Usage of few-shot learning and meta-learning in agriculture: A literature review
João Vitor de Andrade Porto, Arlinda Cantero Dorsa, Vanessa Weber, Karla Rejane de Andrade Porto, Hemerson Pistori
Abstract
This paper examines the potential of using few-shot learning and computer vision techniques for detecting, identifying, and counting agricultural pests and diseases in images. A systematic review of papers published between 2020 and 2022 was conducted to evaluate the applications and results across various fields of agriculture. 24 papers were selected according to inclusion and exclusion criteria, organized similarly to Wang et al.'s proposal. The findings suggest that applying meta-learning and few-shot learning in agriculture holds promise, as demonstrated by recent works. These techniques offer diverse solutions to issues related to plant diseases, insect pests, and morphology using machine learning.