An Enhanced Hybrid Technology for Digital Image Compression Malvika Dixit 1, Harbinder Singh 2 1 M.Tech Student (ECE), 2 Assistant Professor (ECE), Baddi University of Emerging Sciences & Technology, India ABSTRACT: This paper presents discrete approach towards MATLAB implementation of the discrete wavelet transform (DWT) and vector quantization for compression. Rippling analysis provides a lot of precise data regarding signal information than different signal analysis techniques. Compression is one amongst the foremost vital applications of wavelets. A new image compression theme primarily based on discrete wavelet transform is proposed in this paper that provides ample high compression ratios with no considerable degradation of image quality. The effectiveness and robustness of this approach are even employing a set of real pictures. To demonstrate the performance of the projected methodology, a comparison between the projected technique and different common compression techniques applied. Therefore the investigation during this thesis are familiarized towards the an enhanced hybrid technology for digital image compression. Keywords: Wavelet, peak signal to noise ratio, mean square error, lossy and lossless techniques, DWT I. INTRODUCTION: Image compression may be lossy or lossless. Lossless compression is preferred for archival purposes and often for medical imaging, technical drawings, clip art, or comics. Lossy compression methods, especially when used at low bit rates, introduce compression artifacts. Lossy methods are especially suitable for natural images such as photographs in applications where minor loss of fidelity is acceptable to achieve a substantial reduction in bit rate. Methods for lossless image compression are: Methods for lossy compression: Reducing the color space to the most common colors in the image. The selected colors are specified in the color palette in the header of the compressed image. Each pixel just references the index of a color in the color palette : Transform coding: This is the most commonly used method. In particular, a Fourier-related transform such as the Discrete Cosine Transform (DCT) is widely used:n. Ahmed, T. Natarajan and K.R.Rao, "Discrete Cosine Transform," IEEE Trans. Computers, 90-93, Jan. 1974. The DCT is sometimes referred to as "DCT-II" in the context of a family of discrete cosine transforms; e.g. discrete cosine transform. The more recently developed wavelet transform is also used extensively, followed by quantization and entropy coding. Fractal compression : The best image quality at a given bit-rate (or compression rate) is the main goal of image compression, however, there are other important properties of image compression schemes: Scalability: Generally refers to a quality reduction achieved by manipulation of the bitstream or file (without decompression and re-compression). Other names for scalability are progressive coding or embedded bitstreams. Despite its contrary nature, scalability also may be found in lossless codes, usually in form of coarse-to-fine pixel scans. Scalability is especially useful for previewing images while downloading them (e.g., in a web browser) or for providing variable quality access to e.g., databases. There are several types of scalability: Run-length encoding used as default method in PCX and as one of possible in BMP, TGA, TIFF Area image compression DPCM and Predictive Coding Entropy encoding Adaptive dictionary algorithms such as LZW used in GIF and TIFF Deflation used in PNG, MNG, and TIFF Chain codes Quality progressive or layer progressive: The bitstream successively refines the reconstructed image. Resolution progressive: First encode a lower image resolution; then encode the difference to higher resolutions. Component progressive: First encode grey; then color. Region of interest coding. Certain parts of the image are encoded with higher quality than others. This may be ISSN: 2348 8549 www.internationaljournalssrg.org Page 21
combined with scalability (encode these parts first, others later). Meta information. Compressed data may contain information about the image which may be used to categorize, search, or browse images. Such information may include color and texture statistics, small preview images, and author or copyright information. Processing power. Compression algorithms require different amounts of processing power to encode and decode. Some high compression algorithms require high processing power. The quality of a compression method often is measured by the Peak signal-to-noise ratio. It measures the amount of noise introduced through a lossy compression of the image, however, the subjective judgment of the viewer also is regarded as an important measure, perhaps, being the most important measure. In information technology, "lossy" compression is the class of data encoding methods that uses inexact approximations (or partial data discarding) for representing the content that has been encoded. Such compression techniques are used to reduce the amount of data that would otherwise be needed to store, handle, and/or