• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
Ct31628631
 

Ct31628631

on

  • 164 views

IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com

IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com

Statistics

Views

Total Views
164
Views on SlideShare
164
Embed Views
0

Actions

Likes
0
Downloads
2
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Ct31628631 Ct31628631 Document Transcript

    • V.Meenakshi, N.Harathi, K.Sravya / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 3, Issue 1, January -February 2013, pp.628-631Single Scale Decomposition And Improving Contrast Of Satellite Images Using Svd V.Meenakshi1, N.Harathi2, K.Sravya3 1 Asst.Professor Dept.of EIE, Sree Vidyanikethan Engineering College, Tirupati, 2Asst.Professor Dept.of EIE, Sree Vidyanikethan Engineering College, Tirupati.3Asst.Professor Dept.of ECE, Sree Vidyanikethan Engineering College, TirupatiAbstract— In this letter, a new satellite image therefore, we can perceive the world similarlycontrast enhancement technique based on the regardless of the considerable changes indiscrete wavelet transform (DWT) and singular illumination conditions.value decomposition has been proposed. The If the contrast of an image is highly concentrated ontechnique decomposes the input image into the a specific range, the information may be lost infour frequency sub bands by using DWT and those areas which are excessively and uniformlyestimates the singular value matrix of the low– concentrated. The problem is to optimize thelow sub band image, and, then, it reconstructs contrast of an image in order to represent all thethe enhanced image by applying inverse DWT. information in the input image. There have beenThe technique is compared with conventional several Techniques to overcome this issue, such asimage equalization techniques such as standard general Histogram equalization (GHE) and localgeneral histogram equalization and local histogram equalization (LHE). In this letter, we arehistogram equalization, as well as state-of-the-art comparing our results with two State-of-the-arttechniques such as brightness preserving techniques, namely, brightness preserving dynamicdynamic histogram equalization and singular Histogram equalization (BPDHE) and ourvalue equalization. The experimental results previously In many image processing applications,show the superiority of the proposed method the GHE technique is one of the simplest and mostover conventional and state-of-the-art effective primitives for contrast Enhancement,techniques. which attempts to produce an output Histogram thatIndex Terms—Discrete wavelet transforms, is uniform. One of the disadvantages of GHE is thathistogram equalization, image equalization, the information laid on the histogram or probabilitysatellite image contrast enhancement distribution function (PDF) of the image will be lost. Demirel and Anbarjafari showed that the PDF ofI. INTRODUCTION face images can be used for face recognition; hence, Satellite images are used in many preserving the shape of the PDF of an image is ofapplications such as Geosciences studies, vital importance. Techniques such as BPDHE orastronomy, and geographical information systems. SVE are preserving the general pattern of the PDFOne of the most important quality factors in satellite of an image. BPDHE is obtained from dynamicimages comes from its contrast. Contrast histogram specification Which generates theenhancement is frequently referred to as one of the specified histogram dynamically from the Inputmost important issues in image processing. Contrast image. The singular-value-based image equalizationis created by the difference in luminance reflected (SVE) technique, is based on equalizing the singularfrom two adjacent surfaces. In visual perception, value matrix Obtained by singular valuecontrast is determined by the difference in the color decomposition (SVD). SVD of an Image, which canand brightness of an object with other objects. Our be interpreted as a matrix, is written as Follows:visual system is more sensitive to contrast thanabsolute luminance; therefore, we can perceive the A  U A  A VA T (1)world similarly regardless of the considerablechanges in illumination conditions. If the contrast of Where UA and VA are orthogonal square matricesan image is highly concentrated on aspecific range, known as Hanger and aligner, respectively, and thethe information may be lost in those areas which are ΣA matrix contains The sorted singular values on itsexcessively and uniformly concentrated. The main diagonal. The idea of Using SVD for imageproblem is to optimize the contrast of an image in equalization comes from this fact that ΣA Containsorder to represent all and brightness of an object the intensity information of a given image. In ourwith other objects. Our visual system is more earlier work SVD was used to deal with ansensitive to contrast than absolute luminance; Illumination problem. The method uses the ratio of 628 | P a g e
