基于特征点相对位置的指纹图像感知哈希算法

Fingerprint Image Perceptual Hashing Algorithm Based on Relative Position of Feature Points

生物指纹图像在身份认证、安全支付、访问控制等方面有着广泛的应用,而指纹图像感知哈希技术在指纹图像检索、生物特征识别与保护中具有重要作用。提出并实现了一种基于特征点相对位置的指纹图像感知哈希算法。首先利用短时傅里叶变换对生物指纹图像进行增强预处理;然后根据形态学变换提取指纹的特征点和中心核,在密钥的控制下随机选取特征点和中心核形成三角形,并计算各个三角形的形心;最后将形心坐标编码为感知哈希值,通过计算感知哈希值之间的欧几里得距离来判断指纹图像的相似度。实验结果表明,算法能够较好地满足感知哈希的各项性质要求。

This paper proposes a fingerprint image perception hashing algorithm based on the relative position of fingerprint feature points.Firstly,the short-time Fourier transform is applied to enhance the biometric fingerprint image,and then the feature points and the center core of the fingerprint are extracted according to the morphological transform.Under the control of the encryption key,the feature points and the center core are randomly selected to form triangles,and the centroid of each triangle is calculated.Finally,the centroid coordinates are encoded as perceptual hashing value,and the Euclidean distance between perceptual hashing value is calculated to measure the similarity of fingerprint images.The experimental results show that the algorithm can satisfy the requirements of perceptual hashing.