DCGAN-based spectral imagery secure retrieval method

Abstract

The invention discloses a DCGAN (Deep Convolutional Generative Adversarial Network)-based spectral imagery secure retrieval method, and belongs to the field of spectral imageries. According to the method, the features of a spectral imagery are highly expressed by utilizing a DCGAN; and a new encrypted domain spectral imagery secure retrieval method is proposed. Firstly the deep spectral-spatial features of the spectral imagery are jointly extracted by utilizing the DCGAN, and the contents of the spectral imagery are accurately represented; in order to ensure the security in a remote sensing image retrieval process, the deep features are encrypted by adopting a Min-Hash method based on a criterion that the similarity of the encrypted features is unchanged, thereby protecting the deep features; and finally under the non-decryption condition, Jaccard similarity distance measurement is performed on image features directly by comparing the number of same Min-Hash values, and images similar to a query image are returned. Therefore, the information security is ensured while the retrieval is realized.

Description

technical field [0001] The invention takes the imaging spectral image as the research object, uses the DCGAN deep network to realize the high-level expression of the characteristics of the imaging spectral image, and proposes a new encrypted domain imaging spectral image security retrieval method. First, the deep convolutional generative adversarial network (Deep Convolutional Generative Adversarial Network, DCGAN) is used to jointly extract the deep spectral-spatial features of the imaging spectral image, and accurately characterize the content of the imaging spectral image; at the same time, in order to ensure the security of the remote sensing image retrieval process, based on Based on the principle that the feature similarity remains unchanged after encryption, the minimum hash (Min-Hash) method is used to encrypt the deep feature to realize the protection of the deep feature; finally, without decryption, by comparing the number of equal minimum hash values The Jaccard sim...

AI Technical Summary

Problems solved by technology

However, due to the hyperspectral characteristics of imaging spectral image data and limited human eye recognition ability, it is difficult to obtain a large number of supervised calibration sample data in practice, which also restricts the existing deep learning technology in imaging spectral image feature extraction, classification and recognition. Applied Research and Extension

Images