Wavelength Selection in Visible and Near Infrared Spectra for Detection of Bruises on Apples
Bruising is one of the most common causes for the degradation of fruit quality. Currently, the online apple grading for bruise detection is still unavailable, which results mainly from the complexity and difficulty of wavelength selection. In this study, three methods of wavelength selection, variable importance on the projection-based selection, genetic algorithms selection, and combination of the two methods, were studied and compared for the classification of bruised and normal areas of ‘Fuji’ apples. The results showed that all of the three sets of selected wavelengths performed better than the full-spectrum wavelengths, 380–1000 nm. The best predictive ability was obtained by the combination of the two methods, with lowest predicted classification error, 3%, and the fewest selected wavelengths, 20. The combination of the two methods, which combines the advantages of the two methods, and gets rid of the disadvantages of both, is effective for selecting the most informative wavelengths with little risk of over-fitting.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
Document Type: Research Article
Publication date: September 1, 2013
More about this publication?
- ADVANCED SCIENCE LETTERS is an international peer-reviewed journal with a very wide-ranging coverage, consolidates research activities in all areas of (1) Physical Sciences, (2) Biological Sciences, (3) Mathematical Sciences, (4) Engineering, (5) Computer and Information Sciences, and (6) Geosciences to publish original short communications, full research papers and timely brief (mini) reviews with authors photo and biography encompassing the basic and applied research and current developments in educational aspects of these scientific areas.
- Editorial Board
- Information for Authors
- Subscribe to this Title
- Ingenta Connect is not responsible for the content or availability of external websites