Soybean images dataset for caterpillar and Diabrotica speciosa pest detection and classification
Maria Eloisa Mignoni,
Aislan Honorato,
Rafael Kunst,
Rodrigo Righi,
Angélica Massuquetti
Affiliations
Maria Eloisa Mignoni
Corresponding author.; Universidade do Estado de Mato Grosso Carlos Alberto Reyes Maldonado, Avenida das Garças, 1192, Centro, Nova Mutum, MT 78450-000, Brazil
Aislan Honorato
Centro Universitário Univag, Av. Dom Orlando Chaves, 2655, Bairro Cristo Rei, Várzea Grande, MT 78118-900, Brazil
Rafael Kunst
Applied Computing Graduate Program University of Vale do Rio dos Sinos (Unisinos), Av. Unisinos, 950, Bairro Cristo Rei, São Leopoldo, RS 93022-750, Brazil; Graduate Program in Economics University of Vale do Rio dos Sinos (Unisinos), Av. Dr. Nilo Peçanha, 1600 - Boa Vista, Porto Alegre, RS 91330-002, Brazil
Rodrigo Righi
Applied Computing Graduate Program University of Vale do Rio dos Sinos (Unisinos), Av. Unisinos, 950, Bairro Cristo Rei, São Leopoldo, RS 93022-750, Brazil
Angélica Massuquetti
Graduate Program in Economics University of Vale do Rio dos Sinos (Unisinos), Av. Dr. Nilo Peçanha, 1600 - Boa Vista, Porto Alegre, RS 91330-002, Brazil
This article presents a dataset of insect-damaged soybean leaves. The capture of images was carried out on several soy farms, under realistic weather conditions, using two cell phones and a UAV. The dataset consists of 3 (three) folders with a total of 6,410 images. The dataset is divided into three categories: (I) healthy plants, (II) plants affected by caterpillars, and (III) images of plants damaged by Diabrotica speciosa. This dataset allows training and validation of machine learning models to diagnose, recognize, and classify soybeans affected by caterpillars or Diabrotica speciosa. The images can be processed according to the user’s need since only the size was standardized during the pre-processing phase.
