A Systematic Review of Noise Types, Denoising Methods, and Evaluation Metrics in Images | IEEE Conference Publication | IEEE Xplore

A Systematic Review of Noise Types, Denoising Methods, and Evaluation Metrics in Images


Abstract:

The removal of noise from images while retaining their quality and information is a crucial task in the field of image processing, and is commonly known as image denoisin...Show More

Abstract:

The removal of noise from images while retaining their quality and information is a crucial task in the field of image processing, and is commonly known as image denoising. As the use of digital images has become increasingly widespread in various fields, including medical imaging, astronomy, and microscopy, the problem of image noise has become more prominent. Image noise, which is any unwanted variation in brightness or color in an image, can significantly reduce the image quality and usefulness. To address this issue, various noise reduction techniques have been developed, such as filtering, wavelet denoising, and deep learning-based methods. However, selecting the appropriate denoising method for a specific type of noise and image can be challenging, and objective performance metrics are required to evaluate the effectiveness of different methods. Therefore, a thorough analysis of the various kinds of noise that can appear in digital images, the denoising techniques that are available, and the performance metrics that are used to assess them, is required. The goal of this review is to shed light on the importance of selecting the best denoising technique for the particular application at hand. This review will help researchers and practitioners choose the best denoising technique for their specific application by illuminating the benefits and drawbacks of various techniques.
Date of Conference: 08-11 November 2023
Date Added to IEEE Xplore: 25 December 2023
ISBN Information:
Conference Location: Kerala, India

Contact IEEE to Subscribe

References

References is not available for this document.