Premium
Effect of synthetic amorphous silica powder on the cuticle of Tribolium castaneum and Sitophilus oryzae using hyperspectral imaging technique
Author(s) -
Li Yanyu,
Agarwal Manjree,
Cao Yang,
Ren Yonglin
Publication year - 2020
Publication title -
pest management science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.296
H-Index - 125
eISSN - 1526-4998
pISSN - 1526-498X
DOI - 10.1002/ps.5517
Subject(s) - sitophilus , cuticle (hair) , hyperspectral imaging , rice weevil , insect , biology , reflectivity , botany , anatomy , artificial intelligence , computer science , optics , physics
BACKGROUND Synthetic amorphous silica (SAS) is safe for human consumption. SAS damages insect cuticles. Qualitative measurement of cuticle properties of insects affected by SAS is essential to understand the mode of action and develop new pesticides. A hyperspectral reflectance imaging approach was used to directly indicate the impact of SAS on the insect cuticle. RESULTS There were significant differences in the LT 95 values of hydrophobic and hydrophilic SAS against Tribolium castaneum and Sitophilus oryzae ( P < 0.01). T. castaneum was more susceptible to hydrophobic SAS while no difference was found for S. oryzae exposed to both SAS types. In a hyperspectral study, the ventral reflectance of control groups was higher than that of SAS‐treated groups in both visible and short‐wave near‐infrared wavelength ranges. The SAS‐treated groups showed much higher dorsal reflectance. The differences in absorption characteristics of cuticular fat and protein may contribute to the varied performance. The effects of both SASs on insect cuticles was significant, as the 100% recognition rate of the back propagation neural network models suggested. Consistent with the assumption that the efficacy was different between the two SAS types, the lowest rates of the model for two treatment groups were 62.2 and 73.3% in the target and output class. CONCLUSION The efficacy varied considerably between the two insect species and the two SASs. Hyperspectral image analyzing coupled with back propagation artificial neural network accurately recorded how SAS impacts the insect cuticle via the effective wavelengths. These findings showed that SAS is a promising candidate for new pesticide products. © 2019 Society of Chemical Industry