Artificial neural networks in online semiautomated pest discriminability: an applied case with 2 Thrips species

Authors: PETER FEDOR, ELADIA MARIA PEÑA-MÉNDEZ, HALINA KUCHARCZYK, JAROMIR VANHARA, JOSEF HAVEL, MARTINA DORICOVA, PAVOL PROKOP

Abstract: Being faced with practical problems in pest identification, we present a methodical paper based on artificial neural networks to discriminate morphologically very similar species, Thrips sambuci Heeger, 1854 and Thrips fuscipennis Haliday, 1836 (Thysanoptera: Thripinae), as an applied case for more general use. The artificially intelligent system may be successfully applied as a credible, online, semiautomated identification tool that extracts hidden information from noisy data, even when the standard characters have much overlap and the common morphological keys hint at the practical problem of high morphological plasticity. Statistical analysis of 17 characters, measured or determined for each Thrips fuscipennis and T. sambuci specimen (reared from larvae in our laboratories), including 15 quantitative morphometric variables, was performed to elucidate morphological plasticity, detect eventual outliers, and visualize differences between the studied taxa. The computational strategy applied in this study includes a set of statistical tools (factor analysis, correlation analysis, principal component analysis, and linear discriminant analysis) followed by the application of a multilayer perceptron artificial neural network system, which models functions of almost arbitrary complexity. This complex approach has proven the existence of 2 separate species: T. fuscipennis and T. sambuci. All the specimens could be clearly distinguished with 2 distinct subgroups for each species, determined by sex. In conclusion, the use of an optimal 3-layer ANN architecture (17, 4, 1) enables fast and reliable 100% classification as proven during the extensive verification process.

Keywords: Artificial neural networks, online semiautomated pest identification, Thysanoptera

Full Text: PDF