Gray Scale Manipulation
This is the simplest form of enhancement in which the transformation operator T directly acts only on the target pixel of the input image such that F’ (x, y) depends only on that pixel. Thresholding is yet another simple technique, wherein the intensity profile of the input image is replaced by a step function, which is controlled by chosen threshold value. In this approach, the intensities of the input image pixels which are below and above the threshold value are respectively mapped to the minimum and maximum intensity of the output’s dynamic range. Other gray scale transformations are outlined in Fig. 1
Histogram Equalization (HE) is one of the most widely used contrast enhancement technique, due to its simplicity and effectiveness. The HE techniques construct the linear cumulative histogram of an input image and redistribute its pixel values over its dynamic intensity range. HE-based enhancement techniques find applications in consumer electronics, medical image processing, speech recognition, texture synthesis, satellite image processing, etc. The traditional histogram equalization technique is described below:
Consider an input image, F (i, j) with a total number of ‘n’ pixels in the gray level range [X0, XN-1]. The probability density function P(rk ) for the level rkis given by:
where, nk represents the number of times the level rk appears in the input image, ‘n’ is the total number of pixels in the image and k = 0, 1, … , N-1. A plot of nk against rk is known as histogram of the image F. Based on above Eqn. , the cumulative density function is calculated as:
HE maps an image into the entire dynamic range, [X0, XN-1] using the cumulative density function which is given as:
Thus, HE flattens the histogram of an image which results in a significant change in its brightness (Fig. 2).
Fig. 2 Woman Image: (a) Original and its Histogram(b) HEed Image and its Histogram