Image Compression using THV method in Lossless Image Compression & in Lossy Image Compression

Download Full-Text PDF Cite this Publication

Text Only Version

Image Compression using THV method in Lossless Image Compression & in Lossy Image Compression

Dr. Anjali Mathur1

Department of Mathematics

Jodhpur Institute of Engineering & Technology Jodhpur, India

Nitesh Agarwal2

Dr. Sandeep Mathur3

Department of Computer Science Jodhpur Institute of Engineering & Technology

Jodhpur, India

Department of Mathematics

Jodhpur Institute of Engineering & Technology Jodhpur, India

AbstractImage compression process reduce required storage size of image. Image compression process use two technique to compress image lossless image compression & lossy image compression. Images that provide numerical, secure & financial information compressed using lossless image compression because we required original data back after decompression process. Lossless image compression use some entropy encoding technique but its compression ratio is low w.r.t lossy image compression. But other images like multimedia images can be compressed using lossy image compression because the human eye is very tolerant of approximation error in an image. Hence we may decide to exploit this tolerance to produce increased compression, at the expense of image quality by reducing some pixel data or information. Using this concept this paper proposed a THV (Threshold Variable) method before entropy encoding technique to get more compressed size using lossless image compression as well as using lossy image compression & give a comparative study in which technique, proposed method is more useful. This paper deals with comparative study of a compressed image on the basis of different value of threshold variable used in THV method.

Key Words: THV, DCT, RLE, MSE, PSNR.

  1. INTRODUCTION

    A digital image is a 2D pixel matrix where each position of pixel gives a color information for image in bits format. On the basis of this bits format image is classified as 2 bit, 6 bit, 8 bit, 16 bit, 24 bit & 32 bit. When an image is design using one of these format each pixel store information in particular bit format in which they are build. To store image in only bits format some time required a high amount of storage device for example if an image is build using 8 bit format then each pixel must have 8 bit storage even it represent information that required less than 8 bit to store. As the digital devices has limited storage & transmission capability we need to compress the image by some suitable method to satisfy this limitation. For example when we play a YouTube video if our communication system have enough bandwidth to play YouTube high resolution video then we can play video without any buffering problem but if bandwidth is low then YouTube use some compression method & transfer the bits of video frame in compressed format according to our network bandwidth size. By lossless image compression we get original image in decompression process without any loss of pixel value. But other images like multimedia images can be

    compressed using lossy image compression because the human eye is very tolerant of approximation error in an image. Hence we may decide to exploit this tolerance to produce increased compression, at the expense of image quality by reducing some pixel data or information.

      1. Lossless Image Compression

        Lossless image compression process compress the image for storage & communication purpose in such a way that original image can be retrieve during decompression process without any loss or modification of information (pixel bits). Lossless image compression use some entropy encoding techniques to compress digital image such as RLE (Run Length Encoding), Huffman Encoding, LZW (Lempel-Ziv- Welch) Encoding, Area Encoding etc. Process of lossless image compression is shown in fig 1 [14].

        Image

        Entropy Encoding

        Image

        Entropy Encoding

        Image

        Image

        Channel Entropy Decoding

        Fig 1 Lossless Image Compression

        Entropy encoding give good compression ratio when image have repeated pixel value sequentially but all the image not have such type of repeated pattern hence present introduce a THV module before encoding technique to make a repeated sequence [5].

        Image

        Proposed THV Method

        RLE

        Image

        Proposed THV Method

        RLE

        Channel

        Image with Some Pixel modification

        Inverse

        Image with Some Pixel modification

        Inverse

        Fig 2 Proposed Lossless Image Compression

        In proposed method as shown in fig 2 image in pixel format transfer to the THV module & then send to the entropy where using RLE proposed method compress input image. THV module modify pixel values of image to get repeated sequence hence output image after the decoding process is equivalent to original image but not equal to the original image data.

