Research On Image Encryption Method Based On Wavelet Fractional Fourier Transform

With the constant development of technology, information transmission, which is abranch of the information processing, becomes a new research topic of informationscience, and the following information secrecy technique is also attached importance toby people. In this paper, supplies a new method for image encryption combining withfractional Fourier transform to address image encryption.As we all know that the fractional Fourier transform is extend of classic Fouriertransform, which is a traditional tool to solve signal processing etc, which has its owntime-frequency properties and diversity. Until now there are many kinds of fractionFourier transforms, such as standard Chirp fractional Fourier transform, multipleparameter fractional Fourier transform, and standard weighted unitary fractional Fouriertransform and so on. In fact, there are two factors to decide fraction Fourier transformdiversity: one of them is choose different kinds of fractional Fourier transformcharacteristic functional, the other factor pays more to divide characteristic value.Wavelet Analysis is a newly developed discipline based on of fractional Fouriertransforms, and is a different developing trend from Fractional Fourier transform. Withthe character of analysis，Wavelet could well show the partial feature of the signal timezoneAs we know, one of the reason why Fourier transform is diversified is that we haveselected different normative characteristic function，so different Fourier transform couldappear when we select them. Hermite-Gaussian function consists of standard orthogonalbasis ofL~2(x) space, which is a group of normative characteristic function. In fact, wecould get new standard orthogonal basis by orthogonal with well expressingHermite-Gaussian function so as to structure new Fractional Fourier transform.The paper’s target is to encrypt the image by using this new wavelet fraction. In thisway, encryption key will not only limit to the characteristic value, but put a full use tocharacteristic vector, which adds matrix encryption key again based on of originalencryption of characteristic value in order to make it more difficult for unconcernedpeople or illegal ones to crack the encryption image. The last image imitation shows thatthis kind of encryption method is robust, reliable and practically valuable.