PhenoHUB is a WeChat Miniapp focused on plant photosynthetic phenotyping research, integrating multiple scientific tools and AI analysis functions to help researchers quickly obtain and analyze plant phenotyping data in the field.
Botanical Extract AI Pro is a project designed to address the challenges of background interference and high manual segmentation costs in traditional plant phenotyping analysis. By leveraging the visual understanding and generation capabilities of multimodal AI models, the system achieves zero-shot high-precision plant image background removal and segmentation. The system supports both single-image interactive processing (Web end) and large-scale batch processing (script end), providing efficient data preprocessing tools for botanical research.
MCTP (Multi‑modal Crop Trait Processing) is a unified data processing platform designed for plant phenotyping workflows. It brings hyperspectral, LiDAR, RGB, and thermal imaging into a single pipeline with consistent GUI experiences, reproducible parameter tuning, batch processing, and standardized outputs—making multi‑modal analysis easier to manage and scale.
MCTP is a self-developed field walking phenotyping platform by Shufeng Bio, and I was responsible for system optimization and data processing and analysis during the development process.
Light distribution within crop canopies determines how efficiently plants convert sunlight into biomass. Our latest study presents a new framework that links leaf anatomy and physiology to optical properties, providing a pathway toward predictive modeling of canopy photosynthesis.
We developed a novel Directional Spectrum Detection Instrument (DSDI) and an ensemble learning (EL) model that accurately predict Bidirectional Reflectance Distribution Function (BRDF) parameters from measurable phenotypic traits.
This work integrates optical physics, phenotyping, and data-driven modeling to enable computational quantification of leaf optical diversity—a key step toward designing crop canopies with higher light-use efficiency.
Deploying AI agents locally offers numerous advantages including data privacy, reduced latency, cost control, and independence from cloud services. This comprehensive guide covers multiple approaches to setting up AI agents on your local infrastructure, from simple chatbots to complex multi-modal systems.
This tutorial introduces a powerful standalone Python tool for agricultural image quantification analysis. The tool provides automated detection and measurement of plant samples from images, including morphological and color metrics.
This guide explains the entire process of publishing a scientific paper, from manuscript preparation to final publication, suitable for graduate students and researchers.
This comprehensive guide covers the deployment, optimization, and academic research applications of Hunyuan3D for plant 3D reconstruction, with a focus on cotton plant point cloud generation and phenotyping analysis.
UAV image acquisition employs a CCO (Cross-Complementary Overlap) flight path design strategy. This strategy enhances viewpoint diversity through multi-directional cross-flight paths to strengthen geometric constraints for 3D reconstruction.
Based on industry-grade UAV platforms, multi-view RGB images of farmland are collected to provide unified data sources for subsequent orthophoto mosaic and 3D modeling.
Root Quantify is a Python-based tool designed for processing root system images. It provides a complete pipeline for analyzing root images with the following capabilities:
