An automatic generation of pre-processing strategy combined with machine learning multivariate analysis for NIR spectral data

Arianti, Nunik Destria; Saputra, Edo; Sitorus, Agustami

doi:10.1016/j.jafr.2023.100625

Cited by 5 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MSC preprocessing can be employed to eliminate noise caused by scattering issues in the spectra [19]. MA preprocessing effectively eliminates noise arising from time series, cyclic variations, and random fluctuations, enabling further analysis of data trends and the development direction [20]. SG preprocessing ensures the preservation of signal shape and width while filtering out unwanted noise components [21].…”

Section: Spectral Preprocessingmentioning

confidence: 99%

Research on the Identification Method of Maize Seed Origin Using NIR Spectroscopy and GAF-VGGNet

Xu,

Fu,

Gao

et al. 2024

Agriculture

View full text Add to dashboard Cite

The origin of seeds is a crucial environmental factor that significantly impacts crop production. Accurate identification of seed origin holds immense importance for ensuring traceability in the seed industry. Currently, traditional methods used for identifying the origin of maize seeds involve mineral element analysis and isotope fingerprinting, which are laborious, destructive, time-consuming, and suffer from various limitations. In this experiment, near-infrared spectroscopy was employed to collect 1360 maize seeds belonging to 12 different varieties from 8 distinct origins. Spectral information within the range of 11,550–3950 cm−1 was analyzed while eliminating multiple interferences through first-order derivative combined with standard normal transform (SNV). The processed one-dimensional spectral data were then transformed into three-dimensional spectral maps using Gram’s Angle Field (GAF) to be used as input values along with the VGG-19 network model. Additionally, a convolution layer with a step size of 1 × 1 and the padding value set at 1 was added, while pooling layers had a step size of 2 × 2. A batch size of 48 and learning rate set at 10−8 were utilized while incorporating the Dropout mechanism to prevent model overfitting. This resulted in the construction of the GAF-VGG network model which successfully decoded the output into accurate place-of-origin labels for maize seed detection. The findings suggest that the GAF-VGG network model exhibits significantly superior performance compared to both the original data and the PCA-based origin identification model in terms of accuracy, recall, specificity, and precision (96.81%, 97.23%, 95.35%, and 95.12%, respectively). The GAF-VGGNet model effectively captures the NIR features of different origins of maize seeds without requiring feature wavelength extraction, thereby reducing training time and enhancing accuracy in identifying maize seed origin. Moreover, it simplifies near-infrared (NIR) spectral modeling complexity and presents a novel approach to maize seed origin identification and traceability analysis.

show abstract

Section: Spectral Preprocessingmentioning

confidence: 99%

Research on the Identification Method of Maize Seed Origin Using NIR Spectroscopy and GAF-VGGNet

Xu,

Fu,

Gao

et al. 2024

Agriculture

View full text Add to dashboard Cite

show abstract

“…This will impact the performance of the developed grading machine if it is to be used directly. Therefore, creating a classification modeling strategy that uses source images from low-cost cameras is necessary, especially machine learning [8]- [11].…”

Section: Introductionmentioning

confidence: 99%

Comparison of feature extraction and auto-preprocessing for chili pepper (Capsicum Frutescens) quality classification using machine learning

Asian,

Arianti,

Ariefin

et al. 2024

IJ-AI

View full text Add to dashboard Cite

The low-cost camera for machine vision, such as a webcam, still has a problem with resolution noise. Therefore, it is important to learn strategies to reduce noise from low-cost camera images so that they can be widely used for grading machines in the future. This paper aims to compare three feature extraction methods with auto-preprocessing to classify chili pepper <em>(Capsicum Frutescens)</em> quality using a machine learning algorithm. Three extraction methods were used, including the color feature, oriented FAST and rotated BRIEF (ORB), and the combination color feature and ORB. A total of 525 image data for quality chili pepper were collected using the webcam. The auto-preprocessing strategy to classify chili peppers can improve the performance of machine-learning algorithms for all data generated by the feature extractor. The performance of the chili paper quality classification model with auto-preprocessing of the variable color feature can improve the performance of machine learning algorithms by up to 64.21%. The performance improvement of the classification model using the ORB feature variable and the auto-preprocessing of up to 4.41%. The performance improvement of the classification model using machine learning algorithms is 11.27% when using the combination color feature and ORB feature and autopreprocessing.

show abstract

Machine learning supported ground beef freshness monitoring based on near‐infrared and paper chromogenic array

Feng,

Wang,

Beykal

et al. 2024

Food Frontiers

View full text Add to dashboard Cite

Maintaining freshness and quality is crucial in the meat industry, as lipid oxidation can lead to undesirable odors, flavors, and potential health risks. Traditional methods for assessing meat freshness often involve time‐consuming and destructive techniques, highlighting the need for rapid, noninvasive approaches. Recent advancements in spectroscopic and chromogenic sensor array technologies have opened up new avenues for monitoring meat quality parameters, offering the potential for real‐time, accurate, and cost‐effective solutions. As thiobarbituric acid reactive substances (TBARS) value is a classic indicator of meat lipid oxidation, this study investigated the data fusion of near‐infrared spectroscopy (NIR) and paper chromogenic array (PCA) for monitoring ground beef TBARS. A standardized PCA was fabricated by photolithography with nine chemoresponsive dyes. Changes in ground beef volatile organic compounds during storage were captured in the shifts of PCA color patterns. Nippy, an open‐source Python module, was used for automated NIR spectra preprocessing. The optimal preprocessing pipeline was found by 10‐fold cross‐validation in machine learning model development. Among optimized models, partial least square regression showed the best performance in coefficient of determination (R2) of .9477, root mean squared error of prediction of 0.0545 mg malondialdehyde/kg meat, and residual prediction deviation of 4.3717. The promising result of this study indicated the potential for NIR and PCA combinations to monitor TBARS values for ground beef freshness assessment.

show abstract

An automatic generation of pre-processing strategy combined with machine learning multivariate analysis for NIR spectral data

Cited by 5 publications

References 21 publications

Research on the Identification Method of Maize Seed Origin Using NIR Spectroscopy and GAF-VGGNet

Research on the Identification Method of Maize Seed Origin Using NIR Spectroscopy and GAF-VGGNet

Comparison of feature extraction and auto-preprocessing for chili pepper (Capsicum Frutescens) quality classification using machine learning

Machine learning supported ground beef freshness monitoring based on near‐infrared and paper chromogenic array

Contact Info

Product

Resources

About