An Efficient and Secure Approach to Image Encryption and Transmission

ABSTRACT- Many challenging computer vision and image processing problems are being solved using compressive sensing with sparse representation algorithms. Compressive sensing integrates sensing and compressing. All natural images have sparse representation in certain basis. Example: Wavelets, DCT etc. This property has been used in this paper to provide compression with sparse representation. An increasing amount of information such as text, image, audio and video are being transmitted over the Internet. To have a lesser communication bandwidth and storage, compression is required. To ensure security and user privacy of the sensed images, encryption becomes an important requirement. The main objective of the work is to develop secure image-encryption techniques, by means of combining the two techniques: encryption and compression. In this technique, a wavelet transform is used to decompose the image and decorrelate its pixels into approximation and detail components. The approximation component is encrypted using chaos based encryption algorithm which provides a good diffusion and confusion properties. The detail components are compressed using Wavelet Transform. This proposed algorithm can provide a high security level for image cryptosystems to provide an efficient and secure approach to real-time image encryption and transmission.

Index Terms- Chaos, Encryption, Wavelet Transform, Compression.

I INTRODUCTION

Today more and more information has been transmitted over the internet. The information is not only text, but also audio, image and other multimedia. To confirm the confidentiality, integrity and usability when transmitted on the network becomes a research hotspot in recent years. The more extensively images will be used, the more important their security, for example it is important to protect the diagrams of important data of bank, military.

When transmitting data over insecure limited bandwidth- channels, compression of data and encryption is always necessary. An encryption algorithm converts the data from comprehensible to incomprehensible structure, thus making the encrypted data difficult to compress using any of the classical compression algorithms, which relies on intelligence embedded in the data. Hence, traditionally the encryption always follows compression. This paper focuses on compressive sensing.

Compressive sensing is a new emerging area, which has gained lot of interest among the signal processing community of late, due to its ability to reconstruct the sparse signal from a relatively smaller sampling set. The protection of information throughout the life span of the information, from the initial creation of the information on through to the final disposal of the information is must.

II LITERATURE SURVEY

A Virtual Image Cryptosystem based upon Vector Quantization focus on to protect the image data. This new cryptosystem can confuse illegal users; compression rate of this cryptosystem will be lost [1].Efficient Quad tree Coding of Images and Video to decompose an image into separate spatial regions to adaptively identify the type of quantizer used in various regions of an image can be well suited for the performance of low bit-rate video coding [2].A novel scheme for the design and analysis of a practical lossless image , where the image data undergoes stream-cipher based encryption before compression and also provides an efficient way to compress encrypted images is through Slepian-Wolf Coding has a disadvantage of increasing the transmission delay i) source coding loss (ii) image coding loss [3].Lossy Compression of Encrypted Image by Compressive Sensing Technique to achieve compression of the encrypted image data based on compressive sensing technique which have a computational complexity[4].An Efficient Chaos-Based Feedback Stream Cipher (ECBFSC) for Image Encryption and Decryption for real-time image encryption and transmission from the cryptographic viewpoint have greater cost [5]. Lossy Compression and Iterative Reconstruction for Encrypted Image provides a novel scheme for lossy compression of an encrypted image with flexible compression ratio which has a disadvantage that the security of encryption used here is weaker [6].Commutative Encryption and Watermarking in Video Compression based on H.264/AVC codec has the drawback is that it distorts media greatly and thus enlarges the attacks difficulty greatly [7].

To overcome this type of difficulties, a new way is proposed. In this paper, the wavelet transform is used to decompose and correlate it pixels into approximate and detailed components (horizontal, vertical, diagnol). The approximate component is encrypted using chaos based encryption algorithm and the detailed components are decomposed by means of wavelet transform.

III SYSTEM OVERVIEW

3.1 Characteristics of an Image Cryptosystem

A good information secure system is able to not only protect confidential messages in the text form, but also in image form.

1. Privacy: an unauthorized user cannot disclose a message.
2. Integrity: an unauthorized user cannot modify or corrupt a message.
3. Availability: messages are made available to authorized users faithfully.

Besides the above characteristics, the image security also requires the following characteristics:

1. The encryption system should be computationally Secure. It must require an extremely long computation time to break, for example. Unauthorized users should not be able to read privileged images.
2. Encryption and decryption should be fast enough not to degrade performance of the system. The encryption and decryption algorithms must be simple enough to be done by users with a personal computer.
3. The security mechanism should be as broad as possible. It must be widely acceptable to design a cryptosystem like a commercial product.
4. The security mechanism should be flexible.
5.There should not be a large expansion of the encrypted image data.

3.2 Chaos and Cryptography

The similar relationship that exist between chaos and cryptography makes chaos based cryptographic algorithms as a natural candidate for secure communication and cryptography chaos based encryption techniques are taken to be good for practical use as these techniques provide a good combination of speed, high security, complexity, reasonable computational overheads and computational power etc.

