Application of Convolutional Neural Networks on the Development of Plant-Parasitic Nematode Image Identification System

Lai, Hsien-Hua; Chang, Yu-Tang; Yang, Jiue-in; Chen, Shih‐Fang

doi:10.13031/aim.202100870

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Next to the method presented here, the application of artificial intelligence (AI), also referred to as machine learning or deep learning, has been put forward as a very promising approach in nematode identification (Qing et al, 2022). However, its application still has several limitations, such as the highly similar gross morphology of closely related nematode genera, the declining number of nematologists who can provide pictures for correct identifications, its low accuracy for PPNs, and especially the requirement of a huge training image data set (Lai et al, 2021; Qing et al, 2022; Thevenoux et al, 2021; Uhlemann et al, 2020). Interestingly, our browser‐based key can be used to validate AI‐based PPN identification and as such facilitates the identification process, not just relying on experts, in order to provide the much‐needed AI training data set.…”

Section: Discussionmentioning

confidence: 99%

Diversity of plant‐parasitic nematodes (PPNs) associated with medicinal plants in Vietnam, Vietnamese PPN checklist and a pictorial key for their identification

Nguyen,

Trinh,

Nguyen

et al. 2023

Plant Pathology

View full text Add to dashboard Cite

Plant‐parasitic nematodes (PPNs) are one of the most damaging pests to plants and are able to cause significant damage to all parts of plants, including stems, leaves, flowers, fruits and roots. Studies on the diversity, host range, distribution and identification methods of PPNs are therefore vital in order to create a basis for management. This current study represents the first dedicated investigation of PPNs from medicinal plants in Vietnam, focusing on the diversity of nematodes associated with 23 different plant species. In combination with a literature review of PPNs in Vietnam, this work has resulted in an updated list of 217 PPN species belonging to 40 genera, 15 families and three orders and also provides a pictorial online key for the identification of 52 most common and important PPN genera of the world. This key is based on the most crucial diagnostic features of PPN females, including female body shape, cuticle, labial shape, cephalic framework, stylet, stylet base, pharynx, median bulb, pharyngeal gland, vulva, tail shape and phasmid. Pictorial representations of these genera and their diagnostic characters are included in the browser‐based key to benefit users from all levels in nematology, be they beginners or experts.

show abstract