Smart Agriculture and AI

A miniaturized Arabidopsis thaliana - Botrytis cinerea seedling pathosystem and the GreenDayz workflow for longitudinal image-based phenotyping

A high-throughput UAV millimetric imaging pipeline for plant density phenotyping reveals high heritability in wheat-pea sole crops and intercrops.

The paper develops a UAV-based imaging pipeline to predict plant density in wheat and pea crops, showing high heritability across diverse intercropping scenarios.

Multistage AI Crop Inspection and Segmentation Framework (MACISF)

MACISF is a deep learning framework for automated fruit quality evaluation, achieving high accuracy in classification, health detection, and defect localization.

Vision Normalizing Flows for the probability-informed detection of banana diseases from in-field images

The paper develops a vision normalizing flow model to detect banana diseases from in-field images with high accuracy, using pre-trained embeddings and probabilistic density estimation.

Grazing cattle respond more strongly to a quadruped robot than to a drone but habituate over consecutive days

A Large Yield Model for Crop Production and Design in Western Canada

This paper introduces LYM-1, a multi-crop yield prediction model for Canadian prairies using extensive yield, weather, and soil data, showing improved predictive performance and revealing agronomically relevant interactions.

Evaluating low-cost hyperspectral and RGB fusion with PSNet for presymptomatic detection of Albugo candida in Arabidopsis thaliana

This paper evaluates PSNet, a multimodal framework combining hyperspectral and RGB imaging, for early detection of Albugo candida in Arabidopsis thaliana, showing high accuracy but no consistent advantage of hyperspectral fusion over RGB alone.

A Novel Phenotyping Approach for Reconciling Precision and Variance in Disease Severity Estimates from High-resolution Imaging

The paper develops an imaging approach to improve disease severity estimation in wheat by combining high-resolution imaging and focus bracketing, reducing measurement uncertainty and enabling efficient sampling.

Petal to the metal: The slow road to automating large-scale phenology labeling for herbarium specimens

The paper presents a machine learning pipeline for automating flower detection on herbarium specimens, resulting in 11.1 million labeled records but highlighting the need for full metadata digitization.

From Plant Detection to Satellite Mapping: A Multi-Scale AI Toolkit for Ragweed Surveillance under Climate Change

This paper develops and evaluates an AI toolkit for detecting and mapping ragweed under climate change, showing improved performance with multi-domain training and satellite-based models.

Drought-Spec-Net: Early Tomato Drought Detection and Potential Yield-Impact Assessment Using Vis NIR Data

This paper develops Drought-Spec-Net, a model for early tomato drought detection using Vis NIR data, achieving high accuracy and providing yield impact assessments.

YieldLearn: An interpretable machine-learning pipeline for estimating yield-related traits based on physiological indicators

YieldLearn estimates yield-related traits in durum wheat using physiological indicators and machine learning, showing moderate accuracy and feature importance.

DIRTNet: Enabling Root Phenotyping with Fiber Bragg Grating Sensors

DIRTNet, a hybrid deep learning model, accurately predicts root depth and diameter from FBG sensor data in both simulated and real maize root growth scenarios.

A breeder in the sky: scoring flowering with fewer flights

This paper introduces a regression-based framework to predict flowering time from sparse UAV observations, reducing the need for repeated data collection.

Embeddings from standardized sorghum leaf images capture variation in disease response that human scoring misses

The paper uses standardized leaf imaging and vision encoders to capture disease response variation missed by human scoring, identifying genomic hotspots linked to disease symptoms.

Segment Any Plant (SAP): Foundation-Model Segmentation for Plant Time-Series Phenotyping

SAP provides a transferable, training-free framework for automated segmentation of plant time-series imagery across species and conditions.

OneGrow: Unified Temporal Plant Image and Mask Generation

OneGrow generates temporal plant images and organ masks, addressing the lack of paired data by jointly modeling wheat images and masks over time.

Vision Transformers Enable Advanced Plant Phenotyping in Controlled Environments

This paper compares segmentation methods for plant phenotyping, finding that pretrained vision transformers outperform color thresholding and supervised U-Nets in generalization across species and conditions.

A physics-informed hybrid deep learning model for spatiotemporal rice disease prediction using multi-source data

The paper introduces a physics-informed hybrid deep learning model for spatiotemporal rice disease prediction, improving accuracy and interpretability using multi-source data.

Text guidance is powerful but prompt-sensitive for weakly-supervised leaf symptom segmentation

The paper evaluates text-guided segmentation using SAM3 for weakly supervised leaf symptom segmentation, finding text guidance matches or outperforms conventional methods but is prompt-sensitive, with hybrid approaches improving robustness.

Automated Detection of Livestock Gastrointestinal Parasite Eggs and Cysts Using YOLOv8-Based Deep Learning

This study applies YOLOv8 for automated detection of livestock gastrointestinal parasite eggs and cysts in microscopic images, achieving moderate performance metrics but highlighting data limitations.

Quantifying Crop Disease Trait Dynamics through Longitudinal Imaging and Temporal Analytics

The paper presents a pipeline for longitudinal plant disease phenotyping using time-series imaging, enabling objective quantification of disease progression in wheat rusts.

AI-Powered Acoustic Surveillance for Early Detection of Calf Respiratory Disease

This paper develops an AI-powered acoustic monitoring system to detect coughs in calves, achieving 92% accuracy and enabling early detection of respiratory disease 1-2 days before clinical signs.

From video-derived feeding behaviour to cow-level nutritional deviation signals: A dairy digital-twin decision-support framework

This paper develops a digital-twin framework that integrates video-derived feeding behavior with physiological data to improve precision nutrition in dairy cows, showing reduced prediction errors in dry matter intake.

RADIX: a deep learning framework that maps root barriers across species and reveals genetic and environmental contributions

RADIX is a deep learning framework that segments root anatomical structures across species and platforms, revealing genetic and environmental influences on root barriers.

MIRA: an open source and user-friendly software to automate counting and sizing of fungal spores

MIRA automates fungal spore counting and sizing using YOLO deep learning, achieving high accuracy and efficiency compared to manual methods.

A Data-Driven Image Extraction and Analysis Pipeline for Plant Phenotyping in Controlled Environments

The paper develops an automated image analysis pipeline for plant phenotyping in controlled environments, using multispectral imaging and AI to extract and analyze phenotypic features across multiple crops.

A Bayesian method for estimation of plant soil water content with application to low-cost horticultural robotics

The paper develops a Bayesian method to estimate soil water content from capacitive sensor data, enabling low-cost plant watering systems without destructive calibration.

PolliCrop: A high-throughput computer vision pipeline for pollinator monitoring in agroecosystems

PolliCrop is a computer vision pipeline that detects and identifies flower-visiting insects on sunflower plants, showing accurate performance in predicting visitation frequencies across genotypes.

ROOTQUANT: AUTOMATED ROOT TRAIT QUANTIFICATION FROMMINIRHIZOTRON IMAGES USING DEEP LEARNING

RootQuant uses deep learning to automatically quantify root traits from minirhizotron images, achieving high accuracy across species without pixel-level annotations.