Spectroscopic and Imaging Technologies Combined with Machine Learning for Intelligent Perception of Pesticide Residues in Fruits and Vegetables

Abstract

Pesticide residues in fruits and vegetables pose a serious threat to food safety. Traditional detection methods have defects such as complex operation, high cost, and long detection time. Therefore, it is of great significance to develop rapid, non-destructive, and efficient detection technologies and equipment. In recent years, the combination of spectroscopic techniques and imaging technologies with machine learning algorithms has developed rapidly, providing a new attempt to solve this problem. This review focuses on the research progress of the combination of spectroscopic techniques (near-infrared spectroscopy (NIRS), hyperspectral imaging technology (HSI), surface-enhanced Raman scattering (SERS), laser-induced breakdown spectroscopy (LIBS), and imaging techniques (visible light (VIS) imaging, NIRS imaging, HSI technology, terahertz imaging) with machine learning algorithms in the detection of pesticide residues in fruits and vegetables. It also explores the huge challenges faced by the application of spectroscopic and imaging technologies combined with machine learning algorithms in the intelligent perception of pesticide residues in fruits and vegetables: the performance of machine learning models requires further enhancement, the fusion of imaging and spectral data presents technical difficulties, and the commercialization of hardware devices remains underdeveloped. This review has proposed an innovative method that integrates spectral and image data, enhancing the accuracy of pesticide residue detection through the construction of interpretable machine learning algorithms, and providing support for the intelligent sensing and analysis of agricultural and food products.

Keywords: deep learning; imaging; machine learning; pesticide residue; spectral technology.

Figure 6

The operation of the HSI imaging sensor [56].

Figure 1

Scheme of the spectral analysis system (VIS imaging, NIRS imaging, HSI imaging, terahertz time-domain imaging) for the detection of agricultural and food products with spectral data and imaging data.

Figure 2

A typical flow chart of spectroscopy technology combined with machine learning for pesticide residue detection [16]. (A) Scheme of spectrometer; (B) The architecture of generative adversarial networks to obtain the spectral datasets; (C) The improved generator; (D) The spectrum generated by improved generative adversarial networks (a) and visualization of generated spectra (b); (E) The confusion matrix of classification results on spectral dataset.

Figure 3

(A) Spectral analysis of pesticide residues in Chinese cabbage by VIS/NIR spectroscopy [22]; (B) Identification of chlorpyrifos residues on Chinese cabbage by combining NIRS with PLS-DA, SVM, ANNs, and principal component ANNs [24].

Figure 4

(A) The binding mechanism of detection materials and carrier materials for the identification of pesticide residues in fruits and vegetables using SERS [31]; (B) (a) The MC-CNNs-GRU discrimination model for pesticide/fungicide residues; (b) The t-SNE analysis results of the MC-CNNs-GRU model; (c) The confusion matrix of the training set and (d) the test set [32].

Figure 5

Scheme of processing for food hyperspectral image datasets [40].

Figure 7

The flowchart of detecting multiple benzimidazole pesticide residues in Toona sinensis leaves using terahertz imaging and deep learning [59].

Figure 8

The integration of spectral and image data with machine learning algorithms for pesticide residue identification in fruits and vegetables: challenges and future prospects.