Author_Institution :
Siemens Corp. Res. Inc., Princeton, NJ, USA
Abstract :
How to deal with variations of image contents, which contributes to uncertainty about shapes, locations, and boundaries of objects of interest in the image, is one of key issues in solving real-world image analysis and processing problems. A method is presented to automatically identify the target objects from images, with variations, low contrast, low resolution, and low signal-to-noise ratio, for tracking, quantitatively measurement, and diagnosis
Keywords :
image recognition; medical image processing; beating heart; image analysis; image contents; low contrast; low resolution; low signal-to-noise ratio; snow-storm images; target objects; tracking; variations; Biomedical imaging; Computer vision; Heart; Image resolution; Image segmentation; Nuclear medicine; Shape; Signal resolution; Signal to noise ratio; Target tracking;