transmit the represented content. The different versions of the photo of the cat at the right demonstrate how the approximation of an image becomes progressively coarser as more details of the data that made up the original image are removed. The amount of data reduction possible using lossy compression can often be much more substantial than what is possible with lossless data compression techniques. Using well-designed lossy compression technology, a substantial amount of data reduction is often possible before the result is sufficiently degraded to be noticed by the user. Even when the degree of degradation becomes noticeable, further data reduction may often be desirable for some applications (e.g., to make real-time communication possible through a limited bit-rate channel, to reduce the time needed to transmit the content, or to reduce the necessary storage capacity). Lossy compression is most commonly used to compress multimedia data (audio, video, and still images), especially in applications such as streaming media and internet telephony. By contrast, lossless compression is typically required for text and data files, such as bank records and text articles. In many cases it is advantageous to make a master lossless file that can then be used to produce compressed files for different purposes; for example, a multi-megabyte file can be used at full size to produce a full-page advertisement in a glossy magazine, and a 10 kilobyte lossy copy can be made for a small image on a web page. II. PROPOSED WORK AND ENVIRONMENT USED: In most of the compression technique proposed so far, either quantization or coding transformation is used. This selection may lead to inefficiencies in Compression. A new image compression scheme based on discrete wavelet transform is proposed in this research which provides sufficient high compression ratios with no appreciable degradation of image quality. The effectiveness and robustness of this approach will be justified using a set of real images. To demonstrate the performance of the proposed method, a comparison between the proposed technique and other common compression techniques will be carried out. Therefore, the investigation in this thesis will be oriented towards the design of an Enhanced Hybrid System DWT-VQ. Implementation of Discrete Wavlet Transform is done using matlab: 1. Import the image and rename it as lena and colormap as map. 2. Perform single wavelet decomposition using Haar wavelet transformation. 3. Construct Approximates and coefficients 4. Perform multilevel wavelet decomposition. 5. Reconstruct decomposition Values 6. Reconstruct the image from multilevel decomposition. 7. Recompress the Image 8. Apply vector quantization. 9. Metrics for comparing the quality of compressed images. The Mean Square Error (MSE) and the Peak Signal to Noise Ratio (PSNR) are the two error metrics used to compare image compression quality. The MSE represents the cumulative ISSN: 2348 8549 www.internationaljournalssrg.org Page 22
squared error between the compressed and the original image, whereas PSNR represents a measure of the peak error. The lower the value of MSE, the lower the error. To compute the PSNR, the block first calculates the meansquared error using the following equation: In the previous equation, M and N are the number of rows and columns in the input images, respectively. Then the block computes the PSNR using the following equation: In the previous equation, R is the maximum fluctuation in the input image data type. For example, if the input image has a double-precision floating-point data type, then R is 1. If it has an 8-bit unsigned integer data type, R is 255, etc. Fig. 2. Image and rename A. Import the image and rename it as lena and colormap as map. Image (lena); Colormap (map); III. IMPLEMENTATION RESULTS: SIMULATION RESULT: 1. Implementation of Discrete Wavelet Transform using Matlab Splash Window designed in MATLAB GUI environment B. Perform single wavlet decomposition [ca,ch,cv,cd]=dwt2(lena,'haar'); C. Construct Approximates and coefficients a=upcoef2('a',ca,'haar',1); h=upcoef2('h',ch,'haar',1); v=upcoef2('v',cv,'haar',1); d=upcoef2('d',cd,'haar',1); D.Perform multilevel wavelet decomposition. [c,s]=wavedec2(lena,1,'haar'); Now extract the coefficients [ch,cv,cd]=detcoef2('all',c,s,1); E.Reconstruct decomposition Values h=wrcoef2('h',c,s,'haar',1); Fig. 1 Splash Screen 2. Following steps are performed in Wavelet Transformation button in the GUI created in MATLAB. v=wrcoef2('v',c,s,'haar',1); d=wrcoef2('d',c,s,'haar',1); F.Reconstruct the image from multilevel decomposition. x0=waverec2(c,s,'haar'); G.Recompress The Image ISSN: 2348 8549 www.internationaljournalssrg.org Page 23
