A Parallel Fuzzy Algorithm for Real-Time Medical Image Enhancement
- PDF / 2,579,659 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 93 Downloads / 195 Views
A Parallel Fuzzy Algorithm for Real-Time Medical Image Enhancement Josep Arnal1 • Mo´nica Chillaro´n2 • Estı´baliz Parcero3 • Luis B. Su´car1 Vicente Vidal2
•
Received: 14 December 2018 / Revised: 15 July 2020 / Accepted: 31 August 2020 Ó Taiwan Fuzzy Systems Association 2020
Abstract Medical images may be corrupted by noise. This noise affects the image quality and can obscure important information required for accurate diagnosis. Effectively apply filtering techniques can facilitate diagnosis or reduce radiation exposure. In this paper, we introduce a parallel method designed to reduce mixed Gaussian-impulse noise from digital images. The method uses fuzzy logic and the fuzzy peer group concept. Implementations of the method on multi-core interface using the open multi-processing (OpenMP) and on graphics processing units (GPUs) using CUDA are presented. Efficiency is measured in terms of execution time and in terms of MAE, PSNR and SSIM over medical images from the mini-MIAS database and over computed radiography (CR) images generated at different exposure levels. These images have been contaminated with impulsive and/or Gaussian noise. Experiments show that the proposed method obtains good performance in terms of the above mentioned objective quality measures. After applying multi-core and GPUs optimization strategies, the observed time shows that the new filter allows to remove mixed Gaussian-impulse noise in real-time.
& Josep Arnal [email protected] 1
Departamento de Ciencia de la Computacio´n e Inteligencia Artificial, Universidad de Alicante, San Vicente del Raspeig 03690, Spain
2
Departamento de Sistemas Informa´ticos y Computacio´n, Universitat Polite`cnica de Vale`ncia, Valencia 46022, Spain
3
Instituto de Tecnologı´as de la Informacio´n y Comunicaciones, Universitat Polite`cnica de Vale`ncia, Valencia 46022, Spain
Keywords Filter design Medical image processing Fuzzy logic Noise reduction
1 Introduction Filtering methods, i.e., techniques to detect and reduce noises, are essential in medical imaging (e.g., X-rays, magnetic resonance imaging (MRI), computer tomography (CT)) because the quality of the image can have repercussions on the diagnosis of a disease (for example, detecting microcalcifications in a mammogram). Moreover, noise reduction filters can be used to improve images when a reduced radiation dose is used [14, 15]. This fact is especially crucial in CT images in order to reduce the exposure to X-rays because the amount of radiation tends to be very high. Two specially common types of noise are the impulsive noise and the Gaussian noise. The impulsive noise is introduced during the transmission process and the Gaussian during the acquisition process [2, 27]. A large number of algorithms have been introduced to reduce either Gaussian (see e.g., [9, 10, 19, 27, 29, 38]) or impulse noise (see e.g., [3–5, 21, 23, 30–34, 39]). However, not all methods are useful when images are contaminated simultaneously with impulsive and Gaussian noise. A possible approach to address this problem is
Data Loading...
