Researchers have recently advanced precision techniques that integrate optical sensors and AI to enhance drying efficiency. A recent study from the University of Illinois Urbana-Champaign highlights three innovative smart drying methods, offering valuable insights for the food industry. The study was published in the journal Food Engineering Reviews.
Illinois researchers studied optical sensing techniques for smart food drying, testing the systems on apple slices. Image Credit: University of Illinois Urbana-Champaign
Food drying is a popular method of preserving various foods, such as meat and fruits, but it can change the food's nutritional content and quality.
With traditional drying systems, you need to remove samples to monitor the process. But with smart drying or precision drying, you can continuously monitor the process in real time, enhancing accuracy and efficiency.
Mohammed Kamruzzaman, University of Illinois Urbana-Champaign
Mohammed Kamruzzaman is an Assistant Professor 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.
The researchers examined scholarly research on various equipment types that use precision techniques to improve smart drying capabilities in the food industry.
This study explores three novel optical sensing systems – near-infrared (NIR) spectroscopy, RGB imaging with computer vision, and near-infrared hyperspectral imaging (NIR-HSI) – for precision food drying. The researchers discussed each system's mechanisms, applications, advantages, and limitations while providing an overview of common industrial drying techniques, including spray, microwave, freeze, and hot-air oven drying, that can be integrated with these precision monitoring approaches.
You can use each of the three sensors separately or in combination. What you choose will depend on the particular drying system, your needs, and cost-effectiveness.
Marcus Vinicius da Silva Ferreira, Lead Author and Postdoctoral Fellow, Department of Agricultural and Biological Engineering, University of Illinois Urbana-Champaign
RGB with computer vision utilizes a standard camera to capture visible light within the RGB color spectrum. It cannot measure moisture content but can give information about surface-level characteristics like size, shape, color, and flaws.
Using near-infrared light, NIR spectroscopy measures the absorbance of various wavelengths, which can be linked to a product's specific physical and chemical properties. It can also measure internal attributes like moisture content. However, NIR scans one point at a time.
According to Kamruzzaman, this can initially be effective for a single product, such as an apple slice.
“But as the drying progresses, the material will shrink and become heterogeneous, because of cracking and bending. If you use NIR at that stage, and if you only scan a single point, you cannot measure the drying rate,” noted Kamruzzaman.
Among the three methods, NIR-HSI is the most comprehensive approach, providing significantly more accurate information about the drying rate and other crucial parameters than NIR alone. This superior accuracy stems from its ability to extract three-dimensional spatial and spectral information, effectively scanning the entire product surface.
However, this enhanced capability comes at a considerable cost. NIR-HSI equipment is substantially more expensive than its counterparts, with costs exceeding those of RGB cameras by a factor of 100 and surpassing NIR sensors by a factor of 10 to 20 times. Furthermore, the overall cost is substantially increased by HSI systems' significantly higher maintenance and computing requirements.
All three approaches necessitate the integration of AI and machine learning algorithms for data processing, with the specific models requiring customization for each application. Due to its high data generation, HSI demands significantly greater computational resources for data processing compared to the other two systems.
To test the different approaches, the researchers also created their own drying system. The scientists constructed a convective heat oven and experimented with drying apple slices. The system was first combined with RGB and NIR. Later, the NIR-HSI system was tested, and the results will be covered in a forthcoming paper.
“For real-time monitoring, the convergence of RGB imaging, NIR spectroscopic sensors, and NIR-HSI with AI represents a transformative future for food drying. Integrating these technologies overcomes conventional drying process monitoring limitations and propels real-time monitoring capabilities,” concluded the scientists.
The scientists pointed out that the development of portable, hand-held NIR-HSI devices in the future will allow continuous monitoring of drying systems and real-time quality control in a range of operating environments.
The Center for Advanced Research in Drying (CARD), a U.S. National Science Foundation Industry University Cooperative Research Center based at Worcester Polytechnic Institute and the University of Illinois at Urbana-Champaign, provided financial support for this study.
Journal Reference:
Kamruzzaman, M., et al. (2024) AI-Enabled Optical Sensing for Smart and Precision Food Drying: Techniques, Applications and Future Directions. Food Engineering Reviews. doi.org/10.1007/s12393-024-09388-0