• DocumentCode
    3746542
  • Title

    The optimization and design of the auto-exposure algorithm based on image entropy

  • Author

    Jingyi Ning;Tiejun Lu;Liyan Liu;Liye Guo;Xiaofeng Jin

  • Author_Institution
    Beijing Microelectronics Tech. Institute (BMTI) No.2. Siyingmen N. Rd Donggaodi Fengtai District, Beijing, China, 100076
  • fYear
    2015
  • Firstpage
    1020
  • Lastpage
    1025
  • Abstract
    In order to make the CMOS image sensor exposure automatically and correctly in different environment, this paper presents an improved algorithm based on image entropy. Image entropy is the information measure of image. Thus, with the purpose of getting the most appropriate exposure time, we can search the maximum information entropy. What´s more, by using the formula manipulation of image entropy and the piecewise linearization of the log function, the algorithm grasps the overall change rule and avoids the traditional precise calculation, which can save the computing time greatly and reduce the use of the resource. Meanwhile, simulation results demonstrated the effectiveness of the algorithm, which leaves a strong foundation for the hardware realization of the CMOS image sensor auto-exposure.
  • Keywords
    "Entropy","Gray-scale","Hardware","Signal processing algorithms","IP networks","CMOS image sensors","Approximation algorithms"
  • Publisher
    ieee
  • Conference_Titel
    Image and Signal Processing (CISP), 2015 8th International Congress on
  • Type

    conf

  • DOI
    10.1109/CISP.2015.7408029
  • Filename
    7408029