Lossless Image Compression Using Data Folding Followed By Arithmetic Coding
1 like189 views
This paper presents a novel lossless image compression technique that combines data folding and arithmetic coding to achieve improved compression ratios and reduced bits per pixel. The proposed method involves an initial data folding process followed by entropy coding using arithmetic coding, which shows better results compared to conventional methods. However, this technique faces challenges related to high computational time.
![IOSR Journal of Computer Engineering (IOSR-JCE)
e-ISSN: 2278-0661,p-ISSN: 2278-8727, Volume 17, Issue 2, Ver. III (Mar – Apr. 2015), PP 67-70
www.iosrjournals.org
DOI: 10.9790/0661-17236770 www.iosrjournals.org 67 | Page
Lossless Image Compression Using Data Folding Followed By
Arithmetic Coding
Richa Goyal1
, Sourav Garg2
1,2,
(CSE,ACET/ PTU, Jalandhar India)
Abstract : The paper presents a lossless image compression technique using the hybridization of two different
entropy coding techniques. Initially data folding technique has been applied to the image. A row folding is
applied on the image matrix followed by a column folding. Multiple iterations of this process is applied on the
image. After completing the data folding process another entropy coding technique known as arithmetic coding
has been applied to the resultant image to get better results.
Keywords: Lossless image compression, data folding, arithmetic coding, compression ratio, bits per pixel.
I. Introduction
Compression can be defined as reducing the size of data so that the amount of space required to store
the data becomes less and it takes less time to transmit the data. With the help of compression it becomes easy to
store large files. These files may be data files, images, videos, audios and other multimedia files. For
transmitting a data, compression can be performed either on just the data content or on the entire transmission
unit. There are different types of data content on which compression can be applied.
Image compression can be referred to as reducing the size of the image so as it requires less space for
storage and is easily transmitted. This can be done by reducing the memory size required to store each pixel.
There are several algorithms to perform image compression. These algorithms can be categorized as lossy and
lossless compression techniques. Lossy compression causes loss of information after compression. When the
compressed data is decomposed to its original version then both are not same but close. It is an irreversible
process. In lossless compression the compressed data can be easily decomposed to its original version. The loss
of data is very less. It is a reversible process.
Image compression techniques reduce the number of bits required to represent an image by taking
advantage of these redundancies. An inverse process called decompression (decoding) is applied to the
compressed data to get there constructed image. The objective of compression is to reduce the number of bits as
much as possible, while keeping the resolution and the visual quality of the reconstructed image as close to the
original image as possible. Image compression systems are composed of two distinct structural blocks: an
encoder and a decoder.
II. Compression Techniques
A. Quantization Technique
Quantization refers to the process of converting the continuous pixel values (such as decimal values) to
discrete values (such as integers). The quantizer performs a lossy image compression. The input to a quantizer is
the original data, and the output is always one among a finite number of levels. The quantizer is a function
whose set of output values are discrete, and usually finite. This is a process of approximation, and a good
quantizer is one which represents the original signal with minimum loss or distortion. The different types of
quantization are–
Scalar Quantization
Vector Quantization.[7]
B. Entropy Coding Technique
After the quantization has been applied to the image, a symbol encoding technique is applied to the
image. Entropy is the amount of information present in the data, and an entropy coder encodes the given set of
symbols with the minimum number of bits required to represent them.[7] Entropy Coding techniques mostly
provide lossless compression. Some of the entropy coding algorithms are-
Huffman Coding
Arithmetic Coding
Run Length Coding
Data Folding](https://image.slidesharecdn.com/k017236770-151208120519-lva1-app6891/85/Lossless-Image-Compression-Using-Data-Folding-Followed-By-Arithmetic-Coding-1-320.jpg)
![Lossless Image Compression using Data Folding followed by Arithmetic Coding
DOI: 10.9790/0661-17236770 www.iosrjournals.org 70 | Page
Fireworks.jpg Leaves.jpg
Fig.5 Database being used
After applying the proposed technique on the above database following observations has been
concluded on the basis of bits per pixel (BPP):
Image Wavelet Huffman Coding Data Folding Proposed Technique
Cmp5 txt 256X256 2.5 2.13 6.31 1.11
Baboon 512X512 1.28 1.5 7.22 1
Boat 512X512 1.003 0.946 5.84 0.638
Elaine 512X512 3.986 3.84 5.86 3.07
Pod rear draw 512X512 0.584 0.627 2.28 0.229
Cathedral 1024X1024 0.618 0.796 5 0.534
Flower 1024X1024 0.581 0.594 2.95 0.326
Temple 1024X1024 0.233 0.328 2.66 0.198
Deer 2048X2048 0.249 0.231 4.47 0.203
Fireworks 2048X2048 0.308 0.56 2.4 .287
Leaves 2048X2048 0.504 0.516 5.78 0.426
Table 1 Results based on BPP
V. Conclusion
We had analysis the different compression techniques and concluded that these techniques are either
lossy or lossless. After this analysis we have proposed a lossless image compression technique known as data
folding followed by arithmetic coding. In this technique we have performed the hybridization of these two
techniques. First row folding followed by column folding has been performed, after that arithmetic coding has
been applied on the resultant image. This technique has been applied on different images of different
dimensions. This technique gives better results in terms of Compression Ratio and Bits Per Pixel. But this
technique has the problem of high computational time.
