A fuzzy histogram weighting method for efficient image contrast enhancement
- PDF / 4,518,019 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 18 Downloads / 212 Views
A fuzzy histogram weighting method for efficient image contrast enhancement Sedighe Mirbolouk 1 & Morteza Valizadeh 1 & Mehdi Chehel Amirani 1 & Mohammad Amin Choukali 1 Received: 15 October 2019 / Revised: 25 August 2020 / Accepted: 2 September 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Image contrast enhancement is an important step in digital image processing applications. In this paper, we present an efficient contrast enhancement approach, which employs a histogram weighting method based on fuzzy system. It is able to enhance the contrast of input images while preserving their details. The proposed method divides the histogram of the original image into three sub-histograms using Fuzzy clustering. The obtained subhistograms are weighted based on the Mamdani Fuzzy inference system, and then they are summed to generate a new histogram. The produced histogram is modified to reduce undesirable effects of its spikes and pits. Finally, the enhanced image is obtained by equalization of the modified histogram. The Mamdani fuzzy inference system assigns an appropriate dynamic range to each input interval of gray levels (sub-histogram), hence enhancing the image details. Experimental results for different types of images verified the merit of the proposed method in terms of preservation the input image details and improving its contrast. Keywords Image enhancement . Fuzzy system . Histogram weighting . Fuzzy clustering
1 Introduction The aim of contrast enhancement techniques is bringing out the hidden details of image by increasing the gray level differences. It produces an output image that looks better than input image and leads to better visual interpretation. Contrast enhancement (CE) is an important preprocessing step in digital image processing [13]. It is widely used for fingerprint and face recognition [4, 19, 24, 29], medical image processing [12, 18, 31], satellite image enhancement [21, 43], video surveillance systems [11] and many others [5, 37]. Many contrast enhancement
* Morteza Valizadeh [email protected]
1
Department of Electrical and Computer Engineering, Urmia University, Urmia, Iran
Multimedia Tools and Applications
methods have been proposed in recent years. These methods can be broadly divided into two categories: direct enhancement [9, 15, 46] and indirect enhancement [2, 6, 35]. Direct enhancement algorithms define some predefined contrast measurement criteria and modify the gray levels of the image to improve these criteria. Indirect algorithms improve image contrast by redistributing the probability density of gray levels without defining any contrast measurements [2]. Indirect algorithms due to good performance and ease of implementation is pursued by the most researcher [2, 6]. They can be classified into two groups; transform domain techniques (like logarithmic and power-law transformations and so on) [35, 47] and histogram modification techniques (like histogram equalization (HE), histogram specification (HS), etc.) [7, 22]. Histogram
Data Loading...
