Similarity and Ranking Preserving Deep Hashing for image Retrieval

Hasit codes based on deep learning can effectively learn image features. For supervised deep learning methods, the label information of the image OM be used to farther learn the semantic Information of the image. However, the current supervised deep learning methods often use I and 0 (or -1) to represent the sindlariry of two images In fact, these two extreme values do not fully reflect the similarity between bnages. Thus, we proposed a novel dinibuity and ranking preserving deep hashing method (SRPDH). In order to enrich and more comprehensively reflect the semantic information between Merges, we refine the single-label information into mold -label Information, and use Jaceard coefficient model to calculate the sbnflarity between label information. In the loss hmction model, we use the cress entropy model and consider the loss caused by the bbuiry quantization of the network output- The experimental results show that our method can farther *improve the mean average precision (MAP) of image retrieval compared with the existing methods,