Hyperspectral imaging could predict incubator chick mortality | WATTPoultry.com

New approach improves monitoring of hatchery environments.

A new technique that simplifies and reduces the cost of hyperspectral imaging could advance egg quality analysis and improve the accuracy of chick mortality predictions in the incubator and hatchery.

“Incubation is a time-consuming and energy-intensive process and a large number of eggs do not hatch due to embryonic mortality,” said Md Toukir Ahmed, a doctoral student in the Department of Agricultural and Biological Engineering (ABE), part of the College of Agricultural, Consumer and Environmental Sciences and The Grainger College of Engineering at Illinois.

“Once optimized adequately with a machine learning interface, the technology can be easily integrated into modern hatcheries with minimum modifications to the existing industrial setup without compromising current hygiene and biosafety standards for a high throughput screening of up to 20,000 eggs per hour.”

Hyperspectral imaging is a form of spectroscopy that collects and processes information about different wavelengths in the light spectrum. This information reveals the chemical composition of an object, such as an egg.

“The technology offers extraordinary data for identifying, classifying, detecting and mapping the tested objects in more detail than any other optical sensing technology,” Ahmed said. “However, the high cost and operational complexity compared to conventional RGB imaging are significant bottlenecks in the widespread adoption.”

The new technology, developed by Ahmed, reconstructs hyperspectral images from standard RGB images using deep machine learning. One of the benefits of this approach is that all it requires is a low-cost handheld device to predict and assess products.

Smart sensing, which includes the new hyperspectral imaging method could be beneficial in the hatchery, according to a paper published in Smart Agricultural Technology,

For broilers, the technology could improve monitoring of the incubation and hatchery environments and allow for data driven decisions when it comes to adjusting incubation conditions and identifying the signs of chick mortality in the egg before hatch.