基于生物特征的随机序列发生器

Biometric-Based Random Sequence Generator

利用生物特征本身内在的独特性、不可预测性和不可重复性等随机特性和外部随机噪声(光照、损伤等)给出了基于生物特征的随机序列发生器.以虹膜为例,通过Canny算子对虹膜图像进行边缘检测,得到二值化图像,提取值为“1”的坐标(i,j),利用f(i,j)=cos(i×j),得到[-1,1]之间的随机序列.使用标准随机统计测验FIPS 140-1 tests检验产生的随机序列.结果表明:和其他随机序列发生器相比,该发生器不仅具有优良的随机性,而且实现简单、方便,提供了真随机序列的一种产生方法.

The randomness of biometric features including uniqueness, unpredictability, irreproducibility and external noises such as lighting and trauma are all used to develop biometricbased random sequence generators. The generator is discussed in detail with iris as an example. The Canny operator is used to do boundary test with the iris images, then the binaryzed iris images are gotten and the pixel coordinates values （i, j ） are fetched as ＂1＂ and the random sequence over the interval [ - 1,1 ] can be acquired by the function f（ i ,j ） = cos（ i x j ）, Then, the FIPS 140-1 tests are carried out for the sequences so as to verify the outstanding randomness. The results show that the generator is not only good at randomness, but also simple and easy to implement, in comparison to other similar generators, thus providing a new way to generate true random sequence.