Revolutionizing Rice Breeding and Cultivation: An Open and AI-Powered Cloud Computing Platform for Quantifying Rice Panicles and Classifying Yield Performance

Rice (Oryza sativa) is a staple food for over half of the world’s population. However, rice production is facing numerous challenges, including climate change, population growth, and the need for increased yield. To address these challenges, researchers have turned to computer vision and artificial intelligence (AI) to quantify yield-related traits, such as panicle number per unit area (PNpM2), in rice fields.

Recent studies have shown that deep learning techniques can effectively detect rice panicles in small-scale trials. However, these methods face scalability challenges, as they require large, high-quality training datasets and diverse rice varieties. Overcoming these challenges is crucial for developing robust, large-scale phenotypic analysis tools to improve rice production in a rapidly changing climate.

Plant Phenomics, a leading research organization in the field of plant phenomics, has developed an open and AI-powered cloud computing platform called Panicle-Cloud to address these challenges. The platform provides an efficient and accurate solution for quantifying rice panicles from drone-collected imagery and classifying yield performance based on the PNpM2 trait.

The first step in developing the Panicle-Cloud platform was to create an open expert-annotated diverse rice panicle detection (DRPD) dataset. This dataset consisted of over 10,000 annotated rice panicle images, which were used to train and validate several deep learning (DL) models. Through an iterative improvement process, the researchers found that the Panicle AI model, a custom DL model, demonstrated superior panicle detection accuracy.

To determine the optimal conditions for panicle phenotyping, the researchers analyzed drone flights at different altitudes and key growth stages. They found that an altitude of 7m during early grain-filling stages provided the most accurate results. This finding is crucial for ensuring that the Panicle-Cloud platform can accurately quantify rice panicles under various growing conditions.

The researchers then tested the correlation between Panicle-AI-derived scoring and manual counts. They found a high correlation coefficient (R2 = 0.945) between the two methods, particularly at the 7m height. This result confirmed the effectiveness of the Panicle AI model in detecting rice panicles and quantifying yield-related traits.

The Panicle-Cloud platform was designed to be user-friendly, allowing non-experts to select from different AI models for panicle detection using a simple web-based interface. Users can process images individually or in batches, and the platform optimizes computation by cropping larger images.

To classify yield performance based on the PNpM2 trait, the researchers used a supervised machine learning model, specifically the CatBoost algorithm. The platform successfully classified rice yield production into low, medium, and high categories with an overall accuracy of 84.03%. This feature allows rice breeders to effectively screen and select preferred varieties based on predicted yield performance.

The Panicle-Cloud platform represents a significant advancement in agricultural technology. It demonstrates the potential of integrating AI, cloud computing, and drone technology to revolutionize rice breeding and cultivation. The platform improves the efficiency and accuracy of quantifying yield-related traits and makes advanced phenotyping tools more accessible to a wider range of users.

The ability to classify yield performance based on the PNpM2 trait is particularly important in the context of global food demand challenges. By providing rice breeders with accurate and timely information on yield performance, the Panicle-Cloud platform can help increase the production of high-yielding rice varieties, thereby contributing to food security and sustainability.

In conclusion, the Panicle-Cloud platform is a game-changer in the field of rice research and agriculture. It provides an open and AI-powered cloud computing solution for quantifying rice panicles from drone-collected imagery and classifying yield performance based on the PNpM2 trait. The platform’s accuracy, efficiency, and user-friendliness make it an essential tool for rice breeders and researchers working to improve rice production in the face of global food demand challenges.

More information: Zixuan Teng et al, Panicle-Cloud: An Open and AI-Powered Cloud Computing Platform for Quantifying Rice Panicles from Drone-Collected Imagery to Enable the Classification of Yield Production in Rice, Plant Phenomics (2023). DOI: 10.34133/plantphenomics.0105

Provided by Plant Phenomics.

Explore further:

Transforming rice phenotyping: Advanced deep learning models enhance panicle analysis and nitrogen impact studies
Facebook
Twitter
Email

Feedback to editors

Featured
Last Comments
Popular
Slight male navigational advantage likely due to cultural differences, researchers find
2 HOURS AGO

0

Meta-analysis seeks to explain why dogs wag their tails
3 HOURS AGO

0

TESS discovers nine new ‘hot Jupiters’
4 HOURS AGO

0

Efficiency asymmetry: Scientists report fundamental asymmetry between heating and cooling
7 HOURS AGO

New details of the history of the Melanesian Border Plateau revealed by undersea rocks
JAN 16, 2024

0

Modeling framework finds fungal ‘bouncers’ patrol plant-microbe relationship
18 MINUTES AGO

Insect populations flourish in the restored habitats of solar energy facilities
22 MINUTES AGO

Big dogs vs. small dogs: Which sizes face higher risks of which diseases?
43 MINUTES AGO

Predicting others’ preference-based choices is cross-cultural and uniquely human, suggests study
43 MINUTES AGO

Wooly mammoth movements tied to earliest Alaska hunting camps
43 MINUTES AGO

Certain personality traits linked to college students’ sense of belonging
43 MINUTES AGO

Molecular sensor enables water bear hardiness by triggering dormancy, study finds
43 MINUTES AGO

New technology for conducting deep-sea research on fragile organisms
43 MINUTES AGO

Scientists develop luminescent sensor to detect ‘forever chemicals’ in water
1 HOUR AGO

Searching for dark matter in gaps between stars
1 HOUR AGO

Relevant PhysicsForums posts
A phosphoric acid question
JAN 16, 2024
And Now, here comes COVID-19 version BA.2, BA.4, BA.5,…
JAN 16, 2024
Nick Lanes on Sean Carroll’s podcast
JAN 15, 2024
Innovative Evolution of the Bamboo-Eating Pandas: A Unique Survival Strategy
JAN 6, 2024
A potent antibiotic for a drug-resistant superbug
JAN 3, 2024
How does one check life signs of an insect?
JAN 2, 2024

More from Biology and Medical

1
2

Medical Xpress
Medical research advances and health news

Tech Xplore
The latest engineering, electronics and technology advances

Science X
The most comprehensive sci-tech news coverage on the web

Newsletters
Email
Science X Daily and the Weekly Email Newsletter are free features that allow you to receive your favorite sci-tech news updates in your email inbox
Follow us
Top
Home
Search
Mobile version
Help
FAQ
About
Contact
Science X Account
Premium Account
Archive
News wire
Android app
iOS app
RSS feeds
Push notification
© Phys.org 2003 – 2024 powered by Science X Network
Privacy policy Terms of use