Dataset Information

Detection of Pesticide Residue Level in Grape Using Hyperspectral Imaging with Machine Learning.

ABSTRACT: Rapid and accurate detection of pesticide residue levels can help to prevent the harm of pesticide residue. This study used visible/near-infrared (Vis-NIR) (376-1044 nm) and near-infrared (NIR) (915-1699 nm) hyperspectral imaging systems (HISs) to detect the level of pesticide residues. Three different varieties of grapes were sprayed with four levels of pesticides. Logistic regression (LR), support vector machine (SVM), random forest (RF), convolutional neural network (CNN), and residual neural network (ResNet) models were used to build classification models for pesticide residue levels. The saliency maps of CNN and ResNet were conducted to visualize the contribution of wavelengths. Overall, the results of NIR spectra performed better than those of Vis-NIR spectra. For Vis-NIR spectra, the best model was ResNet, with the accuracy of over 93%. For NIR spectra, LR was the best, with the accuracy of over 97%, but SVM, CNN, and ResNet also showed closed and fine results. The saliency map of CNN and ResNet presented similar and closed ranges of crucial wavelengths. Overall results indicated deep learning performed better than conventional machine learning. The study showed that the use of hyperspectral imaging technology combined with machine learning can effectively detect the level of pesticide residues in grapes.

SUBMITTER: Ye W

PROVIDER: S-EPMC9180647 | biostudies-literature | 2022 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Detection of Pesticide Residue Level in Grape Using Hyperspectral Imaging with Machine Learning.

Ye Weixin W Yan Tianying T Zhang Chu C Duan Long L Chen Wei W Song Hao H Zhang Yifan Y Xu Wei W Gao Pan P

Foods (Basel, Switzerland) 20220530 11

Rapid and accurate detection of pesticide residue levels can help to prevent the harm of pesticide residue. This study used visible/near-infrared (Vis-NIR) (376-1044 nm) and near-infrared (NIR) (915-1699 nm) hyperspectral imaging systems (HISs) to detect the level of pesticide residues. Three different varieties of grapes were sprayed with four levels of pesticides. Logistic regression (LR), support vector machine (SVM), random forest (RF), convolutional neural network (CNN), and residual neural ...[more]

PMID: 35681359

Similar Datasets

Project description:Cadmium (Cd) is a toxic element that can accumulate in edible plant tissues and negatively impact human health. Traditional Cd quantification methods are time-consuming, expensive, and generate a lot of toxic waste, slowing development of methods to reduce uptake. The objective of this study was to determine whether hyperspectral imaging (HSI) and machine learning (ML) can be used to predict Cd concentrations in plants using kale (Brassica oleracea) and basil (Ocimum basilicum) as model crops. The experiments were conducted in an automated phenotyping facility where all environmental conditions except soil Cd concentration were kept constant. Cd concentrations were determined at harvest using traditional methods and used to train the ML models with data collected from the imaging sensor. Visible/near infrared (VNIR) images were also collected at harvest and processed to calculate reflectance at 473 bands between 400 to 998 nm. All reflectance spectra were subject to the feature selection algorithm ReliefF and Principal Component Analysis (PCA) to generate data and provide input to evaluate three ML classification models: artificial neural network (ANN), ensemble learning (EL), and support vector machine (SVM). Plants were categorized according to Cd concentrations higher or lower than the safety threshold of 0.2 mg kg−1 Cd. Wavelengths with the highest ranks for Cd detection were between 519 and 574, and 692 and 732 nm, indicating that Cd content likely altered the plants’ chlorophyll content and altered leaf internal structure. All models were able to sort the plants into groups, though the model with the best F1 score was the ANN for the validation subset that utilized reflectance from all wavelengths. This study demonstrates that HSI and ML are promising technologies for the fast and precise diagnosis of Cd in leafy green plants, though additional studies are needed to adapt this approach for more complex field environments. Graphical abstract Highlights • Classification modeling can be used to estimate Cd concentrations in leafy greens.• Neural network model was the best algorithm in this study for classifying plants.• Cadmium induced changes in wavelengths linked with chlorophyll and leaf structure. ReliefF; Ensemble learning; Support vector machine; Neural networks basil; Kale.

Dataset Information

Detection of Pesticide Residue Level in Grape Using Hyperspectral Imaging with Machine Learning.

Publications

Detection of Pesticide Residue Level in Grape Using Hyperspectral Imaging with Machine Learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets