2D Gradient Algorithms for Noise Reduction in Radiological Images

In areas such as biomedical image processing, the techniques or methods used to recover the content in noise-contaminated signals are essential. One of them has been adaptive filtering, which, by adjusting to the desired signal through real-time updating of the coefficients, allows improvement and deconvolution in the recovery of degraded or contaminated images, attracting the attention of researchers in inverse problems. In this paper, the 2D-AR  gradient algorithm is used in noise reduction in dental radiological images, for which simulations are performed to obtain the best configuration of the hyperparameters, and a statistical analysis of the values obtained is performed. Based on the simulation results and the established metrics, it is demonstrated that the algorithm achieves a slightly higher noise reduction than the other 2D gradient algorithms (LMS and NLMS).

• Fecha de publicación: 2023
• Institución: Universidad Pedagógica y Tecnológica de Colombia
• Repositorio: RiUPTC: Repositorio Institucional UPTC
• Idioma: eng
• OAI Identifier: oai:repositorio.uptc.edu.co:001/14376
• Acceso en línea: https://revistas.uptc.edu.co/index.php/ingenieria/article/view/16178; https://repositorio.uptc.edu.co/handle/001/14376
• Palabra clave: 2D adaptive filter; Noise cancellation; Signal processing; Radiological images; Gradient algorithm; 2D filtro adaptativo; Cancelación de Ruido; Procesamiento de señales; Imágenes Radiológicas; Algoritmo Gradiente
• License: Copyright (c) 2023 Jhonatan Collazos-Ramírez, Pablo-Emilio Jojoa, Juan-Pablo Hoyos