A study on CNN-based detection of psyllids in sticky traps using multiple image data sources.

Vista sencilla de ítem
dc.contributorJAYME GARCIA ARNAL BARBEDO, CNPTIA; GUILHERME BARROS CASTRO, CromAI, São Paulo.
dc.creatorBARBEDO, J. G. A.
dc.creatorCASTRO, G. B.
dc.date2020-10-07T09:14:58Z
dc.date2020-10-07T09:14:58Z
dc.date2020-10-06
dc.date2020
dc.date.accessioned2026-07-07T04:16:22Z
dc.descriptionAbstract: Deep learning architectures like Convolutional Neural Networks (CNNs) are quickly becoming the standard for detecting and counting objects in digital images. However, most of the experiments found in the literature train and test the neural networks using data from a single image source, making it difficult to infer how the trained models would perform under a more diverse context. The objective of this study was to assess the robustness of models trained using data from a varying number of sources. Nine different devices were used to acquire images of yellow sticky traps containing psyllids and a wide variety of other objects, with each model being trained and tested using different data combinations. The results from the experiments were used to draw several conclusions about how the training process should be conducted and how the robustness of the trained models is influenced by data quantity and variety.
dc.identifierAI, v. 1, n. 2, p. 198-208, June 2020.
dc.identifierhttp://www.alice.cnptia.embrapa.br/alice/handle/doc/1125315
dc.identifierhttps://doi.org/10.3390/ai1020013
dc.identifier.urihttp://hdl.handle.net/123456789/456101
dc.languageeng
dc.rightsopenAccess
dc.subjectAprendizado profundo
dc.subjectRobustez de modelo
dc.subjectVariedade de dados
dc.subjectRedes neurais
dc.subjectRedes Neurais Convolucionais
dc.subjectCitrus huanglongbing
dc.subjectHLB
dc.subjectImagens digitais
dc.subjectDeep learning
dc.subjectModel robustness
dc.subjectData variety
dc.subjectConvolutional Neural Networks
dc.subjectCitrus
dc.subjectNeural networks
dc.subjectDigital images
dc.titleA study on CNN-based detection of psyllids in sticky traps using multiple image data sources.
dc.typeArtigo de periódico

Archivos

Colecciones

Embrapa Agricultura Digital (CNPTIA)