Image enhancement

3Picture Summary

Dynamic Histogram equalization:

ps0

Gray level range allocation to sub-histograms:

ps1

Full description

Our key observation is to eliminate the domination of higher histogram components on lower histogram components in the image histogram and to control the amount of stretching of gray levels for reasonable enhancement of the image features. In an image there can be several dominating portions. These are separated in sub-histograms. Each sub-histogram receives a dynamic range. This range is used to map the corresponding sub-histogram while applying Histogram Equalization (HE). In this way small features of the image cannot become dominant. It also prevents image getting washed out. We calculate the dynamic range and weight with the following equations.
\(S_i\): The summation of all histogram values of \( i \)th sub-histogram
\( X \): The coefficient to control the strength of image contrast
In summary the method can be divided in three main steps

  1. Selecting sub-histograms
  2. Select each sub-histogram’s dynamic range
  3. Applying Histogram Equalization to each sub-histogram

 

By changing the value of \(x\) in equation (2) we can control the strength of contrast enhancement. Here we show the experimental result of histogram enhancement for different values of \(x\). Also we show result of a traditional Global Histogram Equalization (GHE) for comparison purpose.
TXSzIxuThvemQevKvFyl2rRcr2ktrbgCaamz-Pph1seKRhjmkciRyX23w5_5_EQmApry3soNTSlTU6-e-KbgwcouPEQaCyAPpPx4OEEZkc_jZrTpgAE* xS83uaf2c_I1Xdt6_1yb8Q1iUGptFIiEXvYCyjR40mcBQFk28uUpk4Y_TO-oEmyKU5Zq91iubhvUnsWfjMz5F2pILlJdc4o_BBvflnuP4StEtcLsqI4*

(a)

(b)

tFWxdkUVSHU-LLXflIw4a2eUkKp5e0TNSMev8q6xYUvbBMkR7GCA64RjODIKL5x9fXdJ2GWGZ3ikYQ2Zw4NlEePxZPnjmT27J-g3AsyGLhbr8HS7W9E* sIVfCNJjPyGHDAtPuBGINDDMMfeiTV4Mt8cM7GIjqtkFTRobi_JjlWWXAoSnL4LuG71pqImR81shKYSHOIrfaK24vV4Wseml2axgd3MjQsLCik1wDKE*

(c)

(d)

mQ4NfedpmjqWBCBRV6OMRBB1SqpQaPvYsFEJhF51upuQYEZ5hnXwZMkGclxQ20VIP7T0XfYNKpCgDcIo_VLJMtz9bDehbzS6wRl2DFi0ahy6NpedA6o* hyWZryO9xa_iiCj5D-pd3ylOjdc8r7rQlMYU2QQNVXbaoiv3cbOXbqO6saXmV2eNyZkrKPnY7s4yx4VuI_5LuOdCLbgxRMnr4XsvG1pcIuulkhtxAa0*

(e)

(f)

5a0EDBwS_LScPEsckQblhwxJjADvQM4znreFvaz4Bafrpz_Z3_bGqoeMlHsqlVYou2RjocmwfbEfM4_O2hTVa7J2jeZJ2kHRUU2SnQDJn-DGKxWTE3A* q18Sh_mf7WdN5Hh2LTZtMFGyICCJ8qONW17LdRdmy-gISCnBTcg2uP-OTsp8-AimidDnlB4Fij0e1pxqHcL6NXPBntNmQ0SfWyBN2sO8y973E_i_nws*

(g)

(h)

Fig. Simulation results using a natural image. (a) Original image, (b) GHEed image, (c) BBHEed image, (d) RMSHEed image (\(r= 2\)), (e)-(h) DHEed image ( \( x= 0\), \(1\), \(2\), \(4\), accordingly).

 

References

SCI/SCIE Indexed Journal:
▷ Md. Hasanul Kabir, M. Abdullah-Al-Wadud, and Oksam Chae, “Brightness Preserving Image Contrast Enhancement using Weighted Mixture of Global and Local Transformation Functions”, International Arab Journal of Information Technology (IAJIT).(SCIE)
▷ M. Abdullah-Al- Wadud, Md. Hasanul Kabir, M. Ali Akber Dewan, and Oksam Chae, “A Dynamic Histogram Equalization for Image Contrast Enhancement”, IEEE Transactions on Consumer Electronics, Vol. 53, Issue 2, pp. 593-600, May 2007. (SCI)
▷ Md. Hasanul Kabir, M. Abdullah-Al-Wadud, Oksam Chae “Image Contrast Enhancement Based on Block-Wise Intensity-Pair Distribution with Two Expansion Forces”, Springer Verlag Lecture Notes in Computer Science, Volume. 4225, pp. 247-256, November 2006 (SCIE)

International Conference:
▷ Md. Hasanul Kabir, M. Abdullah-Al-Wadud, and Oksam Chae, “Global and Local Transformation Function Mixture for Image Contrast Enhancement”, IEEE International Conference on Consumer Electronics (ICCE), pp. 185-186, Jan 2009.
▷ M. Abdullah-Al-Wadud, Md. Hasanul Kabir, Oksam Chae, “A Spatially Controlled Histogram Equalization for Image Enhancement”, In proc. of the 23rd IEEE International Symposium on Computer and Information Sciences ( ISCIS), Istanbul, Turkey, 2008.

Comments are closed.