Data Fusion and Dimensionality Reduction for Pest Management in Pitahaya Cultivation

Computation, Год журнала: 2025, Номер 13(6), С. 137 - 137

Опубликована: Июнь 3, 2025

This study addresses the critical need for effective data fusion strategies in pest prediction pitahaya (dragon fruit) cultivation Ecuadorian Amazon, where heterogeneous sources—such as environmental sensors and chlorophyll measurements—offer complementary but fragmented insights. Current agricultural monitoring systems often fail to integrate these streams, limiting early detection accuracy. To overcome this, we compared late approaches using comprehensive experiments. Multidimensionality is a central challenge: datasets span temporal (hourly sensor readings), spatial (plot-level samples), spectral (chlorophyll reflectance) dimensions. We applied dimensionality reduction techniques—PCA, KPCA (linear, polynomial, RBF), t-SNE, UMAP—to preserve relevant structure enhance interpretability. Evaluation metrics included proportion of information retained (score) cluster separability (silhouette score). Our results demonstrate that yields superior integrated representations, with PCA KPCA-linear achieving highest scores (0.96 vs. 0.94), KPCA-poly best definition (silhouette: 0.32 0.31). Statistical validation Friedman test (χ2 = 12.00, p 0.02) Nemenyi post hoc comparisons (p < 0.05) confirmed significant performance differences. KPCA-RBF performed poorly (score: 0.83; silhouette: 0.05), although t-SNE UMAP offered visual insights, they underperformed clustering 0.12). These findings make three key contributions. First, better captures cross-domain interactions before reduction, improving robustness. Second, offers an non-linear mapping suitable tropical agroecosystem complexity. Third, our framework, when deployed Joya de los Sachas, improved accuracy by 12.60% over manual inspection, leading more targeted pesticide use. contributes precision agriculture providing low-cost, scalable smallholder farmers. Future work will explore hybrid pipelines sensor-agnostic models extend generalizability.

Язык: Английский

---

Advanced sound-based pest detection in agriculture using deep learning and adaptive optimization

International Journal of Information Technology, Год журнала: 2025, Номер unknown

Опубликована: Июнь 4, 2025

Язык: Английский

---

Role of Big Data Analytics in Intelligent Agriculture

Studies in computational intelligence, Год журнала: 2024, Номер unknown, С. 19 - 48

Опубликована: Янв. 1, 2024

Язык: Английский

---

AI for Smart Agriculture – A Deep Learning Based Turmeric Leaf Disease Detection

2022 7th International Conference on Communication and Electronics Systems (ICCES), Год журнала: 2024, Номер unknown, С. 1491 - 1495

Опубликована: Дек. 16, 2024

Язык: Английский

---