Deep learning in plant phenotyping: the first ten years

2025/08/21 by Jordan Ubbens, Ian Stavness, Michael P. Pound +1 · 1 voice
Agricultural and Biological Sciences · Biochemistry, Genetics and Molecular Biology · Environmental Science · #Cell Image Analysis Techniques #Remote Sensing in Agriculture #Smart Agriculture and AI

paper · doi:10.1016/j.plaphe.2025.100062

openalex publication_date 2025/08/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/04

Abstract

As with many fields of science, plant science and agriculture have seen a rapid adoption of deep learning in recent years. The present moment is significant as it marks one decade since the first applications of deep learning began to appear in the literature on plant phenotyping. In this short time, a new research community was founded and new connections between computer vision and biology were established. In this letter, we reflect on this critical period of time from the inception of the field to where it stands today.

Citations

• Image Analysis: The New Bottleneck in Plant Phenotyping [Applications Corner]
• ImageNet classification with deep convolutional neural networks
• Denoising Diffusion Probabilistic Models
• NeRF
• High-Resolution Image Synthesis with Latent Diffusion Models
• 3D Gaussian Splatting for Real-Time Radiance Field Rendering
• Adding Conditional Control to Text-to-Image Diffusion Models
• Segment Anything
• High-fidelity wheat plant reconstruction using 3D Gaussian splatting and neural radiance fields
• AgriNeRF: Neural Radiance Fields for Agriculture in Challenging Lighting Conditions
• FruitNeRF: A Unified Neural Radiance Field based Fruit Counting Framework
• T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy
• Evaluating Neural Radiance Fields (NeRFs) for 3D Plant Geometry Reconstruction in Field Conditions
• 3D Gaussian Splatting for Real-Time Radiance Field Rendering
• Leaf Only SAM: A Segment Anything Pipeline for Zero-Shot Automated Leaf Segmentation
• DINOv2: Learning Robust Visual Features without Supervision
• Micrograph segmentations for DDEVD
• Adding Conditional Control to Text-to-Image Diffusion Models
• High-Resolution Image Synthesis with Latent Diffusion Models
• Extending the WILDS Benchmark for Unsupervised Adaptation
• LoRA: Low-Rank Adaptation of Large Language Models
• WILDS: A Benchmark of in-the-Wild Distribution Shifts
• Unsupervised Domain Adaptation For Plant Organ Counting
• Global Wheat Head Detection (GWHD) dataset: a large and diverse dataset of high resolution RGB labelled images to develop and benchmark wheat head detection methods
• Deep Leaf Segmentation Using Synthetic Data
• Using Deep Learning for Image-Based Plant Disease Detection
• An open access repository of images on plant health to enable the development of mobile disease diagnostics

Discussions

• A decade of deep learning in plant phenotyping: from first papers to a thriving research community bridging computer vision and biology. Our reflection on this transformative journey. 🌱🤖 #PlantPheno [bsky, 1 points, 0 comments]

Related

• Image Analysis: The New Bottleneck in Plant Phenotyping [Applications Corner]
• Applications of Machine Learning Methods to Quantifying Phenotypic Traits that Distinguish the Wild Type from the Mutant Arabidopsis Thaliana Seedlings during Root Gravitropism
• Deep-Plant: Plant Identification with convolutional neural networks
• Adaptive Transfer Learning for Plant Phenotyping
• PhenoAssistant: A Conversational Multi-Agent AI System for Automated Plant Phenotyping