[thr,sorh,keepapp]= ddencmp('cmp','wv',lena); [lena0comp,cxc,lxc,perf0,perfl2] =wdencmp('gbl',c,s,'haar',1,thr,sorh,keepapp); H.Plot original Image colormap(map); subplot(1,2,1); image(lena);title('original Image'); Fig 4 : Comparing Compressed and Decompressed Images 4. Calculate PSNR and MSE value for this hybrid transformation and using different threshold values for obtaining decompressed images: Fig. 3 Plot Original Image 3. Plotting Compressed Image colormap(map); axis square subplot(1,2,2); colormap(map); image(lena0comp);title('compressed Image'); Fig. 5 PSNR and MSE with threshold value 2.0000 Now we have implemented the work using different threshold values. In above case threshold generated is 2.0000. As we increased the threshold to 52.000 the quality of decompressed image is: ISSN: 2348 8549 www.internationaljournalssrg.org Page 24
Fig. 6 Decompressed Image generated after increasing the threshold. Fig. 9: Decompressed Image generated with With thr=.2500 The PSNR block computes the peak signal-to-noise ratio, in decibels, between two images. This ratio is often used as a quality measurement between the original and a compressed image. The higher the PSNR, the better the quality of the compressed or reconstructed image. Table 1 : Results obtained with different threshold values Thr MSE PSNR MAXERR L2rat 2.0000 0.5442 50.7733 3.0000 3.0000 Fig. 7 Decompressed Image generated With thr=1.0000 1.0000 0.0800 59.0972 1.5000 1.0000 0.5000 0.0132 66.9399 0.7500 1.000 0.2500 1.737 315.7324 0 1.000 Fig. 8 Decompressed Image generated with thr = 0.5000 Fig.10 Graphical representation ISSN: 2348 8549 www.internationaljournalssrg.org Page 25
In matlab created Thr and MSE arrays Thr =[2.0000 1.000 0.5000 0.2500] MSE=[0.5442 0.0800 0.0132 1.737e-27] The plot( Thr, MSE), xlabel ('Thr'), ylabel ('MSE'),title ('Thr vrs MSE') method will plot MSE and Threshold values.mse is decreasing with increase in Theshold. images. The images are taken with a digital camera. To demonstrate the performance of the proposed method, a comparison between the proposed technique and other common compression techniques has been revealed. From the experimental results it is evident that, the proposed compression technique gives better performance compared to other traditional techniques. Wavelets are better suited to time-limited data and wavelet based compression technique maintains better image quality by reducing errors. V ACKNOWLEDGEMENT I am highly grateful to Mr. Harbinder Singh Assistant Professor for giving me invaluable guidance in the field of image compression and providing me the opportunity to carry out this work further. It was the essential encouragement that enables me to pursue my work in this field. REFERENCES: Fig. 11 MSE is Decreasing with increase in Thr Now to plot graphs for Threshold and PSNR Again two array variables Thr and PSNR are taken. Thr=[2.0000 1.000 0.5000 0.2500] PSNR=[50.7733 59.0972 66.9399 315.7324] The plot(thr,psnr),xlabel('thr'),ylabel('psnr'),title( 'Thr vrs PSNR') method will plot the graph. 1. Bradley J. Erickson,Armando Manduca, Wavelet compression of medical images, Journal of Radiology, vol. 206, pp. 599-607, 1998. 2. Abhishek Kaushik, Maneesha Gupta, Analysis of Image Compression Algorithms,International Journal of Engineering Research and Applications,march-april 2012. 3. Anilkumar Katharotiya, Swati J. Patel Mahesh Goyani, Comparative Analysis between DCT &DWT Techniques of Image Compression,Journal of Information Engineering and Applications,Vol 1, No.2, 2011. 4. Mukesh Mittal, Ruchika Lamba, Image Compression Using Vector Quantization Algorithms: A Review, Department of Electrical & Instrumentation Engineering Thapar University, Patiala, India ; Volume 3, Issue 6, June 2013. 5. Md. Rubaiyat Hasan, Data Compression using Huffman based LZW Encoding Technique,International Journal of Scientific & Engineering Research, Volume 2, Issue 11, 1 ISSN 2229-5518 IJSER, November 2011. Fig. 12 Increasing PSNR with increase in Thr(Threshold) IV CONCLUSION A new image compression scheme based on discrete wavelet transform and vector quantization is proposed in this research which provides sufficient high compression ratios with no appreciable degradation of image quality. The effectiveness and robustness of this approach has been justified using a set of real 6. Rafael C. Gonzalez, Richard E. Woods, Digital Image Processing (2nd edition), NJ:Prentice Hall 1992. 7. Karen Lees, Image compression using Wavelets", Report of M.S. 2002. ISSN: 2348 8549 www.internationaljournalssrg.org Page 26
8. Gonzalez, R.C. and Woods, R.E, Digital Image Processing using MATLAB, Pearson Education, India,2006. 9. Locker Gnome, Real World Application of Image Compression, 2011. Subramanya A, Image Compression Technique, 10. Potentials IEEE, Vol. 20, Issue 1, pp 19-23, Feb-March 2001, David Jeff Jackson & Sidney Joel Hannah, Comparative Analysis of image Compression 11. Techniques, System Theory 1993, Proceedings SSST 93, 25th Southeastern Symposium,pp 513-517, 7 9 March 1993. 12. Hong Zhang, Xiaofei Zhang & Shun Cao, Analysis & Evaluation of Some Image Compression Techniques, High Performance Computing in Asia Pacific 13. Region, 2000 Proceedings, 4th Int. Conference, vol. 2, pp 799-803,14-17 May, 2000 Ming Yang & Nikolaos Bourbakis, An Overview of Lossless Digital Image Compression Techniques, Circuits & Systems, 2005 48th Midwest Symposium ISSN: 2348 8549 www.internationaljournalssrg.org Page 27