Easy MPE: Extraction of Quality Microplot Images for UAV-Based High-Throughput Field Phenotyping

Area of Science:

• Agricultural Science
• Remote Sensing
• Image Processing

Background:

• Microplot extraction is crucial for plant breeding research using drone (UAV) imagery.
• Current manual methods using GIS software are time-consuming and labor-intensive.
• Accurate plot data is essential for reliable field research outcomes.

Purpose of the Study:

• To develop an intuitive and efficient semiautomatic program for microplot extraction (MPE).
• To provide researchers with reliable plot data from UAV images under diverse field conditions.
• To streamline the image processing workflow in plant breeding studies.

Main Methods:

• The Easy MPE program employs a four-step process: binary segmentation, microplot extraction, .shp file generation, and projection onto raw UAV images.
• The method successfully identifies crop rows in various field conditions.
• Performance was validated using the intersection-over-union (IOU) metric against manual extraction.

Main Results:

• The Easy MPE program demonstrated high accuracy in microplot extraction.
• The average IOU ratio between Easy MPE and manual extraction was 91% (±3%).
• Crop rows were consistently identified across all tested fields.

Conclusions:

• Easy MPE offers a reliable and efficient solution for semiautomatic microplot extraction from UAV imagery.
• The program significantly reduces the time and effort required for processing plant breeding field data.
• This tool enhances accessibility to accurate plot data for researchers, improving plant breeding research efficiency.