METHOD FOR REDUCING NOISE IN SEQUENCES OF FLUOROSCOPIC IMAGES

摘要

A method for reducing noise in flouroscopic images by detection of the traits in the curvelet domain, matching of the detected traits and time-domain filtering adapted to the type of coefficients associated with a trait. The method uses the discrete curvelet transforms of the images from a sequence. The denoised coefficients are detected, then transmitted to a step for locally matching the traits, the matched data subsequently undergo a step for time-domain filtering allowing the movement of the objects to be preserved in the image and a step for a 2D IDCT (Inverse Discrete Curvelet Transform) in order to produce the final image.

主权项

1. A method for reducing noise in a sequence of fluoroscopic images acquired by an X-ray detector, an image comprising one or more traits, an image being represented by curvelet coefficients ct,s,θsb,i,j, where s is the scale of the image, θsb its sub-band linked to the angle of a trait in the image and (i,j) its relative position, the method comprising in combination at least the following steps:
detection of the traits in an image in the curvelet domain, using the curvelet coefficients sub-band by sub-band and by determining the maximum response of a filter associated with a sub-band θsb, and the detected traits are stored in memory, locally matching the detected traits,
considering a set of curvelet coefficients, testing whether a curvelet coefficient being considered belongs to a trait detected at the time t or at an earlier time,if the curvelet coefficient belongs to a detected trait, testing whether the trait, to which the coefficient belongs, is matched to a trait detected at an earlier time, if this is the case, then the curvelet coefficient is categorized in a first set,if the coefficient is not detected, or if the curvelet coefficient belongs to a trait detected but non matched to a trait detected at an earlier time, then forming at least a second set,applying a first coefficient for time-domain filtering the curvelet coefficients belonging to the first set and at least a second coefficient for time-domain filtering the curvelet coefficients belonging to the second set,determining the denoised image using the time-domain filtered curvelet coefficients.