Automated Health Estimation of Capsicum annuum L. Crops by Means of Deep Learning and RGB Aerial Images

Abstract

Recently, the use of small UAVs for monitoring agricultural land areas has been increasingly used by agricultural producers in order to improve crop yields. However, correctly interpreting the collected imagery data is still a challenging task. In this study, an automated pipeline for monitoring C. Annuum crops based on a deep learning model is implemented. The system is capable of performing inferences on the health status of individual plants, and to determine their locations and shapes in a georeferenced orthomosaic. Accuracy achieved on the classification task was 94.5. AP values among classes were in the range of [63,100] for plant location boxes, and in [40,80] for foliar area predictions. The methodology requires only RGB images, and so, it can be replicated for the monitoring of other types of crops by only employing consumer-grade UAVs. A comparison with random forest and large-scale mean shift segmentation methods which use predetermined features is presented. NDVI results obtained with multispectral equipment are also included.

Evolução e tendências do uso de Aeronaves Remotamente Pilotadas no Brasil (2017-2022) e suas implicações para o geoprocessamento

“…No início, os trabalhos científicos publicados, tinham uma ideia de reaplicação do que já era feito com os dados tradicionais, aqueles coletados por sensores a bordo de satélites ou aviões, sendo a inovação estritamente atrelada a altíssima resolução espacial. São exemplos mapeamentos de uso e cobertura da terra (Al-Najjar et al, 2019;Gonçalves, 2021;Wyard et al, 2022), análises de degradação na vegetação (Paz et al, 2021) e a geração de índices de vegetação, majoritariamente o Índice de Vegetação da Diferença Normalizada (NDVI), mas também com outras propostas como o Índice Resistente à Atmosfera na Região Visível (VARI) e o Green Leaf Index (GLI) (Barbosa et al, 2021;Meivel & Maheswari, 2020;Singh et al, 2022;Sosa et al, 2022). Exemplos desses produtos são apresentados na Figura 5.…”

“…Esse índice de vegetação, que já era extremamente explorado com dados orbitais (principalmente da família LandSat), passou a ser replicado com dados coletados pelas RPAs, gerando resultados distintos da aplicação tradicional. Embora muito bem documentado sua aplicação da agricultura (Barbosa et al, 2021;Shafi et al, 2020;Singh et al, 2022;Sosa et al, 2022) Propostas de aplicação mais adequadas à elevada resolução espacial dos dados derivados das RPAs surgem em momento posterior. Exemplo marcante refere-se às análises de vegetação agora embasada por mapeamentos de espécies (Luo et al, 2022;Takeshige et al, 2022).…”

Crop Growth Analysis Using Automatic Annotations and Transfer Learning in Multi-date Aerial Images and Orthomosaics

“…Therefore, this approach is preferable to classification of individual crop specimens which can result in challenges due to spectrum overlap problems due to intra-crop occlusions [11]. It has also been demonstrated that the instance segmentation pipeline trained over nadir RGB images for prediction and analysis of plant shapes can be ensembled with other ML models to perform similar and additional analysis over orthomosaics to suit different agricultural environments [12]. Another work based on combinational machine learning techniques coupled classification and regression trees to enhance the accuracy of ground coverage (GC) estimation from UAV imagery for wheat phenotyping [13].…”