Huanglongbing (HLB, ex-greening) is the most devastating citrus disease around the world. We modelled HLB spread in scenarios with different populational levels of the main alternative host ( Murraya paniculata ) and Diaphorina citri Kuwayama, vector of HLB associated bacteria; and removal of HLB-symptomatic plants. A compartmental deterministic mathematical model was built for representing the HLB dynamics in the Reconcavo Baiano, Bahia State, Brazil. The model encompasses delays on latency and incubation disease periods and on the D. citri nymphal stages. The simulations indicated that the presence of alternative hosts at low proportion would not play a crucial role in HLB dynamics in situations of poor D. citri management, regardless of HLB-symptomatic plants eradication. Symptomatic citrus plants contribute more to increase the HLB-incidence than the alternative host in scenarios without a suitable D. citri management.

Dada a importância da citricultura na economia do Brasil, uma das potências mundiais nesse setor, é essencial estar sempre renovando as técnicas de cultivo para tornar a produção mais eficiente. Este trabalho testa a eficiência da rede neural YOLOv8 para a detecção de objetos, com o propósito de obter uma estimativa de safra de laranjas mais precisa. O desempenho geral apresentado pela rede foi satisfatório, porém não nos permitiu demonstrar que ela supera sua versão anterior YOLOv5 em todas as métricas.

Traditionally, sweet orange crop forecasting has involved manually counting fruits from numerous trees, which is a labor-intensive process. Automatic systems for fruit counting, based on proximal imaging, computer vision, and machine learning, have been considered a promising alternative or complement to manual counting. These systems require data association components that prevent multiple counting of the same fruit observed in different images. However, there is a lack of work evaluating the accuracy of multiple fruit counting, especially considering (i) occluded and re-entering green fruits on leafy trees, and (ii) counting ground-truth data measured in the crop field. We propose a non-invasive alternative that utilizes fruit counting from videos, implemented as a pipeline. Firstly, we employ CNNs for the detection of visible fruits. Inter-frame association techniques are then applied to track the fruits across frames. To handle occluded and re-appeared fruit, we introduce a relocalization component that employs 3-D estimation of fruit locations. Finally, a neural network regressor is utilized to estimate the total number of fruit, integrating image-based fruit counting with other tree data such as crop variety and tree size. The results demonstrate that the performance of our approach is closely tied to the quality of the field-collected videos. By ensuring that at least 30% of the fruit is accurately detected, tracked, and counted, our yield regressor achieves an impressive coefficient of determination of 0.85. To the best of our knowledge, this study represents one of the few endeavors in fruit estimation that incorporates manual fruit counting as a reference point for evaluation. We also introduce annotated datasets for multiple orange tracking (MOrangeT) and detection (OranDet), publicly available to foster the development of novel methods for image-based fruit counting.