A novel hashing scheme via image feature map and 2D PCA

Hashing scheme is a high-efficiency technique for processing massive images. Two critical metrics of the hashing are discrimination and robustness, but most schemes do not get satisfied classification performance between them. This paper proposes novel via image feature map 2D PCA. First, proposed extracts local phase quantization (LPQ) features in frequency domain ternary pattern (LTP) spatial domain, combines them to construct an map. Second, conducts dimension reduction PCA learning from Last, learned compressed generate hash sequence. Performances tested on open datasets. The results demonstrate that can make good balance robustness. In addition, copy detection both superior those some famous schemes.