Title :
Textural segmentation of medical images
Author_Institution :
Dept. of Electr. Eng., Massachusetts Univ., Lowell, MA, USA
Abstract :
The segmentation process is formulated as a configurational optimization problem that is solved using simulated annealing optimization methods. This allows information about the degree of correspondence between a candidate segment and an assumed textural model, as well as morphological information about the candidate segment to be used in the segmentation process. To apply this optimization method, a Gerchberg-Papoulis bandlimited extrapolation fractal estimation technique, capable of processing irregularly shaped candidate segments, is utilized.
Keywords :
fractals; image segmentation; image texture; mathematical morphology; medical image processing; simulated annealing; Gerchberg-Papoulis bandlimited extrapolation fractal estimation technique; configurational optimization problem; irregularly shaped candidate segments; medical images; morphological information; radiographs; simulated annealing; textural segmentation; Biomedical imaging; Bones; Fractals; Image segmentation; Optimization methods; Partitioning algorithms; Pixel; Radiography; Simulated annealing; Temperature control;
Conference_Titel :
Bioengineering Conference, 1992., Proceedings of the 1992 Eighteenth IEEE Annual Northeast
Conference_Location :
Kingston, RI, USA
Print_ISBN :
0-7803-0902-2
DOI :
10.1109/NEBC.1992.285990