AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING

International Journal of Computer Engineering and Applications, Volume XI, Issue V, May 17, www.ijcea.com ISSN 2321-3469 AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING Scholar in the Dept of Computer Science & Technology, S.K.U., Anantapuram, ABSTRACT: The propose procedure is Pixel Based Leveling (PBL) Technique. PBL is a lossy strategy and appropriate for still pictures. In PBL strategy picture is size is decreased in every level. PBL method is finished with two methodologies. One of the methodologies is leveling and another is leveling with bit cut. We have utilized 8 levels, up to 3 levels picture won't exasperate and gives more pressure rate. Be that as it may, in level 4 onwards picture is bothered a considerable measure and gives preferred pressure over level 3.In every level unique picture can be recovered with loss of information. In first phase a sample image is taken and bit slicing technique is applied from bit slice 1 to bit slice 8. On the application of bit slice 1 and bit slice 2 there is no much changes in size and image is disturbed a lot. And on the application of bit slice 3 onwards size of the image is goes on reducing and clarity of image is also increasing In second phase, a bit slice 8 image is taken and leveling technique i.e. only level 1 is applied for further compression. This approach is showing better compression than first approach. Key words: Pixel Based Leveling,, slice 1. INTRODUCTION A binary image is a digital image that has only two possible values for each pixel. Typically the two colors used for a binary image are black and white though any two colors can be used. The color used for the object(s) in the image is the foreground color while the rest of the image is the background color. In the document-scanning industry this is often referred to as "bi-tonal" [1,6]. Parallel pictures are additionally called bi-level or two-level. This implies every pixel is put away as a solitary piece i.e., 0 or 1. The names highly contrasting, B&W, mono chrome or monochromatic are frequently utilized for this idea, however may likewise assign any pictures that have just a single specimen for every pixel, for 1

AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING example, grayscale pictures[8,9]. In Photoshop speech, a parallel picture is the same as a picture in "map" mode. There are numerous operations which are performed on paired pictures. A whole class of operations on paired pictures works on a 3 3 window of the picture. This contains nine pixels, so 512 conceivable qualities. Considering just the focal pixel, it is conceivable to characterize whether it stays set or unset, in view of the encompassing pixels. Cases of such operations are diminishing, widening, discovering branch focuses and endpoints, evacuating disconnected pixels, moving the picture a pixel in any bearing, and breaking H-associations. Conway's Game of Life is additionally a case of a 3 3 window operation. A very important characteristic of a binary image is the distance transform. This gives the distance of every set pixel from the nearest unset pixel. The distance transform can be efficiently calculated. It allows efficient computation of Voronoi diagrams, where each pixel in an image is assigned to the nearest of a set of points. It also allows skeletonization, which differs from thinning in that skeletons allow recovery of the original image. The distance transform is also useful for determining the centre of the object, and for matching in image recognition. 2. Pixel Based Leveling (PBL) Technique This technique is based on spatial domain of the image and is lossy compression technique. The main objective of this technique is to compress the image with loss of data and suitable for digital still images. PBL technique is performed in two approaches. In the first approach, PBL is applied directly on an image and in second approach, PBL is applied on bit sliced image.both the approaches are showing better compression than other lossy compression technique. The two approaches of PBL is shown below[3] Figure 1. Pixel based leveling method LZW Compression Compressed Image Source Image (a) slicing technique Pixel based leveling method LZW Compression Compressed Image Source Image Figure 1: PBL Compression model 2

International Journal of Computer Engineering and Applications, Volume XI, Issue V, May 17, www.ijcea.com ISSN 2321-3469 2.1 PBL Technique Pixel based leveling (PBL) method comprises of applying leveling request at every phase to the pixels in the picture. The new esteem for each pixel is acquired by doing a move operation dictated by the request of the level. The resultant picture landed subsequent to playing out the leveling request is a compacted picture of littler size than the source picture. The request of leveling is expanded at every phase to get a more packed picture bringing about a littler picture measure. At every stage the proportion of pressure increments with the expanding request of leveling. This strategy holds great till the request of leveling achieves 8.The distinction in the source picture and packed picture can't be perceived by human eye till the third request of leveling is appeared in figure 2. After the 3rdorder, the distinction in source picture and compacted picture can be taken note. Since, there is a distinction in the source picture and the packed picture in the higher requests this system falls under the lossy approach of pressure. Level1 Level2 Level3 Figure 2 : Leveling images from level 1 to level3 In PBL technique a sample images are taken which are suitable for lossy compression. This approach involves two phases. Figure 1 shows model of compression[2]. In first phase, leveling technique is applied and in second phase LZW compression technique [5,7] is applied to reduce the size of the leveled image. In PBL, sample images are taken, one of the sample image, leveling technique is applied from level 1 to level 8.The below Figure 3.18 shows output of each level output. In level 1, level2 and level 3 images is not disturbed in clarity but reduced in size. From level 4 onwards image will disturbed a lot and size is also reduced is shown Table 3.1.In our proposed work PBL, we applied up to level 3. 3

AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING Level1 Level2 Level3 Level4 Level5 Level6 Level7 Level8 Figure 3. : Levels of image from 1-8 Size of the some of sample images before compression and after compression is shown table.1. Sizes are measured in terms of bytes. Table.1 Sizes of images in each level Images RAW Level1 Level2 Level3 Level4 Level5 Level6 Level7 Level8 Lena 21,025 12,408 9,625 7,102 5,163 3,348 1,876 820 46 Baboon 16,384 11,494 9,435 7,434 5,413 3,705 2,253 1,667 46 Konga 65,536 12,654 8,979 6,411 3,990 2,815 2,016 801 112 Lion 215,550 125,172 98,239 73,697 52,171 34,201 19,923 8,337 47 Pepper 16,384 9,105 7,095 5,230 3,654 2,641 1,748 868 46 4

International Journal of Computer Engineering and Applications, Volume XI, Issue V, May 17, www.ijcea.com ISSN 2321-3469 The below graph shows up to 5 levels of the some of sample images before compression and after compression is. Sizes are measured in terms of bytes. Graph 1: Levels of images 2.2 PBL Technique with Slicing (Approach II) This approach is also involves two phases. It is also same as approach one but need add one more step is bit slice. Procedure for compression are follows the below steps. Take Source image Apply the Slicing technique from bit slice 1 to bit slice 8. Apply PBL(PBL) technique of the respective leveling order to each pixel in the image only level1 is applied. Apply the LZW technique In first phase a sample image is taken and bit slicing technique is applied from bit slice 1 to bit slice 8. On the application of bit slice 1 and bit slice 2 there is no much changes in size as shown in figure 4 and image is disturbed a lot. And on the application of bit slice 3 onwards size of the image is goes on reducing and clarity of image is also increasing is shown figure 4. 16740bytes 16740bytes 16725bytes 16621bytes 16360bytes 5

AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING 15521bytes 13449bytes 10306 bytes 6369bytes Figure 4 Slice 1 images to slice 8 images In second phase, a bit slice 8 image is taken and leveling technique i.e. only level 1 is applied for further compression. This approach is showing better compression than first approach. and level1. Images The below table 3.2 shows sizes of an images after compression using bit slice RAW Slice1 Table 2. Sizes of images in each bit slice Slice 2 Slice 3 Slice 4 Slice 5 Slice 6 Slice 7 Slice 8 Level1 Lena 16725 16621 16360 16360 15521 13449 10306 6369 5,334 16725 Foot 16,740 5,335 4,429 3,627 2,892 2,257 1,736 1,198 747 1,736 Kneejoint 18,225 9,325 7,588 5,935 4,423 3,189 2,180 1,317 695 2,180 Headscan 15,625 7,641 5,778 4,308 3,126 2,288 1,627 46 46 1,627 Shoulder 18,225 8,254 6,626 5,252 3,899 2,847 2,013 1,384 769 2,013 The below graph shows up to 5 levels of the some of sample images after compression using bit slice. 6

International Journal of Computer Engineering and Applications, Volume XI, Issue V, May 17, www.ijcea.com ISSN 2321-3469 Graph 2: Sizes of images in each bit slice This phase primarily involves taking the source image which is the output of bit slice 8 and then applying the PBL technique of the order 2 n, the range on n varies from 1 to 8.The image obtained after PBL technique can further be compressed by applying LZW compression [4]. 2.3 Decompression Decompression is the reverse process of the compression to get back the original image from the compressed image. This also has two approaches, figure 5 shows the decompression models of approach I and approach II. Compressed Image Inverse LZW Compression Inverse Pixel based leveling method (a) Reconstructed Image Compressed Image Inverse LZW Compression Inverse Pixel based leveling method Figure 5 : Decompression models (a & b) Inverse slicing technique Reconstructed Image The following are steps followed in decompression approach I using PBL 7

AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING Take Compressed image Apply LZW decompression Apply IPBL(Inverse PBL) technique of the respective leveling order to each pixel in the image The following are steps followed in decompression approach II using PBL with Slicing [4] Take Compressed image Apply LZW decompression Apply IPBL(Inverse PBL) technique of the respective leveling order to each pixel in the image Apply the Inverse bit slicing technique Consider the compressed image, applied inverse technique of LZW and inverse of PBL technique is applied with respect to their levels. In PBL technique with bit slicing one more inverse bit slicing is applied to get the original image. Retrieved image is same as original image but with loss data. 2.4. Procedure to compress using PBL technique //read the image Call READ_INPUT_FILE () //convert image to digital format Call CONVERT_INT () //applying PBL techniques Call PIXEL_BASED_LEVEL () //compress encoded file by compress method1 Call COMPRESS_METHOD () END Procedure for CONVERT_INT () Procedure for PIXEL_BASED_LEVEL () Procedure to compress using PBL technique with bit slicing. Procedure for PIXEL_BASED_LEVEL () Decompression Procedure for PBL_DECOMPRESS() 8

