Title :
Tanning Leather Classification using an Improved Statistical Geometrical Feature Method
Author :
Wang, Ling ; Liu, Chang
Author_Institution :
Sichuan Normal Univ., Chengdu
Abstract :
Tanning is an important procedure in the leather production process. Tanning agent can change the physical properties and chemical compositions of leather, and ultimately alter the texture of the surface. In order to classify chrome tanning and vegetable tanning leather, this paper presents improved statistical geometrical features (ISGF) to proceed texture analysis for tanning leather and use two classifiers based on Fisher criterion and LVQ Net to compare chrome tanning and vegetable tanning leather, Our results suggest that ISGF can discriminate chrome tanning and vegetable tanning leather with less computation cost and more fit for tanning leather texture feature than SGF.
Keywords :
feature extraction; image classification; image texture; leather industry; production engineering computing; statistical analysis; tanning; Fisher criterion; LVQ Net; chemical composition; chrome tanning leather; leather production; statistical geometrical feature; surface texture; tanning agent; tanning leather classification; texture analysis; vegetable tanning leather; Chemicals; Cybernetics; Frequency; Gray-scale; Image coding; Machine learning; Noise reduction; Pixel; Production; Surface texture; Compactness; Fisher criterion; Irregularity; LVQ Net; Leather texture; Statistical geometrical features;
Conference_Titel :
Machine Learning and Cybernetics, 2007 International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-0973-0
Electronic_ISBN :
978-1-4244-0973-0
DOI :
10.1109/ICMLC.2007.4370433