        1. THV (Threshold Variable) Method

          THV method take pixel value from 2D pixel matrix row by row. THV method uses two node 1st node store the 1st pixel & 2nd node move forward row by row in pixel matrix. THV method take pixel difference between pixel value store in these two node & repeat the pixel value store in node 1 until difference between these two nodes are not greater than threshold variable t used in THV method. One a difference greater than t occurs node 1 store that pixel & node 2 traverse pixel one by one from neighboring pixel of pixel store in node 1 & same process is follow until complete pixel matrix not traversed. For example

          101

          100

          105

          110

          201

          200

          205

          210

          306

          304

          307

          306

          406

          404

          407

          406

          502

          501

          500

          502

          602

          601

          600

          602

          728

          729

          730

          732

          828

          829

          830

          832

          201

          205

          202

          203

          905

          904

          900

          902

          555

          554

          553

          550

          551

          552

          557

          558

          602

          605

          604

          602

          601

          701

          705

          709

          905

          905

          905

          905

          905

          906

          912

          911

          Table 1: Pixel matrix of input image

          Table 1 shows a 2D pixel matrix but this matrix cannot be compressed using RLE because pixel value not repeated sequentially. THV method convert this matrix so that it can be compressed using RLE. Let the value of variable t in THV method is 10 then converted pixel matrix is

          602

          101

          101

          101

          101

          201

          201

          201

          201

          306

          306

          306

          306

          406

          406

          406

          406

          502

          502

          502

          502

          602

          602

          602

          728

          728

          728

          728

          828

          828

          828

          828

          201

          201

          201

          201

          905

          905

          905

          905

          555

          555

          555

          555

          555

          555

          555

          555

          602

          602

          602

          602

          602

          701

          701

          701

          905

          905

          905

          905

          905

          905

          912

          912

          Table 2: Pixel matrix of After THV method with

          variable t =10

          After the THV method pixel matrix contain a good no of repeated pixel as shown in table 2. This repeated pixel helps the RLE to compress pixel matrix.

          THV algorithm

          Input: Pixel matrix of input image. Output: Modified Pixel Matrix.

          {

          h = height of pixel matrix;

          pixel[h][w] = pixel matrix of original image; t = THV variable;

          for(i=0; i<h; i++)

          {

          j=0;

          tmp=pixel[i][j]; for(j=0; j<w; j++)

          {

          if( (difference between tmp & pixel[i][j]) > t

          ) pixel[i][j] = tmp;

          else tmp=pixel[i][j];

          }

          }

          } [1]

          THV method used in such type of image where modifying some pixel data does not cause any big problem.

        2. RLE (Run Length Encoding)

    This is a very simple compression technique method used for compressing sequential data. Many digital image consist pixel values that are repeats sequentially for such type of image RLE is useful. In proposed THV method RLE receive sequential data from pixel matrix modified by THV method & store pixel value that repeats & no of time that pixel value repeat sequentially. For example table 2 by RLE compressed as

    Pixel Value

    Repetition

    Pixel Value

    Repetition

    101

    4

    201

    4

    201

    4

    905

    4

    306

    4

    555

    8

    406

    4

    602

    5

    502

    4

    701

    3

    602

    4

    905

    6

    728

    4

    912

    2

    828

    4

    Table 3: Compressed Data after RLE

    Table 3 required less storage space as compare to table 1. Table 1 require total 64 values to store but table 3 require only 30 values to store [5].

    Compression Ratio (CR) = 64/30 =

    2.13 1.2 Lossy Image compression

    Lossy image compression gives a high compression ration then lossless image compression. It is say as lossy because it modify or destroy some pixel information by using some transformation like DCT, DST, KLT, DFT etc. & using quantization table before entropy encoding.

    w = width of pixel matrix;

    Fig 3: Lossy Image Compression

    In this paper we use DCT (Discrete Cosine Transformation) to transform image in frequency domain [4].

        1. DCT (Discrete Cosine Transform) [11]

          DCT convert an image into its equivalent frequency domain by partitioning image pixel matrix into blocks of size N*N. An image is a 2D pixel matrix hence 2D DCT is used to transform an image.

          2-D DCT can be defined as

          N 1 N 1 2x 1u 2y 1v

          Where (i,t) define position of input & output value, QDCT is DCT block after quantization, QT is standard quantization matrices & defined as

          Cu,vu (v) f x, ycos cos

          x0 y 0 2N 2N

          for u, v = 0,1,2,,N 1.

