The digital image processing is a widespread applicable technique especially in the area where tools are used for feature extraction and to obtain patterns of studied images. Initially segmentation is used to separate the image into parts that represent an interest object that can be used for further specific study. There are various techniques present which performs such task but a common technique that adapt to all images is required, especially for complex or specific images. Hence our project basically aims to obtain a technique which is convenient for complex and different images. We tend to obtain a more specific result of the input image using histogram quantization, calculating valleys from analysis of histogram slope percentage, calculating threshold using maximum entropy. This approach provides more specific results over the already proposed technique which will be of great importance to the doctors, pathologists and surgeons to detect the potential cell rejection.

[1] S. R. Aboud Neta, L. V. Dutra, G. J. Erthal1, “Limiarização automática em histogramas ultimodais”. Proceeedings of the 7th Brazilian Conference on Dynamics, Control and Applications, FCT – Unesp de Presidente Prudente, May, 2008.
[2] Otsu N A 1979 Threshold selection method from graylevel histogram IEEE Transactions on
[3] Pun T 1980 A New Method for Gray-Level Picture Thresholding Using the Entropy.

@article{yogesh02041007,
title = "Facial Feature Extraction Based on Local Color and Texture for Face Recognition using Neural Network ",
journal = "International Journal of Science and Engineering Applications (IJSEA)",
volume = "2",
number = "3",
pages = "87 - 89 ",
year = "2013",
author = "Mr.Yogesh pawar, Sahil Dhammani, Dnyaneshwar Wable, Mukund Baheti ",
}