3.3 Confusion and Diffusion Properties

Chaos encryption refers to a condition or place of great disorder or randomness.

Confusion means that the transformation complicates the dependence of the statistics of the output on the statistics of the input.

Diffusion means the spreading out of local information on the entire image, this means that a single input spreads out over many possible outputs.

3.4 Proposed System

To have a lesser communication bandwidth and storage compression is required. To provide security and user privacy encryption is used by means of Selective encryption of an image combined with compression. First, here the image is taken and just decomposes it into two by means of DWT (i) Approximate part and (ii) Detail part. The approximate part is encrypted and detailed part is compressed and combine to get a coded image in Fig1 (a).Here two algorithms are used.

- Encryption-Chaos Based Image Encryption Algorithm.
- Compression-Wavelet Compression.

3.4.1 Chaos based encryption algorithm

Chaos presents two general principles which have been already used for long time in classical encryption algorithm these properties are confusion and diffusion. A commonly used definition says that, for a forcible system to be classified as chaotic, it must have the following characteristics:

- It must be sensitive to initial conditions
- It must be topologically mixing and
- Its periodic orbits must be dense.

Chaotic maps present many desired cryptographic qualities such as simplicity of implementation that leads to high encryption rates, and excellent security.

The block diagram of the algorithm is depicted in Fig 1(a).

illustration not visible in this excerpt

Fig: 1(a) .The process of encryption and compression

The process of decryption and decompression is the reverse of previous one in which we have use IDWT and the diagram is depicted in Fig1 (b).

illustration not visible in this excerpt

Fig: 1(b).The process of decryption and decompression.

3.4.2 Wavelet Compression

Compression is the process in which the external force that tends to crush a material, squeezing its particles closer and shortening the dimension in the direction of its action. Wavelets are a class of functions used to localize a given function in both space and scaling and also a small wave with its energy concentrated in time. The general image-compression technique used here is the wavelet transform, which is designed to differentiate between the visually known information and unknown information. The transform is also meant to reduce the statistical dependence between coefficients so that the source coding will be more efficient.

Every image will be transformed in each level of decomposition to a one low information image and three details image in parallel, perpendicular and diagonal axis image, also the low also the low information image can be dissolved into other four images. These approaches of decomposition process provide us a number of unrealizable features in the original image, which appear in their surface after the application of transformation.

illustration not visible in this excerpt

Fig: 2. splitting of subband into next higher Level four subbands.

There are different types of wavelet transforms is present, of that here the discrete Wavelet Transform (DWT) is used when a signal is being sampled. In a discrete wavelet transform, an image can be evaluated by passing it through an analysis filter bank, which subsists of a low-pass and high-pass filter at each decomposition stage. When a signal passes through this filter, it is split into low –pass and high-pass filter in which low-pass filter fits to averaging operation, which extracts the coarse information of the signal. The high-pass filter, which matches to a differencing operation, extracts the detail information of the signal. The product of the filtering operations is decimated by 2.

A two-dimensional transform can be done by performing two separate one-dimensional transforms. First, the image is filtered along the x -dimension using low pass and high pass analysis filters and decimated by 2. Low pass filtered coefficients are gathered on the left part and high pass filter on the right, because of it the total size of the transformed image is same as the original image. Then, it is succeed by filtering the subimage along the y -dimension and decimated by 2.

Finally, the image has been dividing into four bands expressed by LL, HL, LH, and HH, after one-level decomposition. The LL band is again exposed to the same procedure. The restoration of the image can be carried out by reversing the above procedure, and it is repeated until the image is fully reconstructed.

IV ENCRYPTION AND COMPRESSION STAGE

The encryption and compression stage will consist of the following. They are

- Input Image
- DWT
- Encryption
- Compression
- Coded Image
- Performance Analysis

illustration not visible in this excerpt

V RESULTS AND DISCUSSIONS

Here the simulations results were done by means of using MATLAB version 9.0 because it has powerful graphic tools allows easy matrix manipulation, plotting of functions and data, implementation of algorithms, creation of user interfaces etc...

