♣ digital image processing can transform one image domain into another domain ♣ large structures produce low frequencies (represent contrast information in image) ♣ small structures within patient produce high frequencies (represent detail in image) O frequency= # of cycles per unit length (# of times a signal changes per unit length) images displayed for viewing spatial location domain, numeric images and matrix of pixels.O Spatial location domain (radiography & CT) Images represented in 2 domains (depend on acquisition).Hence, image manipulated by spatial filter emphasis/de-emphasis on specific ranges of spatial frequency Image domain.Low spatial frequency information= larger objects, changes in image darkness.O Also describes major component of image noise High spatial frequency information= small things (pulmonary nodules, markings and bone trabecular patterns).Spatial frequency= # of times object can be repeated per unit distance (line pairs per mm size, sharpness), and amplitude (exposure value).Image recomposed (multi-resolution reconstruction) How is the image produced?.Adjustments made: contrast attenuated/enhanced, edges sharpened, noise minimised.Image split into different frequency components (multi-resolution decomposition).Parameter set for image derived and calculated from raw image.Automated multi-resolution image enhancement software.
0 kommentar(er)
