Digital image correlation method is an important optical technique for surface displacement and strain measurement. An approach based on Particle Swarm Optimization algorithm for sub-pixel correlation search is described in this paper. The new Algorithm does not involve reasonable guess of displacement and deformation gradient and the calculation of second-order derivatives of the digital images. Benefiting from the abilities of global optimum and parallelism searching, and compared with genetic algorithm, the new approach can complete the sub-pixel correlation search with high accuracy and less computational consumption. Computer-simulated images are then used to verify this method. The experimental results show that the new approach is a practicable sub-pixel searching method.