Establishment of a Daqu Grade Classification Model Based on Computer Vision and Machine Learning

Foods, Journal Year: 2025, Volume and Issue: 14(4), P. 668 - 668

Published: Feb. 16, 2025

The grade of Daqu significantly influences the quality Baijiu. To address issues high subjectivity, substantial labor costs, and low detection efficiency in evaluation, this study focused on light-flavor proposed a two-layer classification structure model based computer vision machine learning. Target images were extracted using three image segmentation methods: threshold segmentation, morphological fusion, K-means clustering. Feature factors selected through methods including mean decrease accuracy random forest (RF-MDA), recursive feature elimination (RFE), LASSO regression, ridge regression. evaluation was constructed support vector (SVM), logistic regression (LR), (RF), k-nearest neighbor (KNN), stacking model. results indicated following: (1) In terms performance, fusion method achieved an accuracy, precision, recall, F1-score, AUC 96.67%, 95.00%, 0.95, 0.96, respectively. (2) For Daqu-P, Daqu-F, Daqu-S, RF models performed best, achieving 97.50%, 0.97, 0.99, (3) distinguishing Daqu-P from combination RF-MDA demonstrated best with 90.00%, 94.44%, 85.00%, 0.89, This provides theoretical technical for efficient objective evaluation.

Language: Английский

COVID-19 Pandemic: A Worldwide Critical Review with the Machine Learning Model-Based Prediction

Journal of The Institution of Engineers (India) Series B, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 27, 2024

Language: Английский