arXiv:2511.16031v1 Announce Type: new 
Abstract: Recent advances in plant phenotyping have driven widespread adoption of multi sensor platforms for collecting crop canopy reflectance data. This includes the collection of heterogeneous data across multiple platforms, with Unmanned Aerial Vehicles (UAV) seeing significant usage due to their high performance in crop monitoring, forecasting, and prediction tasks. Similarly, satellite missions have been shown to be effective for agriculturally relevant tasks. In contrast to UAVs, such missions are bound to the limitation of spatial resolution, which hinders their effectiveness for modern farming systems focused on micro-plot management. In this work, we propose a cross modal learning strategy that enriches high-resolution satellite imagery with UAV level visual detail for crop canopy trait estimation. Using a dataset of approximately co registered satellite UAV image pairs collected from replicated plots of 84 hybrid maize varieties across five distinct locations in the U.S. Corn Belt, we train a model that learns fine grained spectral spatial correspondences between sensing modalities. Results show that the generated UAV-like representations from satellite inputs consistently outperform real satellite imagery on multiple downstream tasks, including yield and nitrogen prediction, demonstrating the potential of cross-modal correspondence learning to bridge the gap between satellite and UAV sensing in agricultural monitoring.

أدت التطورات الأخيرة في تصنيف النباتات إلى زيادة استخدام المنصات متعددة المستشعرات لجمع بيانات انعكاس مظلة المحاصيل. تعتبر الطائرات بدون طيار (UAV) فعالة بشكل خاص في مراقبة المحاصيل، بينما تواجه المهام الفضائية قيودًا في الدقة المكانية. تقترح هذه الدراسة استراتيجية تعلم عبر الأنماط تجمع بين الصور الفضائية عالية الدقة مع تفاصيل على مستوى UAV لتحسين تقدير خصائص مظلة المحاصيل، باستخدام مجموعة بيانات من 84 نوعًا هجينًا من الذرة.

Los avances recientes en fenotipado de plantas han impulsado la adopción generalizada de plataformas multi-sensoras para la recolección de datos de reflectancia del dosel de cultivos. Los Vehículos Aéreos No Tripulados (UAV) son especialmente efectivos para el monitoreo de cultivos, mientras que las misiones satelitales enfrentan limitaciones en la resolución espacial. Este trabajo propone una estrategia de aprendizaje cross-modal que enriquece las imágenes satelitales de alta resolución con detalles visuales a nivel de UAV para la estimación de rasgos del dosel de cultivos, utilizando un conj…

Des avancées récentes en phénotypage des plantes ont conduit à une adoption accrue de plateformes multi-capteurs pour la collecte de données de réflectance du couvert végétal. Les véhicules aériens sans pilote (UAV) sont particulièrement efficaces pour le suivi des cultures, tandis que les missions satellites rencontrent des limitations en termes de résolution spatiale. Cette étude propose une stratégie d'apprentissage cross-modal qui combine des images satellites haute résolution avec des détails de niveau UAV pour améliorer l'estimation des traits du couvert végétal, en utilisant des données…

Recent advancements in plant phenotyping have led to the increased use of multi-sensor platforms for collecting crop canopy reflectance data. Unmanned Aerial Vehicles (UAVs) are particularly effective for crop monitoring, while satellite missions face limitations in spatial resolution. This study proposes a cross-modal learning strategy that combines high-resolution satellite imagery with UAV-level detail to enhance crop canopy trait estimation, utilizing data from 84 hybrid maize varieties.

Crossmodal learning for Crop Canopy Trait Estimation

arXiv:2601.07975v1 Announce Type: new 
Abstract: High-resolution UAV photogrammetry has become a key technology for precision agriculture, enabling centimeter-level crop monitoring and point-level plant localization. However, point-level maize localization in UAV imagery remains challenging due to (1) extremely small object-to-pixel ratios, typically less than 0.1%, (2) prohibitive computational costs of quadratic attention on ultra-high-resolution images larger than 3000 x 4000 pixels, and (3) agricultural scene-specific complexities such as sparse object distribution and environmental variability that are poorly handled by general-purpose vision models.
  To address these challenges, we propose the Additive Kolmogorov-Arnold Transformer (AKT), which replaces conventional multilayer perceptrons with Pade Kolmogorov-Arnold Network (PKAN) modules to enhance functional expressivity for small-object feature extraction, and introduces PKAN Additive Attention (PAA) to model multiscale spatial dependencies with reduced computational complexity. In addition, we present the Point-based Maize Localization (PML) dataset, consisting of 1,928 high-resolution UAV images with approximately 501,000 point annotations collected under real field conditions.
  Extensive experiments show that AKT achieves an average F1-score of 62.8%, outperforming state-of-the-art methods by 4.2%, while reducing FLOPs by 12.6% and improving inference throughput by 20.7%. For downstream tasks, AKT attains a mean absolute error of 7.1 in stand counting and a root mean square error of 1.95-1.97 cm in interplant spacing estimation. These results demonstrate that integrating Kolmogorov-Arnold representation theory with efficient attention mechanisms offers an effective framework for high-resolution agricultural remote sensing.

يمثل تقديم المحول الإضافي كولموغوروف-أرنولد (AKT) تقدمًا كبيرًا في تحديد موقع الذرة على مستوى النقاط باستخدام صور الطائرات بدون طيار، حيث يتناول التحديات مثل النسب المنخفضة بين الكائنات والبكسل وتكاليف الحوسبة العالية المرتبطة بالنماذج التقليدية. تعزز هذه الطريقة المبتكرة استخراج الميزات للكائنات الصغيرة وتقدم نمذجة مكانية متعددة المقاييس من خلال الانتباه الإضافي PKAN (PAA).

La introducción del Transformador Aditivo Kolmogorov-Arnold (AKT) representa un avance significativo en la localización de maíz a nivel de puntos utilizando imágenes de UAV, abordando desafíos como las bajas proporciones objeto-píxel y los altos costos computacionales asociados con modelos tradicionales. Este enfoque innovador mejora la extracción de características para objetos pequeños e introduce un modelado espacial multiescalar a través de la Atención Aditiva PKAN (PAA).

L'introduction du Transformateur Additif Kolmogorov-Arnold (AKT) représente une avancée significative dans la localisation de maïs au niveau des points à l'aide d'images UAV, en répondant à des défis tels que les faibles rapports objet-pixel et les coûts computationnels élevés associés aux modèles traditionnels. Cette approche innovante améliore l'extraction des caractéristiques pour les petits objets et introduit une modélisation spatiale multiscale grâce à l'Attention Additive PKAN (PAA).

The introduction of the Additive Kolmogorov-Arnold Transformer (AKT) marks a significant advancement in point-level maize localization using UAV imagery, addressing challenges such as low object-to-pixel ratios and high computational costs associated with traditional models. This innovative approach enhances feature extraction for small objects and introduces multiscale spatial modeling through PKAN Additive Attention (PAA).

An Efficient Additive Kolmogorov-Arnold Transformer for Point-Level Maize Localization in Unmanned Aerial Vehicle Imagery

One More Thing in AI – Your Shortcut to AI Mastery

Crossmodal learning for Crop Canopy Trait Estimation

Was this article worth reading? Share it

One More Thing in AI

LucidQuery AI

Airparser

Attentive AI

Vermeer

Research AI

Ready to build your own newsroom?