How AI and 3D Printing Are Revolutionizing Food Production
Scientists utilize laser scanning to generate 3D models of the aerial parts of sugar beet plants in agricultural fields, pushing forward the advancement of AI-supported improvements in crop pipelines.
Recent research illustrates how 21st-century crop breeding can benefit from emerging technologies by integrating laser scanning and 3D printing to produce detailed 3D models of sugar beet plants.
This approach advances beyond using genetic data alone for informed breeding, capturing the critical above-ground characteristics of sugar beet plants for AI-assisted crop improvement processes.
Open Access and Practical Application in the Field
These reproducible models are practical for field applications, and all research data, methods, and 3D printing files are freely accessible. This advancement provides valuable tools for crop management and allows anyone to print their own 3D sugar beet plant with minimal upkeep.
Modern plant breeding now heavily relies on data, utilizing machine learning algorithms and advanced imaging technology to select desirable traits. “Plant phenotyping“—the science of accurately gathering information and measurements on plants—has significantly advanced in recent years.
Previously, phenotyping depended on labor-intensive human measurements. Today, automated phenotyping pipelines are increasingly utilizing cutting-edge sensor technology, often leveraging artificial intelligence.
Enhanced Measurement Precision and Automation
These measurements can include parameters such as size, fruit quality, leaf shape, and other growth characteristics. Besides improving efficiency by automating the measurement process, computer-assisted sensors can also capture complex plant data that would be challenging for humans to collect on a large scale.
A critical factor in this sensor-driven era of crop breeding is the availability of accurate reference materials.
Sensors require data on a “standard plant” that includes all pertinent characteristics, including complex, three-dimensional traits like leaf orientation angles. Having a tangible “artificial plant” as a full-size reference is more advantageous than relying solely on computer data or flat, 2D representations.
A physical model can also serve as a reference and internal control within a greenhouse or test field among real plants.
Accessibility and Standardization through 3D Printing
The newly created 3D-printed model of a sugar beet plant was designed with these applications in mind. Its printing files are freely available for download and reuse, enabling scientists (and any sugar beet enthusiast) to produce an exact replica of the reference model.
This accessibility helps standardize research across various labs worldwide, enhancing comparability. Additionally, the affordability of 3D printing makes this method adaptable even in resource-limited environments, such as developing countries.
To obtain accurate data for their realistic model, Jonas Bömer and his team from the Institute of Sugar Beet Research (Göttingen) and the University of Bonn used LIDAR (Light Detection and Ranging) technology.
Creating and Evaluating a High-Fidelity Sugar Beet Model
They scanned an actual sugar beet plant with a laser from 12 different angles to generate 3D data. After processing this data, they input it into a commercial-grade 3D printer to produce a full-size model of the sugar beet. The team then evaluated the model’s effectiveness as a reference point in both laboratory and field settings.
Jonas Bömer elaborates: In the realm of three-dimensional plant phenotyping, accurately referencing sensor systems, computer algorithms, and measured morphological parameters is a challenging but crucial task.
Utilizing additive manufacturing technologies to create reproducible reference models offers a new way to develop standardized methodologies for precise and objective referencing, benefiting scientific research and practical plant breeding alike.
AI, 3D Printing, and Sensor Integration
However, this method is not limited to sugar beets. Moreover, the new study in GigaScience illustrates how combining artificial intelligence, 3D printing, and sensor technology can shape the future of plant breeding, supporting the production of healthy, flavorful crops to feed the global population.
Chris Armit, a data scientist at GigaScience, remarks: “The benefit of a printable 3D model is the ability to produce multiple copies, one for each crop field. As a cost-effective phenotyping strategy, where the main expense is the LIDAR scanner, it would be great to see this approach applied to other crops, such as rice or African orphan crops, which require affordable phenotyping solutions.”
Read the original article on: Scitech Daily
Read more: Scientists Boost Crop Yields with CO2-Capturing Rock Dust on Fields
