Please use this identifier to cite or link to this item: http://www.alice.cnptia.embrapa.br/alice/handle/doc/1125315
Research center of Embrapa/Collection: Embrapa Informática Agropecuária - Artigo em periódico indexado (ALICE)
Date Issued: 2020
Type of Material: Artigo em periódico indexado (ALICE)
Authors: BARBEDO, J. G. A.
CASTRO, G. B.
Additional Information: JAYME GARCIA ARNAL BARBEDO, CNPTIA; GUILHERME BARROS CASTRO, CromAI, São Paulo.
Title: A study on CNN-based detection of psyllids in sticky traps using multiple image data sources.
Publisher: AI, v. 1, n. 2, p. 198-208, June 2020.
Language: Ingles
Keywords: Aprendizado profundo
Robustez de modelo
Variedade de dados
Redes neurais
Redes Neurais Convolucionais
Citrus huanglongbing
HLB
Imagens digitais
Deep learning
Model robustness
Data variety
Convolutional Neural Networks
Description: Abstract: 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.
NAL Thesaurus: Citrus
Neural networks
Digital images
Data Created: 2020-10-06
Appears in Collections:Artigo em periódico indexado (CNPTIA)

Files in This Item:
File Description SizeFormat 
AP-CNN-psyllids-2020.pdf4.5 MBAdobe PDFThumbnail
View/Open

FacebookTwitterDeliciousLinkedInGoogle BookmarksMySpace