Methods for Automatic Image-Based Classification of Winged Insects Using Computational Techniques: A Systematic Lireracture Review

• Allan Rodrigues Rebelo USP
• Joao Marcos Garcia Fagundes USP
• Luciano Antonio Digiampietri USP https://orcid.org/0000-0003-4890-1548
• Helton Hideraldo Bíscaro USP https://orcid.org/0000-0003-1565-1044

DOI: https://doi.org/10.5753/sbsi.2020.13775

Resumo

Artificial intelligence has been used in conjunction with computational and statistical techniques for automatic identification of insect species because of the current growing demand for this type of solution. This study presents the results of a Systematic Literature Review conducted to evaluate the state of the art of automatic image-based classification methods of winged insects. One thousand and sixty studies were researched and thirty-six were fully analyzed, and, from the results obtained, it was concluded that the area is consolidating with an uptrend and that there are techniques with satisfactory results for the proposed objectives, but there are still major problems to be solved.

Palavras-chave: image-based insect recognition, classification, insects, arthropods, image, feature extraction, systematic literature review, datasets, deep learning