Chaotic Differential-Evolution-Based Fuzzy Contrast Stretching Method

Introduction

In the realm of digital image processing all the indigenous algorithms must be cognizant of the two main aspects viz. contrast enhancement and brightness preservation of the images experimented. The endeavour of the enhancement algorithms is to achieve an optimal condition using an objective function where the image attains a state of maximum clarity such that it can have a good visual analysis. Then only it can be differentiated from the original image having poor contrast and other technical anomalies.

The first ever approach is to achieve contrast enhancement was Histogram Equalization (HE) technique (Gonzalez, R.C., Woods, R.E. (2002)). HE based techniques have been used in medical image processing, satellite image processing etc. Basically HE procedure flattens the histogram of the original image. Theoretically the entire grey levels are distributed with uniform distribution. As a result of this it improves the contrast of the image, maximizes the image entropy. As the histogram of the output image is uniformly distributed the mean brightness is approximately changed to the middle of the grey level regardless of the mean of the input image (Chen, S. D., Ramli, A. R. (2004), Kim, Y.T. (1997)). To overcome the flattering effect of the histogram in HE method which sometimes results in giving washed-out images, unnatural enhancement and also some undesirable artefacts a new procedure was put forward by Kim (Kim, Y.T. (1997)) known as Brightness Preserving Bi-Histogram Equalization (BBHE). In BBHE, histogram of the image was separated around its mean and then the two divided parts were equalized separately (Chen, S. D., Ramli, A. R. (2004), Shanmugavadivu, P., Balasubramanian, K., Muruganandam, A. (2014)). In Dualistic Sub-Image Histogram Equalization (DSIHE) proposed by Wan (Chen, S. D., Ramli, A. R. (2004), Shanmugavadivu, P., Balasubramanian, K., Muruganandam, A. (2014)), the procedure was same as BBHE, but the histogram was separated by median instead of mean (Chen, S. D., Ramli, A. R. (2004), Shanmugavadivu, P., Balasubramanian, K., Muruganandam, A. (2014), Chen, S. D., Ramli, A. R. (2003), Chen, S. D., Ramli, A. R. (2003)). Chen and Ramli proposed Minimum Mean Brightness Error Bi-Histogram Equalization (MMBEBHE) method (Chen, S. D., Ramli, A. R. (2004), Chen, S. D., Ramli, A. R. (2003)), in which the histogram was separated using a specified threshold which preserved the minimum mean brightness error between input and output images and then the two parts were equalized independently. This technique was better than BBHE and DSIHE, but it still suffered from deficiency of contrast and brightness (Shanmugavadivu, P., Balasubramanian, K., Muruganandam, A. (2014)). In literature, parameterized contrast stretching function was also reported for image enhancement and produced better result than traditional HE (Gorai, A., Ghosh, A. (2009), Gorai, A., Ghosh, A. (2011), Barik, M., Sheta, A., Ayesh, A. (2007), Dhal, K.G, Quraishi, I. M., Das, S. (2015), Dhal, K., G., Das, S. (2015), Dhal, K.G, Quraishi, I. M., Das, S. (2015), Dhal, K.G, Quraishi, I. M., Das, S. (2015)) . The optimal values of these parameters had been found by employing different metaheuristic algorithms with the assistance of properly selected objective function (Gorai, A., Ghosh, A. (2009), Gorai, A., Ghosh, A. (2011), Barik, M., Sheta, A., Ayesh, A. (2007), Dhal, K.G, Quraishi, I. M., Das, S. (2015), Dhal, K., G., Das, S. (2015), Dhal, K.G, Quraishi, I. M., Das, S. (2015), Dhal, K.G, Quraishi, I. M., Das, S. (2015)) . In this study, one novel parameterized fuzzy based bi-histogram contrast stretching function has been employed for enhancing the grey as well as color images and optimal parameters are computed by formulating the enhancement problem as optimization problem.