A Local Type-2 Fuzzy Set Based Technique For He Stain Image Enhancement

Dibya Jyoti Bora

the Assam Kaziranga University

Rubul Kumar Bania

NEHU,Meghalaya

Che Ngoc

Ton Duc Thang University

The proposed approach in the paper comes under “Advanced Soft Computing Based Medical Image Processing Research” and the work has been conducted by Dr. Dibya Jyoti Bora (Assistant Professor), School of Computing Sciences, The Assam Kaziranga University, Jorhat, Assam in the year 2018-2019.

Introduction: HE stain images, although considered as the golden standard for medical image diagnosis, are still found to suffer from poor contrast and degradation in color quality. In this paper, a Type-2 fuzzy set-based enhancement technique is proposed for HE stain image enhancement with special care towards color-based computations and measurements.

Methods: This paper introduces a new approach based on Type-2 fuzzy set for HE stain image enhancement where Bicubic Interpolation plays an important part. Unsharp Masking is also employed as a post enhancement factor.

Results: From the results, it is clearly visible that cell nuclei and other cell bodies are easily distinguishable from each other in the enhanced result produced by our proposed approach. It implies that vagueness in the edges surrounding the objects in the original image is removed to an acceptable level.

Conclusions: The proposed approach is found to be, through both subjective and objective evaluations, an efficient preprocessing technique for a better HE stain image analysis.

Originality: The ideas involved in this paper are original. If work by other researchers are mentioned in any part of the paper, then they are cited properly.

Limitation: The relatively high time complexity is the only limitation associated with the proposed approach.

Keywords: Bicubic Interpolation, HE stains image, color image processing, color image enhancement, HSV color space, Type-2 fuzzy set
Published
2019-09-16
Downloads
Metrics
Metrics Loading ...
https://plu.mx/plum/a/?doi=10.16925/2357-6014.2019.03.02