AUTHOR=Sun Lei , Cui Xiwen , Fan Xiaofei , Suo Xuesong , Fan Baojiang , Zhang Xuejing 

TITLE=Automatic detection of pesticide residues on the surface of lettuce leaves using images of feature wavelengths spectrum

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 13 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.929999

DOI=10.3389/fpls.2022.929999

ISSN=1664-462X

ABSTRACT=The inappropriate application of pesticides to vegetable crops can usually lead to environmental pollution, pesticide residues, etc., which generate serious impacts on the ecological environment, human health and so on. Given the current inaccurate detection of pesticide residues, expensive equipment, complex flow and other problems, a detection method for pesticide residues in lettuce leaves was proposed in this study. To specify this method, the characteristic wavelength of pesticide residues was acquired through the spectral analysis, the machine vision equipment was improved, and then bandpass filter and light source of characteristic wavelength were installed to acquire the leaf image information under the light of characteristic wavelength. Next, the image preprocessing and feature information extraction were automatically implemented through programming. Several links were established for the training model so that the feature information needed could be automatically extracted after the batch input of images. The pesticide residue detected through the chemical method was taken as the output and modeled together with the input image information using the convolutional neural network (CNN) algorithm. Furthermore, a prediction program was rewritten to generalize the input images during the prediction process and directly obtain the output pesticide residue. The experimental results revealed that when the detection device and method designed in this study were used to detect the pesticide residues in lettuce leaves, the coefficient of determination of equation reached 0.883, and the root mean square error (RMSE) was 0.134 mg/L, indicating that the accuracy was high and the proposed method integrated the advantages of spectrum detection and deep learning. Moreover, the improved machine vision equipment lowered the equipment cost and provided powerful support for the application and popularization of the proposed method and equipment.