International Journal of Computer Engineering and Applications, Volume XI, Issue V, May 17, www.ijcea.com ISSN 2321-3469 Decompression Procedure for PBL_DECOMPRESS_BIT_SLICE() 2.5 Histogram and Statistical information of Original and Reconstructed image of LL-SPEST The proposed LL-SPEST techniques are applied on sample medical and nonmedical raw images to compress. The following Figures from 6 and 7 are showing the histogram and statistical information of original and reconstructed medical images. Below Figures are clearly showing the no difference in histogram and statistical information of original and reconstructed image. Hence we can conclude that LL-SPEST is pure lossless technique. The Figures 6 and 7 are showing histogram and statistical information non medical images. (a) Figure 6 : Histogram and statistical information of Original Baboon and reconstructed Lossless Baboon image (a) Figure 7 Histogram and statistical information of Original Lena and reconstructed Lossless Lena image 9

AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING 2.6. Histogram and Statistical information of Original and Reconstructed segmented image of LSY- SPEST The proposed LSY-SPEST techniques are applied on simple and segmented medical and nonmedical raw images to compress. The following Figures 8 and 9 are showing the histogram and statistical information of original and reconstructed lossy images. And these figures are showing the little difference in histogram and statistical information of original and reconstructed lossy medical and non medical image. Original Chest x-ray image and Reconstructed Lossy Chest x-ray image The Figures.8,9 medical images. are showing histogram and statistical information non (a) Figure 8 : Histogram and statistical information of Original Baboon and reconstructed Lossy Baboon image (a) 10

International Journal of Computer Engineering and Applications, Volume XI, Issue V, May 17, www.ijcea.com ISSN 2321-3469 Figure 9 : Histogram and statistical information of Original Lena and reconstructed Lossy Lena image Conclusion: The objective of image compression is to reduce size of image in order to reduce network traffic and improve transfer quality. So we have developed a new approach of image compression using PBL (Pixel Based Leveling)technique. The PBL technique is lossy compression technique to compress the images and suitable for digital still images.we have applied PBL technique in two approaches.firstly, PBL is directly applied on image and second is applied on bit slice images. Both approaches shows better compression than other lossy compression techniques. In the proposed PBL,leveling technique and LZW compression are applied on 8 level bit slice to reduce size of leveled image and up to 3 levels picture gives more pressure rate. Futher, decompression is also done on compressed image to get original images i.e,ipbl(inverse PBL) and LZW decompression are applied.the performance of proposed technique using histogram and statistical information are applied on sample medical and non medical raw images to compress. References [1] http://en.wikipedia.org/wiki/pixel_density, http://www.prepressure.com/ design/ basics/resolution. [2] Image Retrieval Using BIT-Plane Pixel Distribution by N S T Sai and R C Patil. International Journal of Computer Science & Information Technology (IJCSIT), Vol 3, No 3, June 2011. [3] Light-Weight Instruction Set Extensions for -Sliced Cryptography Philipp Grabher, Johann Großsch adl, and Dan Page. [4] International Journal of Information Sciences and Techniques (IJIST) Vol.2, No.4, July 2012 Design and Implementation of LZW Data Compression Algorithm Simrandeep Kaur, Student1 ; V. Sulochana Verma, Project Consultant. [5] Kris Popat And Dan S. Bloomberg. Two-Stage Lossy/Lossless Compression Of Grayscale Document Images. Mathematical Morphology And Its Applications To Image And Signal Processing Computational Imaging And Vision Volume 18, 2000, Pp 361-370. [6] S.Parveen Banu, Dr.Y.Venkataramani, "An Efficient Hybrid Image Compression Scheme based on Correlation of Pixels for Storage and Transmission of Images", International Journal of Computer Applications (0975 8887) Volume 18 No.3, March 2011. 11

AN IMAGE COMPRESSION TECHNIQUE USING PIXEL BASED LEVELING [7] Song Zhao, Yan Xu, Hengjian Li, Heng Yang. A Comparison Of Lossless Compression Methods For Palmprint Images. Journal Of Software, Vol. 7, No. 3, March 2012. [8] M.U. Celik et al., "Gray-level-embedded lossless image compression", Signal Processing: Image Communication 18 (2003) 443 454, Elsevier Science, doi:10.1016/s0923-5965(03)00023-7. [9] Mehmet Utku Celik, Gaurav Sharma, A. Murat Tekalp. Gray-Level-Embedded Lossless Image Compression. Elsevier, Signal Processing: Image Communication, 18 (2003) 443 454. Authors : Mr.S.Vijayanand is an research scholar in the department of Computer Science and Technology at S.KUniversity, Anantapur A.P. He has completed B.Tech from Madras University and M.Tech from Dr.MGR Educational & Research Institute and University. He has published 5 National and International publications. His research interests are in the field of Computer Networking and Image Processing. Mrs.B.Harichandana is research scholar in the department of Computer Science Technology at S.K.University, Anantapur. She acquired M.Sc in Computer Science from S.K. University, Anantapur. She has 10 years of experience in teaching.her research interest is in the field of Image Processing. Miss.K.Lavanya is research scholar in the department of Computer Science Technology at S.K.University, Anantapur. She acquired M.Sc in Computer Science from S.V. University, Tirupathi. She has 2 years of experience in teaching.her research interest is in the field of Image Processing. 12