Remote Sensing Image Anti-Interference Matching Algorithm Based on Deep Hashing

With its efficient storage and computing capabilities, deep hashing algorithms have been widely used in tasks such as remote sensing image matching. However, while this algorithm is convenient and fast, it is easily affected by noise and other interferences, which significantly reduces the remote sensing matching performance of the algorithm. To address the aforementioned shortcomings, we propose a remote sensing image anti-interference matching algorithm based on deep hashing. Guided by class representative hash codes, the algorithm trains the deep hashing network by constructing adversarial samples with sufficient interference capability to enhance its anti-interference ability. Compared to traditional anti-interference algorithms, the algorithm proposed in this paper is more targeted and effective in the field of remote sensing image hashing. The experimental results show that the network trained using the algorithm proposed in this paper can still make correct judgments when facing interfering remote sensing images and has good anti-interference ability.