Grade Determination of Flue-cured Tobacco by Near Infrared Spectroscopy Combined with Teaching-learning-based Optimization Algorithm Optimized Extreme Learning Machine

KeyWord:near infrared spectroscopy  teaching-learning-based optimization algorithm  extreme learning machine  tobacco  grade determination

Abstract:

      The quality evaluation on tobacco is an important work as it is a high-value attribute product£®Therefore£¬it is of a certain application value to ultilize intelligent means for efficient classification of tobacco£®Based on near infrared spectroscopy£¨NIRs£©£¬937 tobacco samples from 13 provinces from 2016 to 2018 were used to compare the extreme learning machine£¨ELM£© model effects of three variable screening methods£¬including competitive adaptive reweighted sampling£¨CARS£© method£¬Monte Carlo uninformed variable elimination£¨MC£­UVE£© method and random frog£¨RF£© algorithm£®Compared with partial least squares£­discriminant analysis£¨PLS£­DA£©£¬the advantages of ELM model were verified£®The ELM model was optimized by teaching-learning-based optimization£¨TLBO£© algorithm£¬thus a TLBO£­ELM classification model for flue-cured tobacco samples was established£®Results showed that the classification accuracy of the validation set was 90.16%£®The external verification effect of the testing set was satisfactory£¬and the TLBO£­ELM model had fast convergence speed and strong generalization ability£¬which could be applied to the classification of flue-cured tobacco£®NIRs combined with TLBO to optimize ELM provides a new idea for intelligent tobacco classification.