms the improved waveguide image quality problem into a series of solving unconstrained optimization problems，and the augmented objective function can be determined by means of PH multiplier algorithm, which makes the weight vector optimized and improves the image quality. Experimental results show that the positioning error about the algorithm is small, defect imaging of sheet is clear and can be intuitively recognized, and the optimal treatment of guided waves imaging is perfect.