Defect imaging algorithms play an important role in Lamb waves based researches of nondestructive testing (NDT) and structural health monitoring (SHM). In classical algorithms, the location or distribution of defects is visualized through mapping the amplitude or phase information of signals gotten by multiple inspection pairs from the time domain to every discrete spatial grid of plates. It is time-consuming in the detection of plates with large size and many transducers. Transforming the defect imaging problem into a scattering source search problem, an intelligent defect localization algorithm was proposed for NDT and SHM with the Lamb waves and sparse array. In the algorithm, the elliptic trajectory-dependent individuals of every inspection pair were extracted first, then the defect position was identified by analyzing the distribution of individuals these located at the intersection of multiply elliptic trajectories. Considering the fuzzy and diversity characteristics in the detection of defects, a fuzzy control parameter and an adaptive individual updating strategy based on the k-means algorithm were introduced to ensure the robustness of the algorithm. The effectiveness of the proposed algorithm was verified by numerical models and experiments. The influences of the fuzzy control parameter and the individual updating strategy on the performance of the algorithm were analyzed furthermore.