POLLEN24_SP_Dataset

[EN] A comprehensive dataset of pollen images has been constructed. This Ground Truth termed POLLEN24_SP, comprised 32.285 pollen/particle images (captured by an expert using optical microscopy) and covers the 24 most prevalent types of pollen grains found in Spanish honeys. Twelve different pre-exi...

Descripción completa

Detalles Bibliográficos
Autores: Valiente González, José Miguel|||0000-0003-4055-8976, Escriche Roberto, Mª Isabel|||0000-0003-0180-0360, Martín-Osuna, Juan José, Peral-Pinto, Ana María
Tipo de recurso: conjunto de datos
Fecha de publicación:2025
País:España
Institución:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:riunet.upv.es:10251/221455
Acceso en línea:https://riunet.upv.es/handle/10251/221455
Access Level:acceso abierto
Palabra clave:Monofloral honey classification
Pollen Classification
Pollen dataset
Labelling and annotating application
HoneyApp
Convolutional neural networks (CNN)
Descripción
Sumario:[EN] A comprehensive dataset of pollen images has been constructed. This Ground Truth termed POLLEN24_SP, comprised 32.285 pollen/particle images (captured by an expert using optical microscopy) and covers the 24 most prevalent types of pollen grains found in Spanish honeys. Twelve different pre-existing Convolutional Neural Networks (CNN) were evaluated in order to discern the relative frequencies of different pollen types, including our new proposals PolleNetV2 and PollenNetV2.mobile.