A popular optimization algorithm, the artificial bee colony algorithm (ABC), has attracted great attention over the recent years for its powerful global search ability. However, its slow convergence rate limits its development. In this paper, to further enhance its performance, we first introduced a new concept of a leading group, which includes some individuals with excellent performance, into the traditional ABC. The updated bee then selects one individual from the group to follow, which accelerates the convergence rate of the population. Furthermore, to enable the ABC algorithm to acquire greater opprotunities to search within a larger space, a logistic chaotic operator was introduced into our algorithm to balance its global and local search abilities. The performance of the algorithm proposed is tested on the traditional 12 benchmark functions and an image registration problem. The results reveal that our algorithm provides more acceptable results compared with the other algorithms.