• DocumentCode
    3114245
  • Title

    A Modified Fuzzy Contrast Enhancement Technique for Retinal Images

  • Author

    Paranjape, Sandeep V. ; Ghosh, Sambuddha ; Ray, A.K. ; Chatterjee, Jyotirmoy ; Lande, Anil R.

  • Author_Institution
    Sch. of Med. Sci. & Technol., IIT Kharagpur, Kharagpur, India
  • fYear
    2015
  • fDate
    26-27 Feb. 2015
  • Firstpage
    892
  • Lastpage
    896
  • Abstract
    The retinal image contrast enhancement is most important when it comes to visual analysis for respective disease diagnosis and as an important prerequisite for further image analysis. The evaluation of images for contrast enhancement depends mostly on image quality and application at hand. Fuzzy inference system used to get more precise information although the irregularity and vagueness in an image. A rule-based Fuzzy Inference System (FIS) is proposed for contrast enhancement of retinal images in addition with the partial histogram. Gaussian membership function and Region of Interest are found to aid in obtaining prominent results. We have obtained promising results by applying proposed method with effective contrast enhancement on DRIVE database images, also on images provided by the Ophthalmologist.
  • Keywords
    Gaussian processes; fuzzy reasoning; fuzzy set theory; image enhancement; visual databases; DRIVE database images; FIS; Gaussian membership function; disease diagnosis; image quality; modified fuzzy contrast enhancement technique; ophthalmologist; partial histogram; region of interest; retinal images; rule-based fuzzy inference system; visual analysis; Bismuth; Electronic mail; Fuzzy logic; Fuzzy sets; Histograms; Image enhancement; Retina; Contrast enhancement; Fuzzy method; Histogram processing; Human retinal images;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computing Communication Control and Automation (ICCUBEA), 2015 International Conference on
  • Conference_Location
    Pune
  • Type

    conf

  • DOI
    10.1109/ICCUBEA.2015.177
  • Filename
    7155975