          & inverse transformation is defined as

          N 1 N 1 2x 1u 2y 1v

          1

          16

          11

          10

          16

          24

          40

          51

          61

          12

          12

          14

          19

          26

          58

          60

          55

          14

          13

          16

          24

          40

          57

          69

          56

          14

          17

          22

          29

          51

          87

          80

          62

          18

          22

          37

          56

          68

          109

          103

          77

          24

          35

          55

          64

          81

          104

          113

          92

          49

          64

          78

          87

          103

          121

          120

          101

          72

          92

          95

          98

          112

          100

          103

          99

          16

          11

          10

          16

          24

          40

          51

          61

          12

          12

          14

          19

          26

          58

          60

          55

          14

          13

          16

          24

          40

          57

          69

          56

          14

          17

          22

          29

          51

          87

          80

          62

          18

          22

          37

          56

          68

          109

          103

          77

          24

          35

          55

          64

          81

          104

          113

          92

          49

          64

          78

          87

          103

          121

          120

          101

          72

          92

          95

          98

          112

          100

          103

          99

          f x, y u(v)cu,vcos

          u 0 v0

          cos

          2N 2N

          2

          1.2.3 Proposed THV method in Lossy Image Compression

          Where Cu, v represents frequency value for u, v &

          f x, y represents pixel color value at position ( x, y ).

          1 for u 0

          Proposed THV method can be used in Lossy Image compression before entropy encoding as

          (u)

          N 3

          2

          N

          for u 0

          (v)

          1 for v 0

          N 4

          Fig 4: Lossy Image Compression with THV Method

          The THV algorithm is work in same way as in lossless image compression the only difference is that in lossless image

          2

          for v 0

          compression it take input from image pixel matrix but in lossy

          N

        2. Quantization

    A Quantizer simply reduces the number of bits needed to store the transformed coefficients by reducing the precision of those values. Since this is a many-to-one mapping, it is a lossy process and is the main source of compression in an encoder.

    The quantization matrix is designed to provide more resolution to more perceivable frequency components ove less perceivable components (usually lower frequencies over high frequencies) in addition to transforming as many components to 0, which can be encoded with greatest efficiency. A DCT block is quantize using following formula

    DCT (i, j)

    QDCTi, j ROUND 5

    QT (i, j)

    & this QDCT block dequantize by following formula

    DCTi, j ROUND QDCT(i, j) *QT(i, j) 6

    For i, j= 0, 1, 2, 3.,N-1

    image compression it take input from quantize block & loop in THV depend upon dimension of quantize block not on dimension of image.

  2. MAIN RESULTS & OUTPUTS

    1. Implementation of Proposed THV method in lossless Image Compression

      Steps involved in this implementation

      1. Create pixel matrix of the image.

      2. Apply THV method on pixel matrix & apply THV algorithm.

      3. Use RLE as entropy encoding on pixel matrix obtain from THV algorithm.

      4. Store matrix obtain by RLE method in to secondary storage.

      5. To get required image read encoded matrix from secondary storage & apply entropy decoding (Run Length Decoding) on that encoded matrix.

      6. Using this decoded matrix make pixel matrix & then using this pixel matrix make required image.

      7. Now we Find MSE (Mean Squared Error), PSNR (Peak Signal To Noise Ratio) & CR (Compression Ration) to determine quality of image obtain by

proposed method for each t variable used in THV method. MSEX, PSNRX & CRX calculated by following formulas [12] –

1

1

H 1 W 1

MSEt

[O(x, y)M t (x, y)]2 7

H *W x0 y0

PSNRt=20*log10 (MAX) – 10*log10 (MSEt) (8)

CR

CR

Original Im age size

t Output Im age size 9

Where H=Height of Image, W= Width of Image, variable MAX shows max value of a pixel for example here image is 8 bit hence MAX=255, MSEt, PSNRt & CRt is MSE, PSNR & CR at variable t used in THV method.

Quality of image obtain by proposed method is depend on MSEt & PSNRt value. If as the MSE value increases PSNR value decreases then we get a bad quality of image by proposed method & if as the MSE value decreases PSNR value increases we get a batter quality image hence on basis of this MSEt & PSNRt value proposed method gives a best value of X on which we get a high compressed image with best quality.

