Efficient Color Cast Correction Based on Fuzzy Logic

Abstract

Color cast defect detection of an image is challenging and of great importance in the field of image quality evaluation. This paper presents a simple and less complex method to improve the visibility in color cast images. An objective measure called color cast could be independent of the illumination condition. There are different assumptions present in the literature to provide an estimate of the Red cast, Blue cast and Green cast present in the image. These assumptions sometimes failed to eliminate the effect of color cast under varying illumination condition. The proposed method performs well under these conditions which may involve two steps. Firstly it performs the color cast detection and thereafter, a power law transformation is introduced to achieve the true color distribution of the image. In our approach a new transformation method which corrects the color cast as well as improves the quality of image. The uncertainty in the color distribution due to the illumination is removed on the basis of fuzzy Gaussian membership functions and further, improves by using weighing distribution. The performance analysis of the proposed approach is carried out by using two objective measures the image distance metric criteria of CIELAB color space and the image quality assessment function. The effectiveness of the proposed approach is illustrated by minimizing the image distance between the reference image and corrected image. The computed higher value of the image quality assessment function obtained in the proposed algorithm application on images, also justifies that the proposed algorithm results the better quality of the image in comparison to existing methods.

Remote sensing image segmentation using feature based fusion on FCM clustering algorithm

“…Remote sensing image interpretation is a challenging task for researchers that are continually working hard but additionally need to improve the quality of advanced algorithms and software to comprehend and accurately detect the region. These constraints of remote sensing image enhancement frequently include factors beyond human control, and it is also challenging to accomplish because of the existence of so many dense regions, and the limitations of human vision [6]. The segmentation approach may deal with the problem of pre-cisely determining a region in a remote sensing image.…”

Generalized Parametric Exponential Entropy Measure of degree on Fuzzy Set

“…Bhandari and Pal [5] presented two fuzzy entropy measures, out of which one is exponential entropy. Kapur [6], Verma and Sharma [7], Hooda [8] and Fan and Ma [9] defined some new fuzzy entropy measure corresponding to probabilistic measure. Tomar and Anshu [10][11][12][13] proposed some entropy measure on fuzzy set.…”