Home // CADAT 2025, The Second International Conference on Accessible Digital Agriculture Technologies // View article

Corn Tassel Emergence Identification and Height Mesurment Based on Unmanned Aerial Vehicles

Authors:
Junfeng Ma
Dongliang Chu
Xukang Lyu
Aiqin Hou
Leping Feng
Xu Wang
Chase Wu

Keywords: Unmanned Aerial Vehicles (UAVs), Real-Time Kinematic (RTK), Deep Learning.

Abstract:
In breeding test fields, where tens or even hundreds of thousands of corns are planted, measurements of numerous phenotypic traits—such as plant height, tassel height, stem thickness,fruiting characteristics (e.g., tassel length, tassel width, awnless tip, row number), disease resistance, and lodging resistance—are typically required. Traditional methods rely on pen-and-paper recordings or basic spreadsheets, which are highly inefficient and prone to human errors, including serial mistakes and incorrect data entries. This makes it difficult to ensure data accuracy and quality. To address these challenges, this paper explores the use of Unmanned Aerial Vehicles (UAVs) and deep learning technologies to monitor the entire growth process of corn plants throughout their life cycle and select high-quality seedlings. Using experiments conducted in corn fields in Henan Province as a case study, the research focuses on identifying the growth and development stages of corn plants, as well as monitoring the timing of tassel emergence. A high-quality dataset covering the entire growth and development process is constructed. Based on UAV remote sensing images with Real-Time Kinematic (RTK) coodinates and timestamps, and 3D point cloud coordinates, we employ You Only Look Once (YOLO)v8 to conduct object detection to accurately identify tassel emergence times during growth. We also collect images of mature corn plants and their point clouds to calculate the height of each mature corn. These approachs aim to achieve precise monitoring of corn growth conditions and facilitate the digital and precise management of the corn cultivation process.

Pages: 9 to 14

Copyright: Copyright (c) IARIA, 2025

Publication date: October 26, 2025

Published in: conference

ISBN: 978-1-68558-328-6

Location: Barcelona, Spain

Dates: from October 26, 2025 to October 30, 2025