Adaptive Intuitionistic Fuzzy Enhancement of Brain Tumor MR Images

Image enhancement techniques are able to improve the contrast and visual quality of magnetic resonance (MR) images. However, conventional methods cannot make up some deficiencies encountered by respective brain tumor MR imaging modes. In this paper, we propose an adaptive intuitionistic fuzzy sets-based scheme, called as AIFE, which takes information provided from different MR acquisitions and tries to enhance the normal and abnormal structural regions of the brain while displaying the enhanced results as a single image. The AIFE scheme firstly separates an input image into several sub images, then divides each sub image into object and background areas. After that, different novel fuzzification, hyperbolization and defuzzification operations are implemented on each object/background area, and finally an enhanced result is achieved via nonlinear fusion operators. The fuzzy implementations can be processed in parallel. Real data experiments demonstrate that the AIFE scheme is not only effectively useful to have information from images acquired with different MR sequences fused in a single image, but also has better enhancement performance when compared to conventional baseline algorithms. This indicates that the proposed AIFE scheme has potential for improving the detection and diagnosis of brain tumors.

FUZZY MEASURE BASED ADAPTIVE METHODS FOR ILLUMINATION INVARIANT FACE RECOGNITION

Varying illumination is one of the well known and challenging problems in Face Recognition applications. Numerous methods have been proposed by researchers, but recognition performance under complex illumination is not yet satisfactory. The paper presents Fuzzy based methods to adaptively normalize illumination in face images for Face Recognition under varying illumination conditions. The paper has main two contributions: (1) Fuzzy measure based Adaptive Single-scale Retinex and (2) Fuzzy measure based Adaptive Single-scale Self Quotient Image method. Also, two more variations of these methods are presented. There are two main advantages of these methods, as compared to multi-scale Retinex and Self Quotient methods. Firstly, due to the adaptive nature of proposed methods, discontinuity in facial feature is smoothed and discontinuity due to shadows is preserved and hence performance is better. Secondly, computational complexity is reduced because of single scale 3∗3 filter instead of multi-scale filters. Rigorous experiments have been performed on CMU PIE face database and Extended Yale B face database. For determining False Acceptance Rate, 529 and 550 imposter faces are used for experiments on PIE and Yale databases respectively. Proposed methods are compared with existing methods under same experimental setup using six performance evaluation parameters. Results have shown that Fuzzy measure based methods performs well.

Fuzzy entropy based interactive enhancement of radiographic images

In certain radiographic images, fuzziness exists due to the vague nature of image characteristics and limitations of visual perception. Accordingly, a gap exists between the information content of an image and the information that can be retrieved. In this paper, a new fuzzy logic based intensification operator has been proposed for enhancement of images. For applying the operator effectively, selection of crossover point is critical. In this paper, the concept of fuzzy entropy has been proposed for objective selection of crossover point. Owing to semantic nature of information content, the methodology has a provision to change levels of enhancement interactively, to help in retrieving the information as required. The effectiveness of the methodology has been demonstrated. Comparison of the results with the Zadeh's INT operator and conventional histogram equalization techniques has established its superiority.

An optimal fuzzy system for color image enhancement

A Gaussian membership function is proposed to fuzzify the image information in spatial domain. We introduce a global contrast intensification operator (GINT), which contains three parameters, viz., intensification parameter t, fuzzifier fh, and the crossover point mu(c), for enhancement of color images. We define fuzzy contrast-based quality factor Qf and entropy-based quality factor Qe and the corresponding visual factors for the desired appearance of images. By minimizing the fuzzy entropy of the image information with respect to these quality factors, the parameters t, fh, and mu(c) are calculated globally. By using the proposed technique, a visible improvement in the image quality is observed for under exposed images, as the entropy of the output image is decreased. The terminating criterion is decided by both the visual and quality factors. For over exposed and under plus over exposed images, the proposed fuzzification function needs to be modified by taking maximum intensity as the fourth parameter. The type of the images is indicated by the visual factor which is less than 1 for under exposed images and more than 1 for over exposed images.

A fuzzy operator for the enhancement of blurred and noisy images

Rule-based fuzzy operators are a novel class of operators specifically designed in order to apply the principles of approximate reasoning to digital image processing. This paper shows how a fuzzy operator that is able to perform detail sharpening but is insensitive to noise can be designed. The results obtainable by the proposed technique in the enhancement of a real image are presented.