References
Journal Papers:
[1]. Suresh Yerva, Smita Nair, Krishnan Kutty “Lossless Image Compression based on Data Folding” IEEE-International Conference on
Recent Trends in Information Technology, ICRTIT 2011 MIT, Anna University, Chennai. June 3-5, 2011
[2]. Sonal, Dinesh Kumar Department of Computer Science & Engineering Guru Jhambheswar University of Science and Technology,
Hisar “A STUDY OF VARIOUS IMAGE COMPRESSION TECHNIQUES”
[3]. V. Singh* VNR Vignana Jyothi Institute of Technology, Hyderabad, India “Recent Patents on Image Compression – A Survey”
[4]. Asadollah Shahbahrami, Ramin Bahrampour, Mobin Sabbaghi Rostami, Mostafa Ayoubi Mobarhan, Department of Computer
Engineering, Faculty of Engineering, University of Guilan, Rasht, Iran “ Evaluation of Huffman and Arithmetic Algorithms for
Multimedia Compression Standards”
[5]. K.Uma, **P.Geetha palanisamy***P.Geetha poornachandran “Comparison of Image Compression using GA, ACO and PSO
techniques” IEEE-International Conference on Recent Trends in Information Technology, ICRTIT 2011 978-1-4577-0590-8/11
©2011 IEEE MIT, Anna University, Chennai. June 3-5, 2011
[6]. Loganathan R and Y.S.Kumaraswamy “An Improved Active Contour Medical Image Compression Technique with Lossless
Region of Interest”
[7]. G.M.Padmaja, P.Nirupama “ Analysis of Various Image Compression Techniques” ARPN Journal of Science and Technology
VOL. 2, NO. 4, May 2012
[8]. “Basic Image compression” by Gleb V. Tcheslavski
[9]. B.C. Vemuri, S. Sahni, F.Chen, C. Kapoor, C. Leonard “ Lossless Image Compression”.
[10]. Jagadish H. Pujar, Lohit M. Kadlaskar “A New Lossless Method Of Image Compression And Decompression Using Huffman
Coding Techniques” Journal of Theoretical and Applied Information Technology © 2005 - 2010 JATIT
[11]. Gaurav Vijayvargiya Dr. Sanjay Silakari Dr.Rajeev Pandey “A Survey: Various Techniques of Image Compression” (IJCSIS)
International Journal of Computer Science and Information Security, Vol. 11, No. 10, October 2013
[12]. Rajesh K. Yadav, S.P. Gangwar and Harsh V. Singh “Study and analysis of wavelet based image compression techniques”
International Journal of Engineering, Science and Technology Vol. 4, No. 1, 2012, pp. 1-7
[13]. Sindhu M, Rajkamal R “Images and Its Compression Techniques – A Review” International Journal of Recent Trends in
Engineering, Vol 2, No. 4, November 2009
[14]. Paul G. Howard and Jeffery Scott Vitter “Arithmetic Coding for Data Compresion”
[15]. Glen G. Langdon “An Introduction to Arithmetic Coding”
Books:
[16] Gonzales Rafeil C., Woods “ Digital Image Processing” Edition third, 416-440, 564-626.](https://image.slidesharecdn.com/k017236770-151208120519-lva1-app6891/85/Lossless-Image-Compression-Using-Data-Folding-Followed-By-Arithmetic-Coding-4-320.jpg)
Lossless Image Compression Using Data Folding Followed By Arithmetic Coding
- 1. IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278-0661,p-ISSN: 2278-8727, Volume 17, Issue 2, Ver. III (Mar – Apr. 2015), PP 67-70 www.iosrjournals.org DOI: 10.9790/0661-17236770 www.iosrjournals.org 67 | Page Lossless Image Compression Using Data Folding Followed By Arithmetic Coding Richa Goyal1 , Sourav Garg2 1,2, (CSE,ACET/ PTU, Jalandhar India) Abstract : The paper presents a lossless image compression technique using the hybridization of two different entropy coding techniques. Initially data folding technique has been applied to the image. A row folding is applied on the image matrix followed by a column folding. Multiple iterations of this process is applied on the image. After completing the data folding process another entropy coding technique known as arithmetic coding has been applied to the resultant image to get better results. Keywords: Lossless image compression, data folding, arithmetic coding, compression ratio, bits per pixel. I. Introduction Compression can be defined as reducing the size of data so that the amount of space required to store the data becomes less and it takes less time to transmit the data. With the help of compression it becomes easy to store large files. These files may be data files, images, videos, audios and other multimedia files. For transmitting a data, compression can be performed either on just the data content or on the entire transmission unit. There are different types of data content on which compression can be applied. Image compression can be referred to as reducing the size of the image so as it requires less space for storage and is easily transmitted. This can be done by reducing the memory size required to store each pixel. There are several algorithms to perform image compression. These algorithms can be categorized as lossy and lossless compression techniques. Lossy compression