2.1.1 Outputs

2.1.1.1 Lossless image compression without THV method

Lossless Image Compression

Uncompressed Compressed Image Image Size=768 KB Size=343 KB

Fig 5: Lossless Image Compression without THV Method

MSE

PSNR

CR

Lossless Image Compression without THV Method

0

Infinity

2.24

Table 6: MSEt, PSNRt, CRt on different value of t

          1. Graphs

            1. THV variable t vs. CRt

              Fig 7: Variation in CRt with different value of THV variable t

            2. THV variable t vs. MSEt

              Fig 8: Variation in MSEt with different value of RVM

              variable t

              Table 5: MSE, PSNR, CR without THV method

              2.1.1.2 Lossless Image Compression with proposed THV method

            3. THV variable t vs. PSNRt

              t=1

              t=2

              Size=247 KB

              Uncompressed t=3 Image

              Size = 768 KB

              Size=191 KB

              Size=159 KB

              Fig 9: Variation in PSNRt with different value of THV

              variable t

            4. THV variable t vs. PSNRt & MSEt

t=4

Size=128 KB

Compressed Images

Fig 6: Proposed Lossless Image Compression with variation

Fig 10: Variation in PSNRt & MSEt with different

value of THV variable t

in THV Variable t

    1. Implementation of Proposed THV method in lossy Image Compression

      Steps involved in this implementation

      1. Create pixel matrix of the image & divided it into blocks of size 8*8

      2. Apply FDCT (Forward Discrete Sine Transform) on each 8*8 block of pixel matrix to get equivalent 8*8 DCT blocks using eq (1).

      3. Apply eq (5) on each block of DCT to get QDCT block.

      4. Apply THV algorithm on each block of QDCT to get THV block.

      5. Combine each THV block & apply RLE on combine block & store this encoded block on secondary storage.

      6. To get required image read encoded matrix from secondary storage & apply entropy decoding (Run Length Decoding) on that encoded matrix.

      7. Divide this decoded matrix in to blocks of size 8*8.

      8. Apply eq (6) on each block to get DCT blocks.

      9. Apply eq (2) on each DCT block to get IDCT blocks.

      10. Combine all IDCT blocks to get pixel matrix.

      11. Using pixel matrix we get required image.

Now we Find MSE (Mean Squared Error), PSNR (Peak Signal To Noise Ratio) & CR (Compression Ration) to determine quality of image obtain by proposed method for each t variable used in THV algorithm.

      1. Outputs

        1. Lossy Image compression without THV method

          Lossy Image Compression

          Uncompressed Image Compressed Image

          Size=768 KB Size=71.9 KB

          Fig 11: Lossy Image Compression without THV Method

          MSE

          PSNR

          CR

          Lossy Image Compression without THV Method

          7.37

          39.45

          10.68

          Table 7: MSE, PSNR, CR without THV

        2. Lossy Image Compression with proposed THV

          method t=1

          Table 8: MSEt, PSNRt, CRt on different value of t

          1. Graphs

            1. THV variable t vs. CRt

              Fig 13: Variation in CRt with different value of THV variable t

            2. THV variable t vs. MSEt

              Fig 14: Variation in MSEt with different value of RVM

              variable t

            3. THV variable t vs. PSNRt

              t=2

              Uncompressed t=3 Image Size=768 KB

              Size=47.9 KB

              Size=39.9 KB

              Size=39.9 KB

              Fig 15: Variation in PSNRt with different value of THV variable t

            4. THV variable t vs. PSNRt & MSEt

t=4

Size=31.9 KB

Fig 16: Variation in PSNRt & MSEt with different

value of THV variable t

Compressed Images

Fig 12: Lossy Image Compression with THV Method

3. CONCLUSION

The result presented in this document shows that

  1. The results shows that as the value of variable t increases storage size of image decreases as shown in Fig 6, Fig 12, Table 6 & Table 8.

    1. Asha Lata & Permender singh Review of Image Compression Techniques, International Journal of

      Emerging Technology & Advanced Engineering (IJETAE), vol 3, issue 7, pp. 461-464, July 2013.

