In order to realize the fast and accurate matching of large array charge coupled-device (CDD) aerial images, a local multi-feature hashing (LMFH) method is proposed. Firstly, the prediction area is constructed according to the course overlap rate, and the feature points detected in the area are described by multi-feature. Then, the hash functions are learned by tens of thousands of existing aerial images. Finally, the high-dimensional feature description vectors are mapped to compact binary hash codes by the learned hash functions. Fast hashing matching is achieved according to the Hamming distance in the Hamming space. Experiments show that compared to the classical speeded up robust features (SURF) algorithm, accuracy is improved about 10%, meanwhile, the matching time is decreased 0.2s. The proposed LMFH algorithm for aerial images matching is much more efficient.