Immune Algorithm for the Singular Sample of Near Infrared Spectroscopy

2013 ◽  
Vol 427-429 ◽  
pp. 770-773
Author(s):  
Fan Yu ◽  
Ji Xin Li
Keyword(s):  
Genetic Algorithm ◽  
Infrared Spectroscopy ◽  
Near Infrared Spectroscopy ◽  
Prediction Accuracy ◽  
Near Infrared ◽  
Immune Algorithm ◽  
Data Sets ◽  
Artificial Immune ◽  
Near Infrared Spectra ◽  
Better Than

This paper proposes a method to identify the singular sample with near infrared spectroscopy based on immune algorithm. T The immune algorithm and genetic algorithm for the same NIR spectral data sets are singular sample identification and comparison of the two methods. Remove the singular sample, immune algorithm is better than the genetic algorithm are used to PRESS of PLS model of water, fat, the protein increased by 25.8%, 32.1%, 21.7%.Experiments show that, artificial immune algorithm is not only suitable for near infrared spectra of singular sample, but also can improve the model prediction accuracy and robustness.

scholarly journals Near-infrared spectroscopy used to predict soybean seed germination and vigour

2018 ◽  
Vol 28 (3) ◽  
pp. 245-252 ◽  
Author(s):  
Maythem Al-Amery ◽  
Robert L. Geneve ◽  
Mauricio F. Sanches ◽  
Paul R. Armstrong ◽  
Elizabeth B. Maghirang ◽  
...  
Keyword(s):  
Infrared Spectroscopy ◽  
Seed Germination ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Prediction Models ◽  
Soybean Seed ◽  
Seed Vigour ◽  
Near Infrared Spectra ◽  
High Germination ◽  
Qualitative Models

AbstractRapid, non-destructive methods for measuring seed germination and vigour are valuable. Standard germination and seed vigour were determined using 81 soybean seed lots. From these data, seed lots were separated into high and low germinating seed lots as well as high, medium and low vigour seed lots. Near-infrared spectra (950–1650 nm) were collected for training and validation samples for each seed category and used to create partial least squares (PLS) prediction models. For both germination and vigour, qualitative models provided better discrimination of high and low performing seed lots compared with quantitative models. The qualitative germination prediction models correctly identified low and high germination seed lots with an accuracy between 85.7 and 89.7%. For seed vigour, qualitative predictions for the 3-category (low, medium and high vigour) models could not adequately separate high and medium vigour seeds. However, the 2-category (low, medium plus high vigour) prediction models could correctly identify low vigour seed lots between 80 and 100% and the medium plus high vigour seed lots between 96.3 and 96.6%. To our knowledge, the current study is the first to provide near-infrared spectroscopy (NIRS)-based predictive models using agronomically meaningful cut-offs for standard germination and vigour on a commercial scale using over 80 seed lots.

scholarly journals Time-of-flight Near-infrared Spectroscopy for Nondestructive Measurement of Internal Quality in Grapefruit

2013 ◽  
Vol 138 (3) ◽  
pp. 225-228 ◽  
Author(s):  
Yohei Kurata ◽  
Tomoe Tsuchida ◽  
Satoru Tsuchikawa
Keyword(s):  
Infrared Spectroscopy ◽  
Near Infrared Spectroscopy ◽  
Prediction Accuracy ◽  
Near Infrared ◽  
Time Of Flight ◽  
Soluble Solids ◽  
Internal Quality ◽  
Citrus Paradisi ◽  
Least Squares Regression ◽  
Time Resolved

We proposed a technique combining time-of-flight (TOF) and near-infrared spectroscopy (NIRS), termed TOF-NIRS, capable of measuring the time-resolved profiles of near-infrared (NIR) light with nanosecond resolution. Analysis of the variation in time-resolved profiles was used to estimate soluble solids concentration (SSC) and acidity in grapefruit (Citrus paradisi), and the prediction accuracy was compared with the conventional NIR measurement device. In data processing, the cross-correlation function, which evaluated the similarity between the reference and transmitted beams, was introduced as an explanatory variable for partial least squares regression. TOF-NIRS predicted both SSC and acidity in grapefruit with higher precision than the conventional NIR measurement with respective r values of 0.72 and 0.85. Specifically, the superiority of TOF-NIRS was attributed to measurement time and prediction accuracy in determining acidity.

scholarly journals Algorithm construction methodology for diagnostic classification of near-infrared spectroscopy data

2011 ◽  
Vol 25 (1) ◽  
pp. 1-11 ◽  
Author(s):  
Ramón Guevara ◽  
Lynn Stothers ◽  
Andrew Macnab
Keyword(s):  
Infrared Spectroscopy ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Pelvic Floor Muscle ◽  
Classification And Regression Tree ◽  
Data Sets ◽  
Linear Discriminant ◽  
Cart Analysis ◽  
Chromophore Concentration ◽  
Nirs Data

Background: Near-infrared spectroscopy (NIRS) has recognized potential but limited application for non-invasive diagnostic evaluation. Data analysis methodology that reproducibly distinguishes between the presence or absence of physiologic abnormality could broaden clinical application of this optical technique.Methods: Sample data sets from simultaneous NIRS bladder monitoring and invasive urodynamic pressure-flow studies (UDS) are used to illustrate how a diagnostic algorithm is constructed using classification and regression tree (CART) analysis. Misclassification errors of CART and linear discriminant analysis (LDA) are computed and examples of other urological NIRS data likely amenable to CART analysis presented.Results: CART generated a clinically relevant classification algorithm (error 4%) using 46 data sets of changes in chromophore concentration composed of the whole time series without specifying features. LDA did not (error 16%). Using CART NIRS data provided comparable discriminant ability to the UDS diagnostic nomogram for the presence or absence of obstructive pathology (88% specificity, 84% precision). Pilot data examples from children with and without voiding dysfunction and women with mild or severe pelvic floor muscle dysfunction also show potentially diagnostic differences in chromophore concentration.Conclusions: CART analysis can likely be applied in other NIRS monitoring applications intended to classify patients into those with and without pathology.

Sign in / Sign up

Export Citation Format

Share Document