Abstract: Robustness against geometric distortion is one of the most important problems in image authentication.In this paper，an image watermarking algorithm robust to geometric transformations is proposed，Radon transformation of edge image is used to detect and invert geometric transformations，then a perceptual hash feature detector is proposed to extract significant geometry preserving feature points，finally compare features from two images by developing a modified Hausdorff distance measure.The experimental results show that the algorithm is robust to image processing，such as JPEG compression，common signal processing operations，global as well as geometric transformations，and it is sensitive to content changing manipulations of image data.

Key words: Radon transformation, perceptual hash, modified Hausdorff distance, authentication

摘要： 提高图像抗几何攻击的能力是当前图像认证算法待解决的重点之一。提出了一种抗几何攻击的图像认证算法，该算法利用图像边界的Radon投影变换来实现图像几何失真的矫正，根据感知hash方法提取图像的特征点，并通过修正Hausdorff距离来实现对图像的认证。实验表明，该算法可以抵抗一定程度的JPEG压缩、叠加噪声等图像处理，也能抵抗旋转、缩放等几何变换，并且对于恶意篡改具有较好的敏感性。

关键词: Radon投影变换, 感知hash, 修正Hausdorff距离, 认证算法