Noise is inevitably common in digital images, leading to visual image
deterioration. Therefore, a suitable filtering method is required to lessen the
noise while preserving the image features (edges, corners, etc.). This paper
presents the efficient type-2 fuzzy weighted mean filter with an adaptive
threshold to remove the SAP noise. The present filter has two primary steps:
The first stage categorizes images as lightly, medium, and heavily corrupted
based on an adaptive threshold by comparing the M-ALD of processed pixels with
the upper and lower MF of the type-2 fuzzy identifier. The second stage
eliminates corrupted pixels by computing the appropriate weight using GMF with
the mean and variance of the uncorrupted pixels in the filter window.
Simulation results vividly show that the obtained denoised images preserve
image features, i.e., edges, corners, and other sharp structures, compared with
different filtering methods.
The article discusses the issue of noise in digital images and the need for an effective filtering method to reduce noise while preserving important image features such as edges and corners. The proposed method in this paper is the efficient type-2 fuzzy weighted mean filter with an adaptive threshold.
This filtering technique consists of two main steps. The first step involves categorizing the images into lightly, medium, or heavily corrupted based on an adaptive threshold. This threshold is determined by comparing the Modified Average Lower Deviation (M-ALD) of processed pixels with the upper and lower Membership Functions (MF) of the type-2 fuzzy identifier. This categorization step helps in determining the level of noise corruption in the image.
In the second step, corrupted pixels are eliminated using the appropriate weights computed using the Geometric Mean Fuzzification (GMF) with the mean and variance of the uncorrupted pixels within the filter window. This step aims to reduce noise in the image without affecting the important image features.
The results of the simulation demonstrate that the proposed filter effectively removes noise while preserving image features like edges, corners, and sharp structures. This is compared to different filtering methods, indicating its superiority in achieving this objective.
This research involves a multi-disciplinary approach by combining concepts from fuzzy logic and image processing. Fuzzy logic provides a robust framework for handling uncertainty and imprecision in the categorization of images based on noise corruption levels. The image processing techniques are used to compute appropriate weights and effectively eliminate corrupted pixels without compromising important image features.
Further research in this area can focus on optimizing the parameters of the proposed filter to enhance its performance. Additionally, exploring the application of this filter to real-world images and evaluating its effectiveness in different scenarios will be valuable. Advancements in machine learning and deep learning may also be incorporated to develop more sophisticated noise filtering techniques.