Compresión de imágenes médicas usando la transformada wavelet de Haar
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Nacional de Trujillo
Abstract
En este trabajo se estudiaron dos tópicos matemáticos. La transformada wavelet de Haar, la cual es una técnica matemática utilizada para analizar y representar señales e imágenes a diferentes escalas y el estudio del umbralamiento, el cual es una técnica utilizada en procesamiento de señales e imágenes que consiste en establecer un valor límite (umbral) para diferenciar entre dos o mas estados o clases en los datos. Con estas dos técnicas se presenta un algoritmo para comprimir imágenes medicas en escala de grises; en donde se realiza una descomposición bidimensional en 4 niveles; luego, se aplica un umbralamiento duro para eliminar coeficientes menos significativos; y finalmente, se lleva a cabo la reconstrucción de la imagen comprimida a partir de los coeficientes truncados
Abstract Two mathematical topics were studied in this work. The Haar wavelet transform, which is a mathematical technique used to analyze and represent signals and images at different scales, and the study of thresholding, which is a technique used in signal and image processing that consists of establishing a limit value (threshold) to differentiate between two or more states or classes in the data. With these two techniques, an algorithm for compressing medical images in gray scale is presented; where a two-dimensional decomposition in 4 levels is performed; then, a hard thresholding is applied to eliminate less significant coefficients; and finally, the reconstruction of the compressed image is carried out from the truncated coefficients
Abstract Two mathematical topics were studied in this work. The Haar wavelet transform, which is a mathematical technique used to analyze and represent signals and images at different scales, and the study of thresholding, which is a technique used in signal and image processing that consists of establishing a limit value (threshold) to differentiate between two or more states or classes in the data. With these two techniques, an algorithm for compressing medical images in gray scale is presented; where a two-dimensional decomposition in 4 levels is performed; then, a hard thresholding is applied to eliminate less significant coefficients; and finally, the reconstruction of the compressed image is carried out from the truncated coefficients
Description
Keywords
transformada wavelet, wavelet de Haar, umbralamiento, análisis autorresolución