Research On Design Of Two-level QR Codes With Xor-based Visual Cryptography Scheme

As a typical two-dimensional machine recognition code,QR code has many characteristics including large storage capacity,fast recognition speed and strong error correction capability,which promotes its widespread usage in applications such as logistics transportation,product marketing and traceability.Two-level QR code is a special symbol image developed from QR code specification,where the QR code is utilized as an carrier to store secret information without affecting the readability of its initial information.Therefore,two-level QR code becomes an excellent choice to efficiently transmit secret or sensitive information via public channels,and promises a high application value in sensitive information transmission and military intelligence communications.Combined with XOR-based visual cryptography scheme,this paper studies the design of two-level QR code with low computational complexity of secret recovery.According to the universal standard specification of QR code,allowable information differences between encoding and decoding processes are deeply investigated.And with the allowable differences,XOR-based visual cryptography scheme and QR code can be combined.On this basis,a variety of two-level QR codes are implemented by designing different sharing algorithms,which are suitable for different application scenarios.1.To guarantee the decoding and decryption capability of two-level QR code,a design model of two-level QR code with XOR-based visual cryptography scheme is proposed.By investigating the encoding and decoding mechanism of QR code,the allowable differences of QR code encoding and decoding processes are analyzed.Then methods of combining XOR-based visual cryptography scheme with QR code are given.Further,according to the characteristics of different methods and their respective influence on two-level QR code performance,requirements of sharing algorithm for each method is discussed.2.Regarding the characteristic of module recognition,the effect of pixel distribution on module recognition result is analyzed.Module recognition unit is designed while the pixel expansion is limited,enhancing the recognition efficiency of two-level QR code.Additionally,an algorithm with perfect recovery performance is presented by using hamming distance of vector space.Therefore,the influence of partial pixel distortion on the accuracy of secret recovery is reduced.Experimental results show that the proposed two-level QR code has higher recognition and secret recovery capabilities with the pixel expansion of 9.3.Regarding the characteristic of encoding redundancy,secret information can be stored in two-level QR code for the XOR operation of Reed-Solomon codes satisfy the closeness feature.Aimed at the size limitation of secret sharing area in two-level QR code,an algorithm without pixel expansion is proposed by simplifying the set of minimal qualified subsets and transforming the deterministic sharing model into a probabilistic one.Therefore,the size of secret image that can be shared reaches the largest.Experimental results show that the proposed two-level QR code improves the recovery performance and the relative difference can reach up to 1.4.Regarding the characteristic of error correction,a condition of single subset sharing is proposed to achieve higher security.It solves the security problem of caused by similar cover images by preventing forbidden subsets from reconstructing the secret.On this basis,the algorithm under general access structure is presented.To solve the problem that single subset sharing algorithm is inefficient for general access structure,multi-subset sharing strategy is proposed to make shares reusable.Experimental results show that the proposed two-level QR code requires for lower execution cost and less storage overhead.