基于自适应格网划分的遥感影像感知哈希认证算法

针对遥感影像大数据量和信息分布不均匀的特点,提出了一种基于自适应格网划分的遥感影像感知哈希算法,并将其用于遥感影像的完整性认证。该算法基于自适应策略对信息丰富的影像区域进行粒度更细的完整性认证,首先对原始影像进行波段融合,然后根据信息熵对波段融合结果进行自适应的格网划分,最后提取格网单元的局部特征构造感知哈希序列。影像接收端通过收到的感知哈希序列实现影像的认证。实验表明,本文算法能够识别遥感影像局部细节篡改,并对保持内容不变的操作具有鲁棒性,实现了遥感影像基于内容的完整性认证。

An Adaptive Grid Partition Based Perceptual Hash Algorithm for Remote Sensing Image Authentication

In the light of the big data characteristics of remote sensing images,aperceptual Hashing al-gorithm based on adaptive grid division for remote sensing image authentication is proposed.Thismethod,based on adaptive strategy,can authenticate information intensive regions at a fine-grainedlevel.First,the bands of the remote sensing image are fused.The fusion image is adaptively dividedinto grid units based on information entropy.Then the local features of each grid unit are extracted togenerate perceptual Hash value.Authentication verifies the content integrity of the remote sensingimage through the received Hash value.Experiments show that the algorithm can detect malicioustampering of remote sensing images,while tolerating legitimate content distortion,thus achievingcontent integrity authentication for remote sensing image.