Smart Agriculture and AI

A multiregional image-text dataset and benchmark for vision-language modeling of plant diseases

The paper introduces LeafMD, a multiregional image-text dataset and benchmark for vision-language modeling of plant diseases, highlighting performance gaps and benefits of agriculture-adapted models.

SeedMeasure: an efficient approach and open-source program to quantify seed size

SeedMeasure quantifies seed size and morphology from images, offering a fast, open-source, and cross-platform tool validated across nine species.

GuavaVision AI: An Explainable Deep Learning Framework for Automated Classification, Lesion Localization, and Segmentation of Guava Diseases

This paper develops a deep learning framework for classifying, localizing, and segmenting guava diseases using image analysis, achieving high accuracy and transparency.

Beyond Accuracy: Reliability-Aware Cross-Farm Evaluation of Dairy Cow Vocalization Models

This paper evaluates dairy cow vocalization models across farms, showing that reliability varies by vocalization type and that cross-farm generalizability is limited for certain categories.

Leaf movements as a quantitative metric for early stress detection

This paper develops a method to detect early plant stress using leaf movement analysis in lettuce and other crops, detecting stress within 1 hour using low-cost imaging and open-source algorithms.

Linking automated image analysis to ecological inference: high-throughput monitoring of soil fauna

This paper develops an efficient image processing pipeline for soil fauna monitoring by combining traditional computer vision with fine-tuned foundation models, reducing annotation effort and enabling large-scale ecological analysis.

AGIcam: An open-source IoT-based camera system for automated in-field phenotyping and yield prediction

AGIcam provides an IoT-based camera system for continuous in-field plant phenotyping and yield prediction using time-series imagery and machine learning models.

An Expert-Informed Synthetic Animal Data Generator: A Physiology-Consistent Generative Framework for High-Fidelity Animal Digital Twins

This paper introduces EICD, a framework that generates physiology-consistent synthetic animal data by embedding mechanistic biological principles, achieving high fidelity and reducing physiological hallucinations.

Aerial imagery and deep learning accurately estimate maize foliar disease severity

This paper uses aerial imagery and deep learning to estimate maize foliar disease severity, achieving results comparable to expert visual assessments.

Image-based leaf SPAD value and chlorophyll measurement using a mobile phone: enabling accessible and sustainable crop management

This paper develops a mobile phone-based app to estimate leaf SPAD and chlorophyll values in rice with accuracy comparable to lab methods and commercial tools.

LeafyVGG-16: Transfer Learning for Plant Disease Detection with Cyber Risk Analysis

This paper introduces LeafyVGG-16, a transfer learning model for plant disease detection, achieving high accuracy and evaluating its vulnerability to adversarial cyber attacks.

EpiReasoner: An Integrated Artificial Intelligence Framework for Phenotype-to-Genotype Reasoning in Plant Epidermal Development

EpiReasoner integrates AI for phenotype-to-genotype reasoning in plant epidermal development, identifying a major QTL for stomatal density in tomatoes.

Easy to use and low cost leaf disease quantification workflow using Ilastik

The paper presents a semi-automated image analysis workflow for quantifying foliar disease symptoms in wheat using Ilastik, achieving high segmentation accuracy with limited training data.

LIME: a fully automated pipeline for high-throughput quantification of leaf lesions

LIME is an automated image analysis pipeline for quantifying leaf lesions in plant disease assays, achieving high accuracy and significantly reducing processing time compared to manual methods.

Multi-Scale Contextual Attention for Robust Crop and Pest Image Classification

This paper introduces a Multi-Scale Contextual Attention module for crop and pest image classification, achieving high accuracy on a field-style dataset while outperforming baselines.

Potato yield can be predicted using drone-captured and environmental measurements early in the growing season

This study predicts potato yield using drone and environmental data early in the growing season, achieving over 80% accuracy with models trained on limited data.

Automated Landmark-Based Root Inoculation in Arabidopsis Using Computer Vision and Robotics

This paper develops an automated system for root inoculation in Arabidopsis using computer vision and robotics, achieving 100% accuracy in inoculating seedlings.

Phenotypic scoring of Canola Blackleg severity using machine learning image analysis

This paper introduces a deep learning model to score Canola Blackleg severity from images, showing it is more consistent than expert raters and maintains similar heritability for breeding purposes.

Transformers Outperform ConvNets for Root Segmentation: A Systematic Comparison Across Nine Datasets

The paper compares Transformer and ConvNet models for root segmentation across nine datasets, finding Transformers significantly outperform ConvNets in accuracy and root-diameter agreement.

PlantCV v4: Image analysis software for high-throughput plant phenotyping

PlantCV v4 improves usability and functionality for plant phenotyping by adding new image types and tutorials for non-coding researchers.

Samplify: A versatile tool for image-based segmentation and annotation of seed abortion phenotypes

Samplify automates seed segmentation and classification in Arabidopsis, improving efficiency and reducing user bias in seed abortion phenotyping.

Towards an AI-based knowledge assistant for goat farmers based on Retrieval-Augmented Generation

This paper develops an AI-based knowledge assistant for goat farmers using Retrieval-Augmented Generation to improve LLM performance on structured knowledge tasks.

Transformer-Based Phenotyping of Rice Root Aerenchyma Across Environments Enables Climate-Smart Rice Selection

This paper develops a transformer-based deep learning pipeline to automate aerenchyma quantification in rice roots across environments, achieving expert-level accuracy and supporting climate-smart breeding.

Smartphone image capture system and image analysis pipelines enable accurate and efficient phenotyping of spaced plant mapping populations

The paper demonstrates a smartphone-based image capture system and two image analysis pipelines for efficient and accurate plant phenotyping in large field experiments.

StomaQuant: Deep Learning-Based Quantification for Stomatal Trait Assessment

StomaQuant uses deep learning to quantify stomatal density and size across diverse plant species, outperforming existing models with high accuracy.

Strategic Segmentation: A Nash Equilibrium based Approach for Weed Segmentation in Agricultural Fields

This paper develops a deep learning method using game theory to improve weed segmentation in agricultural fields by finding a Nash equilibrium to reduce false positives.

Automated detection and quantification of two-spotted spider mite life stages using computer vision for high-throughput in vitro assays

This paper develops a computer vision pipeline for automated detection of two-spotted spider mite life stages, achieving high accuracy in fecundity estimation but lower accuracy in mortality assessment.

CitriBEiTNet: A Hybrid CNN-Transformer Architecture Combining MobileNetV2 with BEiT's Global Attention for Automated Citrus Leaf Disease Diagnosis

CitriBEiTNet combines MobileNetV2 and BEiT for citrus leaf disease classification, achieving high accuracy in detecting diseases and healthy leaves.

Deep Learning for Sorghum Yield Forecasting using Uncrewed Aerial Systems and Lab-Derived Imagery

This paper uses deep learning models to detect sorghum panicles from aerial and lab images, improving yield prediction through machine learning regression.

Research on the Estimation Model of Seedling Sowing Period Based on Different Altitude

This study develops an LSTM model to estimate optimal tobacco seedling sowing periods based on altitude, achieving high accuracy in predicting survival rates.