Title :
The characterization of scanning noise and quantization on texture feature analysis
Author :
Martin, G. ; Pattichis, M.S.
Author_Institution :
Dept. of Electr. & Comput. Eng., New Mexico Univ., Albuquerque, NM, USA
Abstract :
The study of the effects of scanning on texture features is of great interest to computer-based screening systems. A mathematical model is developed for understanding how the original image gets distorted due to the contrast variability and geometric distortion inherent in the scanning process. Both quantitative and qualitative results (for sixty common texture features) are given.
Keywords :
feature extraction; image texture; medical image processing; optical distortion; quantisation (signal); analog radiograph images; computer-based screening systems; contrast variability; digital diagnostic tools; geometric distortion; mathematical model; quantization; scanning noise; texture feature analysis; Biomedical optical imaging; Charge coupled devices; Mathematical model; Optical distortion; Optical films; Optical sensors; Quality control; Quantization; Sensor arrays; Testing;
Conference_Titel :
Image Analysis and Interpretation, 2004. 6th IEEE Southwest Symposium on
Print_ISBN :
0-7803-8387-7
DOI :
10.1109/IAI.2004.1300964