    2. Firas A. Jassim and Hind E. Qassim, FIVE MODULUS METHOD FOR IMAGE COMPRESSION, signal & image processing: An

  2. As the value of t increases CRt also increases as in lossless image compression as shown in Fig 7 but in lossy image compression as the value of t increases

    1. International Journal (SIPIJ), vol 3, no.5, pp. 19-28, Octobar 2012

      Harley R. Myler and Arthur R. Weeks The Pocket Handbook of Image Processing Algorithms in C, ISBN

      0-13-642240-3 Prentice Hall P T R Englewood Cliffs, New Jercy 07632.

      CRt goes to a constant value.

  3. As the value of t increases proposed process add more noises in the image i.e. value of MSEt increases as shown in Fig 8 & Fig 14.

  4. The results shows that proposed method give good results in lossless image compression but in lossy image compression present method not works well because it add high aount of noises in image that cannot be tolerable this can be seen by comparing Table 6 & Table 8.

  5. As the value of t increases PSNRt value decreases as shown in Fig 9 & Fig 15.

  6. As the MSEt value decreases & PSNRt increases quality of image improves but CRt decreases.

  7. Fig 13 show after value 3 of variable t CRt almost constant in lossy image compression.

  8. Image used in this paper for lossless image compression have almost same MSEt & PSNRt value for t=15 as shown in Fig 10 .

  9. Lossy image compression with THV method give more compression ratio than lossless image compression with THV method but lossless image compression with THV method give good results because its MSEt values are less than MSEt values in lossy image compression with THV for same value of t.

  1. Iain E.G. Richardson H.264 and MPEG-4 Video Compression: Video

    Coding for Next-generation Multimedia, ISBN 0470848375, 9780470848371, Wiley, 2003.

  2. Jesse D. Kornblum Using JPEG quantization tables to identify imagery processed by software, ELSEVIER,

    DIGITAL INVESTIGATION 5, pp. S21-S25,2008.

  3. Maneesha Gupta and Dr.Amit Kumar Garg, Analysis Of Image Compression Algorithm Using DCT International

    Journal of Engineering Research and Applications (IJERA), vol.2, pp. 515-521, Jan-Feb 2012

  4. N. Ahmed, T. Natarajan, and K.R. Rao, Discrete Cosine Transform, IEEE Transactions on Computers, vol. C-32, pp. 90-93, Jan. 1974.

  5. Nitesh Agarwal and Dr. A.M. Khan Application of DCT in image processing, International Journal of

    Engineering Research & Technology (IJERT), ISSN: 2278-0181, pp. 185-189, 2014.

  6. Swati Dhamija and Priyanka Jain Comparative Analysis for Discrete Sine Transform as a suitable method for noise estimation IJCSI International Journal of Computer Science Issues, Vol. 8, Issue 5, No 3, September 2011 pp. 162-164.

  7. V.Singh Recent Patents on Image Compression A Survey, Recent Patents on signal Processing (Bentham

Open), vol 2, pp. 47-62, 2010 .

REFERENCES

  1. A. H. Husseen , S. Sh. Mahmud & R. J. Mohammed

    Image Compression Using Proposed Enhanced Run Length Encoding Algorithm , Ibn Al-Haitham Journal For Pure And Applied Science, VOL 24(1), pp. 315-328, 2011.

  2. A.M. Raid, W.M. khedr, M.A. El-dosuky and wesan

    Ahmed JPEG image compression using Discrete Cosine Transform A survey, International Journal of

    Computer Science & Engineering Survey (IJCSES), vol 5, no 2, pp. 29- 47, April 2014.

  3. Andrew B. Watson, Image Compression Using Discrete Cosine Transform, NASA Ames Research Centre, 4(1), pp. 81-88,1994

  4. Anjali Kapoor and Dr. Renu Dhir, Image Compression Using Fast 2-D DCT Technique, International Journal on

Computer Science and Engineering (IJCSE), vol. 3 pp. 2415-2419, 6 June 2011

Leave a Reply

Your email address will not be published. Required fields are marked *