Smart Agriculture and AI

Decrypting the complex phenotyping traits of plants by machine learning

AMULET predicts plant phenotypes using imaging and machine learning before visible changes occur, improving breeding and agricultural management.

Remote Phenotyping Strategies for Sunflower Flowering Assessments Using Deep Learning Approaches

This paper develops a deep learning method to estimate sunflower flowering dates using UAV-acquired RGB images, achieving high correlation with ground measurements and improving heritability estimates.

Breaking the barrier of human-annotated training data for machine-learning-aided plant research using aerial imagery

A machine learning approach accelerates plant phenotyping by reducing the need for large amounts of human-annotated data, enabling faster and more efficient detection of a plant trait using aerial imagery.

Decoding Bovine Communication with AI and Multimodal Systems ~ Advancing Sustainable Livestock Management and Precision Agriculture

This paper develops an AI system to decode bovine vocalizations and behavior using multimodal data, enabling real-time monitoring for improved livestock management and sustainability.

AnimalMotionViz: an interactive software tool for tracking and visualizing animal motion patterns using computer vision

AnimalMotionViz tracks and visualizes dairy cattle movement patterns using computer vision to generate motion heatmaps and analyze space utilization.

Towards high throughput in-field detection and quantification of wheat foliar diseases with deep learning

This paper develops a deep learning method to detect and quantify wheat foliar diseases using field images, achieving high accuracy in keypoint detection and segmentation.

Enhancing tree seed germination prediction with image-driven machine learning models.

This paper uses image-driven machine learning to predict tree seed germination, finding that X-ray images are better at detecting empty seeds while color images struggle with variability and differentiation.

The FIP 1.0 Data Set: Highly Resolved Annotated Image Time Series of 4,000 Wheat Plots Grown in Six Years

The paper introduces a large-scale wheat phenotyping dataset with time series imaging and trait data across six years, enabling studies on genotype-environment interactions and crop modeling.

Assessing nutritional pigment content and plant health status of green and red leafy vegetables by image analysis: Catching the 'red herring' of plant digital color processing

This study develops digital color analysis models to accurately assess pigment content and plant health in green and red leafy vegetables, showing that certain color features correlate strongly with chlorophyll and anthocyanin levels.

Dairy DigiD ~ A Deep Learning-Based, Non-Invasive Biometric Identification System for Dairy Cattle Using Detectron2

Dairy DigiD uses deep learning and facial images to classify dairy cattle into four age categories with high accuracy, offering a non-invasive alternative to traditional identification methods.

CBAM-ResNet34-based classification and evaluation method for developmental processes of greenhouse strawberries

The paper proposes an improved CBAM-ResNet34 model for classifying greenhouse strawberry developmental stages, achieving high accuracy and F1 scores.

RSG-YOLO: Detection of rice seed germination rate based on enhanced YOLOv8 and multi-scale attention feature fusion

This paper introduces RSG-YOLO, an enhanced YOLOv8 model for rice seed germination rate detection, achieving a 4% improvement in mAP50 over YOLOv8.

Cross-dataset evaluation of deep learning models for plant pest and disease diagnosis

This study evaluates deep learning models for plant pest and disease diagnosis across different datasets, finding that performance degrades when models are tested on new datasets.

Weed Instance Segmentation from UAV ortho-mosaic Images based on Deep Learning

This paper develops deep learning models to segment weeds in UAV ortho-mosaic images and analyzes their impact on potato yield.

Highly Accurate and Precise Determination of Mouse Mass Using Computer Vision

This paper demonstrates a computer vision method to estimate mouse body mass from video data with 4.8% error, enabling noninvasive and continuous monitoring.

Development of a Cloud-Based IoT System for Livestock Health Monitoring Using AWS and Python

This paper develops a cloud-based IoT system for livestock health monitoring using AWS and Python, demonstrating its scalability and effectiveness in tracking health indicators and enabling predictive analytics.

Artificial Intelligence driven Benchmarking Tool for Emission Reduction in Canadian Dairy Farms

This study develops an AI-driven benchmarking tool to reduce methane emissions in Canadian dairy farms using machine learning and geospatial analysis, showing LSTM models provide the most accurate emission forecasts.

MIPDB: A maize image-phenotype database with multi-angle and multi-time characteristics

The paper introduces MIPDB, a maize image-phenotype database with multi-angle and multi-time images, providing a high-quality dataset for deep learning-based phenomics research.

StomaVision: stomatal trait analysis through deep learning

StomaVision automates stomatal trait analysis from micrographs, revealing physiological responses to heat stress across plant species.

Fruit-In-Sight: a deep learning-based framework for secondary metabolite class prediction using fruit and leaf images

Fruit-In-Sight uses deep learning to predict secondary metabolite concentration classes from fruit and leaf images, achieving high accuracy without lab equipment.

Sunpheno: a deep neural network for phenological classification of sunflower images

Sunpheno uses a pre-trained ResNet50 model to classify sunflower phenological stages from field images, showing high accuracy and speed.

High-throughput Phenotyping of Soybean Biomass: Conventional Trait Estimation and Novel Latent Feature Extraction Using UAV Remote Sensing and Deep Learning Models

This study uses UAV remote sensing and deep learning to estimate soybean biomass traits and extract latent features for breeding applications.

Temporal Image Sandwiches Enable Link between Functional Data Analysis and Deep Learning for Single-Plant Cotton Senescence

This paper links functional data analysis with deep learning to analyze single-plant cotton senescence using temporally concatenated UAS images.

Impact evaluation of score classes and annotation regions in deep learning-based dairy cow body condition prediction

This study evaluates how deep learning models predict dairy cow body condition scores using image annotations, finding stronger agreement with fewer classes and larger error ranges.

High-Throughput Image-Based Plant Stand Count Estimation Using Convolutional Neural Networks

This paper introduces DeepStand, a deep learning method for high-throughput corn stand counting using images, outperforming existing methods.

High throughput measurement of Arabidopsis thaliana fitness traits using deep learning

This study uses deep learning to accurately measure Arabidopsis fitness traits, revealing gene function effects previously undetected by traditional methods.

Computational Prediction of Disease Detection and Insect Identification using Xception model

The paper compares deep learning models for crop disease and pest detection, finding Xception achieves 82.89% accuracy for disease and 77.9% for pests.

GridFree: A Python Package of Image Analysis for Interactive Grain Counting and Measuring

GridFree is a Python software package that automates grain counting and measurement using unsupervised machine learning, achieving lower error rates than existing tools across multiple crops.

DeepCorn: A Semi-Supervised Deep Learning Method for High-Throughput Image-Based Corn Kernel Counting and Yield Estimation

DeepCorn uses semi-supervised deep learning to accurately count corn kernels in-field, improving yield estimation and data collection for agriculture.

Integrating diverse data sources to predict disease risk in dairy cattle

This study predicts eight diseases in dairy cattle by integrating environmental, behavioural, and physiological data using machine learning and regression analysis.