We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
基于可见- 近红外透射光谱的蜜柚检测中影响因素分析.
- Authors
孙潇鹏; 刘灿灿; 陆华忠; 徐 赛
- Abstract
For the problem of non-destructive testing accuracy of internal quality of thick-skinned fruits, a fast, stable and efficient visible-near infrared transmission spectrum detection method was used and a self-built experimental prototype was utilized to analyze the influence rules of different light intensity, light source distance and peel thickness on transmission spectrum data of honey pomelo.It is found through the test that when the opyimum parameters were set as L10 and H2, the original spectra was preprocessed by Savitzky-Golay polynomial smoothing, and the spectral features were selected by principal component analysis, the partial least squares regression model established to predict the sugar degree accuracy of honey pomelo is the highest, and the model has the best prediction ability. The results can provide a reference for quality testing of thick-skinned fruit.
- Subjects
PARTIAL least squares regression; PRINCIPAL components analysis; NONDESTRUCTIVE testing; LIGHT sources; REGRESSION analysis
- Publication
Packaging & Food Machinery, 2022, Vol 40, Issue 4, p1
- ISSN
1005-1295
- Publication type
Article
- DOI
10.3969/j.issn.1005-1295.2022.04.001