causes loss of information after compression. When the compressed data is decomposed to its original version then both are not same but close. It is an irreversible process. In lossless compression the compressed data can be easily decomposed to its original version. The loss of data is very less. It is a reversible process. Image compression techniques reduce the number of bits required to represent an image by taking advantage of these redundancies. An inverse process called decompression (decoding) is applied to the compressed data to get there constructed image. The objective of compression is to reduce the number of bits as much as possible, while keeping the resolution and the visual quality of the reconstructed image as close to the original image as possible. Image compression systems are composed of two distinct structural blocks: an encoder and a decoder. II. Compression Techniques A. Quantization Technique Quantization refers to the process of converting the continuous pixel values (such as decimal values) to discrete values (such as integers). The quantizer performs a lossy image compression. The input to a quantizer is the original data, and the output is always one among a finite number of levels. The quantizer is a function whose set of output values are discrete, and usually finite. This is a process of approximation, and a good quantizer is one which represents the original signal with minimum loss or distortion. The different types of quantization are– Scalar Quantization Vector Quantization.[7] B. Entropy Coding Technique After the quantization has been applied to the image, a symbol encoding technique is applied to the image. Entropy is the amount of information present in the data, and an entropy coder encodes the given set of symbols with the minimum number of bits required to represent them.[7] Entropy Coding techniques mostly provide lossless compression. Some of the entropy coding algorithms are- Huffman Coding Arithmetic Coding Run Length Coding Data Folding
- 2. Lossless Image Compression using Data Folding followed by Arithmetic Coding DOI: 10.9790/0661-17236770 www.iosrjournals.org 68 | Page C. Optimization Technique The optimization techniques can effectively reduce the encoding time while retaining the quality of the retrieved. Various optimization techniques are listed below: Genetic Algorithm Particle Swarm Optimization Ant Colony Optimization III. Proposed Technique Hybridization of two lossless image compression techniques has been used to obtain better results. Data Folding technique has been applied to the image followed by another entropy coding technique i.e. Arithmetic Coding. These two techniques have been individually used earlier to provide lossless image compression. But using these two techniques in one algorithm are supposed to provide better Compression Ratio (CR) and lesser Bits Per Pixel (BPP). Though none of the technique can be considered as completely lossless but using these techniques the loss has been expected to be minimum. A. Data Folding Data folding is a very effective algorithm that can be used for lossless image compression. The simple method applied on the image is to subtract the even pixels from odd pixels and the store this difference in a separate buffer. This one step is known as folding. Then further folding is applied to pixel values stored in the separate buffer. In this way, a number of iterations have been applied to the data. The process of data folding consists of two steps: Row Folding Column Folding Row folding is the process of folding the rows of the image matrix. Even row is subtracted from the odd row. Then the resultant odd row is stored in a separate buffer. Then column folding is applied on the data stored in this buffer. Column folding is the process of folding the columns of the image matrix. After making a row folding even column is subtracted from the odd column. Then the resultant odd column is stored in a separate buffer. Then further iterations of row folding followed by column folding are applied on the data stored in this buffer. Fig.1 Image Matrix Fig.2 Image after 1st Row Folding Fig.3 Image after 1st Column Folding B. Arithmetic Coding After applying the technique of data folding, an entropy coding technique has been used. This entropy coding technique is known as arithmetic coding. In this approach, the entire source symbol is assigned a single arithmetic code. Initially all the symbols are defined inside a fixed window size of (0, 0.5) and arranged according to their probability distribution. All the pixels lying within the interval are used to represent the image and others are removed. Then the window size is narrowed and again the symbols are arranged according to
