UAV technology works to improve rice breeding and selection

Rice researchers at the Texas A&M AgriLife Research Center in Beaumont have taken rice cultivar breeding and selection to the next level with a project that uses an aerial vehicle (UAV). The team will use drones to take real-time snapshots of rice crops, remove crop phenotypic traits from those images, and analyze the information to reveal high-yielding rice genotypes, a Texas A&M press release shared. AgriLife.

Researchers hope to avoid one of the main hurdles in data collection – the cumbersome and time-consuming procedure of manually collecting data in the field by skilled labor. Yubin Yan, senior biosystems analyst at the Beaumont center, will lead the project with funds provided by a three-year, $650,000 grant from the USDA’s National Institute of Food and Agriculture (NIFA). .

Key research objectives:

  • Calculate key phenotypic traits for rice growth and development.
  • Record UAV images of rice genotypes at crucial rice growth stages.
  • Create advanced image processing algorithms to extract the main phenological, morphological and architectural traits.
  • Generate a digital rice breeding system to select the best performing genotypes using data integration with multi-trait decision making.

The next wave of UAV technology

“Traditional manual measurement of rice phenotypic traits is very, very time-consuming,” Yang explained. “It is becoming more and more difficult to hire qualified and experienced personnel. UAV technology and advanced image processing could potentially provide a cost effective and reliable alternative. We can use drones to capture images of rice at key growth stages and develop algorithms to extract different phenotypic traits for hundreds or even thousands of rice genotypes.

Multiple UAV flights will be executed throughout the rice harvest season to capture thousands of UAV images, as well as ground truth information. Different camera angles will be used to help analyze stand establishment, as well as gaps between plants.

“A considerable amount of data will need to be integrated and analyzed,” Yang added. “It’s the first year of the project and it’s a learning process for us. Timely capture of UAV images for early rice growth has been difficult due to small size of rice seedlings and windy weather conditions. There is a limited window during which you can fly.

The team will also work to develop machine learning algorithms that can identify key traits and determine the best performing rice genotypes. The project will focus on essential phenotypic traits including stand establishment, biomass growth, final grain yield and phenological development.

“We will develop automated algorithms capable of extracting phenotypic traits from UAV images taken at critical stages of rice, including sowing, tillering, flowering, grain filling and maturity,” said Yang. “The digital rice breeding system will be developed through the integration of multiple traits to identify the most successful genotypes.”

Researchers believe another feature of UAV technology could be monitoring plant growth for nitrogen control and disease detection, explained Fugen Dou, a fellow scientist at AgriLife Research.

“This proposed project represents a major effort to provide an integrated decision-making system based on UAV imagery to rice breeders and researchers,” Yang concluded. “It will be an indispensable tool to greatly improve the efficiency of rice breeding and phenotyping.”

Learn more about rice progress:

New mid-density SNP panel developed for US rice

Researchers in Japan reveal tools to improve rice production

Provivi and Syngenta Crop Protection launch pheromone-based Nelvium to control rice pests

Filipino researcher discovers drought resistance gene in rice