5.1 Input Image

The image which is of equal dimensions is given as input and any type of image specification is used. Here the input image is converted into grayscale image in Fig: 5.1.

illustration not visible in this excerpt

Fig: 5.1 Conversion of Input Image into Grayscale Image

illustration not visible in this excerpt

5.2 DWT

The converted grayscale image is decomposing by means of haar transform and has four parts; one approximate and three detailed components are shown.

illustration not visible in this excerpt

Fig: 5.2 Decomposition of an Input Image

5.3 Encryption

The approximate part is encrypted by means of chaos which have a property of good random type of shuffling to encrypt the original image.

illustration not visible in this excerpt

Fig: 5.3 Encryption of an approximate component

5.4 Compressed Image

The detailed part will be selected among three detailed components (horizontal, vertical, diagonal) and compressed by means of wavelet compression

illustration not visible in this excerpt

Fig: 5.4 Compression of detailed component

5.5 Coded Image

Here the encrypted and the compressed image is combined to get a coded image which is depicted in Fig: 7

illustration not visible in this excerpt

Fig: 7 Formation of coded image

5.6 Performance Analysis

Simulation was done with MATLAB using version 9.0, here the images are taken and divided into approximate and detailed components, in which approximate component is encrypted by chaos because of the property of confusion and diffusion, and among the three detailed component the part which is similar to original image i.e., vertical component is taken and compressed by means of wavelet compression. Here the correlation and compression ratio of the following test images is calculated and value is shown in the table 5.7.

illustration not visible in this excerpt

Fig: 5.6 Test images

illustration not visible in this excerpt

Table: 5.7 Correlation and compression ratios of test images

VI CONCLUSION AND FUTURE ENHANCEMENTS

In this paper, a method for the selective encryption of an image with compression uses chaos based encryption and wavelet compression, which results in a significant reduction in encryption time and provides a greater security. Future work may be extended by using of longer key for encryption to further enhance more security.

REFERENCES

[1].T. S. Chen, C. C. Chang, and M. S. Hwang, “A virtual image cryptosystem based upon vector quantization,” IEEE Transactions on Image Processing, vol. 7, no. 10, pp. 1485–1488, 1998.

[2].Gary J. Sullivan and Richard L. Baker ,, “Efficient Quadtree Coding of Images and Video”, in IEEE Transactions On Image Processing, VOL.:3, NO. 3, May 1994; page(s):327-331

[3].MonicaD.N, Shunmuganathan, K.L., Suresh, L.P.,, “Efficient Compression Of Encrypted Grayscale Images,” International Conference on Signal Processing, Communication, Computing and Networking Technologies (ICSCCN) 2011, Page(s): 564 – 568.

[4].A.Anil Kumar and Anamitra Makur, “Lossy Compression Of Encrypted Image By Compressive Sensing Technique” IEEE, Technical Conference on2009,Page(s):1 – 5

[5].H.Hossam El-din, H. M. Kalash, and O. S. Farag Allah, “An efficient chaos-based feedback stream cipher (ECBFSC) for image encryption and decryption,” Informatics, vol. 31, no. 1,pp. 121–129, 2007.

[6].Gary J. Sullivan and Richard L. Baker, “Efficient Quadtree Coding of Images and Video,” IEEE Transactions On Image Processing, VOL.:3, NO. 3, May 1994; page(s):327-331.

[7].J.Fridrich, “Symmetric ciphers based on two-dimensional chaotic maps,” International Journal of Bifurcation and Chaos in Applied Sciences and Engineering, vol. 8, no. 6, pp. 1259–1284, 1998.

[8].H.Hossam El-din, H. M. Kalash, and O. S. Farag Allah, “An efficient chaos-based feedback stream cipher (ECBFSC) for image encryption and decryption,” Informatica, vol. 31, no. 1,pp. 121–129, 2007.

[9].T. Kunkelmann, “Applying encryption to video communication,” in Proceedings of the Multimedia and Security Workshopat (ACMMultimedia ’98), pp. 41–47, England, UK, September1998.

[10].X. Li, J. Knipe, and H. Cheng, “Image compression and encryption using tree structures,” Pattern Recognition Letters, vol.18, no. 11-13, pp. 1253–1259, 1997.

[11]. Y. Mao and G. Chen, “Chaos based image encryption,” in Handbook of Computational Geometry for Pattern Recognition, Computer Vision, Neural Computing and Robotics, E. B. Corrochano,Ed., Springer, Heidelberg, Germany, 2004.

[12].S. Roche, J.-L. Dugelay, and R. Molva, “Multi-resolution access control algorithm based on fractal coding,” in Proceedings of the IEEE International Conference on Image processing (ICIP’96 ), pp. 235–238, IEEE Signal Processing Society, Lausanne, Switzerland, September 1996.

[13].T. S. Chen, C. C. Chang, and M. S. Hwang, “A virtual image cryptosystem based upon vector quantization,” IEEE Transactions on Image Processing, vol. 7, no. 10, pp. 1485–1488, 1998.

[14].J. Scharinger, “Fast encryption of image data using chaotic Kolmogorov flows,” Journal of Electronic Imaging, vol. 7, no.2, pp. 318–325, 1998.

[15].Y. Mao and G. Chen, “Chaos based image encryption,” in Handbook of Computational Geometry for Pattern Recognition,Computer Vision, Neural Computing and Robotics, E. B. Corrochano,Ed., Springer, Heidelberg, Germany, 2004.