- 3. Lossless Image Compression using Data Folding followed by Arithmetic Coding DOI: 10.9790/0661-17236770 www.iosrjournals.org 69 | Page their probability distribution. All the pixels lying within this narrowed range are used to represent the image and others are removed. This process is being repeated for 6 times. Fig.4 Flow Chart for Compression Procedure IV. Results The proposed technique has been applied to the following database: Cmp5 text.jpg Leena.jpg Baboon.jpg Boat.jpg Elaine.jpg Pod rear draw.jpg Cathedral.jpg Flower.jpg Temple.jpg Deer.jpg
- 4. Lossless Image Compression using Data Folding followed by Arithmetic Coding DOI: 10.9790/0661-17236770 www.iosrjournals.org 70 | Page Fireworks.jpg Leaves.jpg Fig.5 Database being used After applying the proposed technique on the above database following observations has been concluded on the basis of bits per pixel (BPP): Image Wavelet Huffman Coding Data Folding Proposed Technique Cmp5 txt 256X256 2.5 2.13 6.31 1.11 Baboon 512X512 1.28 1.5 7.22 1 Boat 512X512 1.003 0.946 5.84 0.638 Elaine 512X512 3.986 3.84 5.86 3.07 Pod rear draw 512X512 0.584 0.627 2.28 0.229 Cathedral 1024X1024 0.618 0.796 5 0.534 Flower 1024X1024 0.581 0.594 2.95 0.326 Temple 1024X1024 0.233 0.328 2.66 0.198 Deer 2048X2048 0.249 0.231 4.47 0.203 Fireworks 2048X2048 0.308 0.56 2.4 .287 Leaves 2048X2048 0.504 0.516 5.78 0.426 Table 1 Results based on BPP V. Conclusion We had analysis the different compression techniques and concluded that these techniques are either lossy or lossless. After this analysis we have proposed a lossless image compression technique known as data folding followed by arithmetic coding. In this technique we have performed the hybridization of these two techniques. First row folding followed by column folding has been performed, after that arithmetic coding has been applied on the resultant image. This technique has been applied on different images of different dimensions. This technique gives better results in terms of Compression Ratio and Bits Per Pixel. But this technique has the problem of high computational time. References Journal Papers: [1]. Suresh Yerva, Smita Nair, Krishnan Kutty “Lossless Image Compression based on Data Folding” IEEE-International Conference on Recent Trends in Information Technology, ICRTIT 2011 MIT, Anna University, Chennai. June 3-5, 2011 [2]. Sonal, Dinesh Kumar Department of Computer Science & Engineering Guru Jhambheswar University of Science and Technology, Hisar “A STUDY OF VARIOUS IMAGE COMPRESSION TECHNIQUES” [3]. V. Singh* VNR Vignana Jyothi Institute of Technology, Hyderabad, India “Recent Patents on Image Compression – A Survey” [4]. Asadollah Shahbahrami, Ramin Bahrampour, Mobin Sabbaghi Rostami, Mostafa Ayoubi Mobarhan, Department of Computer Engineering, Faculty of Engineering, University of Guilan, Rasht, Iran “ Evaluation of Huffman and Arithmetic Algorithms for Multimedia Compression Standards” [5]. K.Uma, **P.Geetha palanisamy***P.Geetha poornachandran “Comparison of Image Compression using GA, ACO and PSO techniques” IEEE-International Conference on Recent Trends in Information Technology, ICRTIT 2011 978-1-4577-0590-8/11 ©2011 IEEE MIT, Anna University, Chennai. June 3-5, 2011 [6]. Loganathan R and Y.S.Kumaraswamy “An Improved Active Contour Medical Image Compression Technique with Lossless Region of Interest” [7]. G.M.Padmaja, P.Nirupama “ Analysis of Various Image Compression Techniques” ARPN Journal of Science and Technology VOL. 2, NO. 4, May 2012 [8]. “Basic Image compression” by Gleb V. Tcheslavski [9]. B.C. Vemuri, S. Sahni, F.Chen, C. Kapoor, C. Leonard “ Lossless Image Compression”. [10]. Jagadish H. Pujar, Lohit M. Kadlaskar “A New Lossless Method Of Image Compression And Decompression Using Huffman Coding Techniques” Journal of Theoretical and Applied Information Technology © 2005 - 2010 JATIT [11]. Gaurav Vijayvargiya Dr. Sanjay Silakari Dr.Rajeev Pandey “A Survey: Various Techniques of Image Compression” (IJCSIS) International Journal of Computer Science and Information Security, Vol. 11, No. 10, October 2013 [12]. Rajesh K. Yadav, S.P. Gangwar and Harsh V. Singh “Study and analysis of wavelet based image compression techniques” International Journal of Engineering, Science and Technology Vol. 4, No. 1, 2012, pp. 1-7 [13]. Sindhu M, Rajkamal R “Images and Its Compression Techniques – A Review” International Journal of Recent Trends in Engineering, Vol 2, No. 4, November 2009 [14]. Paul G. Howard and Jeffery Scott Vitter “Arithmetic Coding for Data Compresion” [15]. Glen G. Langdon “An Introduction to Arithmetic Coding” Books: [16] Gonzales Rafeil C., Woods “ Digital Image Processing” Edition third, 416-440, 564-626.