Title :
Texture analysis techniques for the classification of microcalcifications in digitised mammograms
Author :
Kramer, Dani ; Aghdasi, Farzin
Author_Institution :
Dept. of Electr. Eng., Univ. of the Witwatersrand, Johannesburg, South Africa
Abstract :
We present various image texture analysis techniques for the classification of microcalcifications in digitised mammograms. Three categories of image texture features are extracted from the microcalcifications. The set of features based on a combined statistical and multiresolution approach to texture analysis gave the best results. Both a k-nn classifier and an artificial neural network are used for classification
Keywords :
cancer; feature extraction; feedforward neural nets; image classification; image resolution; image texture; mammography; medical image processing; statistical analysis; artificial neural network; digitised mammograms; features; image texture analysis techniques; k-nn classifier; microcalcification classification; multiresolution approach; statistical approach; Biopsy; Breast cancer; Diagnostic radiography; Feature extraction; Image segmentation; Image texture; Image texture analysis; Lesions; Statistical analysis; Tumors;
Conference_Titel :
Africon, 1999 IEEE
Conference_Location :
Cape Town
Print_ISBN :
0-7803-5546-6
DOI :
10.1109/AFRCON.1999.820877