    • V.Meenakshi, N.Harathi, K.Sravya / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 3, Issue 1, January -February 2013, pp.628-631  LLA   LLAthe largest Singular value of the generatednormalized matrix, with mean Zero and variance of ˆone, over a normalized image which can becalculated according to max  N    0,Var 1  LL A  U LLA  LLA VLLA (5) max   A  Now, the LLA, LHA, HLA, and HHA sub band (2) images of the original image are recombined by applying IDWT toWhere ΣN(μ=0,var=1) is the singular value matrix generateof the synthetic intensity matrix. This coefficient the resultant equalized image Acan be used to regenerate an equalized image using A = IDWT(LLA, LHA,HLA,HHA).(6) equalized A  U A    A VA T (3)II. PROPOSED IMAGE CONTRASTENHANCEMENT There are two significant parts of theproposed method. The first one is the use of SVD.As it was mentioned, the singular value matrixobtained by SVD contains the illuminationinformation. Therefore, changing the singular valueswill directly affect the illumination of the image;hence, the other information in the image will not bechanged. The second important aspect of this workis the application of DWT. As it was mentioned inSection I, the illumination information is embeddedin the LL sub band. The edges are concentrated inthe other sub bands (i.e., LH, HL, and HH). Hence,separating the high-frequency sub bands andapplying the illumination enhancement in the LLsub band only will protect the edge informationfrom possible degradation. After reconstructing thefinal image by using IDWT, the resultant image willnot only be enhanced with respect to illuminationbut also will be sharper.The general procedure of the proposed technique isas follows.The input image A is first processed by using GHEto Generate Aˆ. Then, both of these images aretransformed by DWT into four sub band images.The correction coefficient For the singular valuematrix is calculated by using the following equation: max( LLAˆ )  max( LLA ) (4) We have used the db.9/7 wavelet function as the Mother function of the DWT. In the followingWhere ΣLLA is the LL singular value matrix of the section, the Experimental results and the comparisoninput image of the aforementioned conventional and state-of-the-and ΣLL ˆA is the LL singular value matrix of the art techniques are discussed..output of the GHE. The new LL image is composedby III.EXPERIMENTAL RESULTS AND DISCUSSIONS ORIGINAL IMAGE LOW CONTRAST IMAGE 629 | P a g e
    • V.Meenakshi, N.Harathi, K.Sravya / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 3, Issue 1, January -February 2013, pp.628-631 The grey level distribution of general Histogram equalization was not clearly distributed over the entire scale as shown in fig. ADAPTIVE HISTOGRAM EQUALIZED IMAGE:Figure:1 Original Low Contrast Image Figure:3 Adaptive Histogram Equalized ImageFigure:1.1 Original Low Contrast HistogramThe fig shows the grey level distribution of lowcontrast image. By using this grey level distribution,the contrast adjustments can be studied. The Figure:3.1 Adaptive Histogram EqualizedHISTOGRAM EQUALIZED: Histogram grey level distribution of adaptive Histogram equalization was improved when compared to the general Histogram method as shown in fig. But Probability distribution function (PDF) is not Clearly achieved SVD ENHANCED IMAGE:Figure:2 Histogram Equalized Image Figure:4 SVD enhanced imageFigure:2 .1Histogram Equalized Histogram 630 | P a g e
    • V.Meenakshi, N.Harathi, K.Sravya / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 3, Issue 1, January -February 2013, pp.628-631 DetectionEstimation and Data Analysis. Hoboken, NJ: Wiley, 1988. 6. J. W. Wang and W. Y. Chen, “Eye detection based on head Contourgeometry and wavelet subband Projection,” Opt. Eng., vol. 45, no. 5,pp. 057001-1–057001-12, May 2006. ABOUT THE AUTHORS Mrs. V. Meenakshi received B Tech degreeFigure:4.1 SVD enhanced Histogram from SV University and received her M The peak values from the grey level distribution of Tech degree from JNTSVD enhanced image histogram are almost University, Hyderabad.identical to the peak Values of the original image Her areas of interestover the entire scale. include signal & image processing andIV.CONCLUSION Communications. She is a life A new satellite image contrast enhancement member of ISTE & IETE.technique based on DWT and SVD was proposed. Mrs. N. HarathiSingular Value Decomposition is employed to received B Tech degreeaddress the contrast variation of Images. The Visual from JNT UniversityQuality of images is superior to wavelet Coefficient and is pursuing her Mfiltering & GHE Qualitative Metrics (Information Tech degree from JNTTheory) are computed and can be used to improve University, Hyderabad.decision making and pattern recognition l. Her areas of interestapplications. include signal & image The proposed technique decomposed the processing andinput image into the DWT sub bands, and, after ARM Processors and PLCs.updating the singular value matrix of the LL sub She is a life member of ISTE.band, it reconstructed the image by using IDWT. Ms. K. SravyaThe technique was compared with the GHE, LHE, received B Tech degreeBPDHE, and SVE techniques. The visual results on from JNT Universitythe final image quality show the superiority of the and received her Mproposed method over the conventional and the Tech degree from JNTstate-of-the-art techniques. University, Ananathapur. . Her areas of interestREFERENCES include signal & image1. R. C. Gonzalez and R. E. Woods, Digital Image processing and Processing. Englewood Cliffs, NJ: Prentice- Communications. She is a life Hall, 2007. member of ISTE.2. T. K. Kim, J. K. Paik, and B. S. Kang, “Contrast enhancement system using spatially adaptive histogram equalization with temporal filtering,”IEEE Trans. Consum. Electron.,vol. 44, no. 1 , pp. 82–87, Feb. 1998.3. H. Ibrahim and N. S. P. Kong, “Brightness preserving dynamic histogram equalization for image contrast enhancement,” IEEE Trans. Consum. Electron., vol. 53, no. 4, pp. 1752– 1758, Nov. 2007.4. C. C. Sun, S. J. Ruan, M. C. Shie, and T. W. Pai, “Dynamic contrast enhancement based on histogram specification,” IEEE Trans. Consum. Electron., vol. 51, no. 4, pp. 1300–1305, Nov. 2005.5. K. S. Shanmugan and A. M. Breipohl, Random Signals: 631 | P a g e