In the heart of China, a team of researchers from the School of Physics at Changchun University of Science and Technology, led by Xue Zhicheng, has developed a groundbreaking method for detecting amylose content in fresh corn ears using near-infrared spectroscopy. This innovative approach, published in the journal *智慧农业* (translated as *Smart Agriculture*), promises to revolutionize the agricultural and food processing industries by providing a rapid, non-destructive, and accurate means of quality control.
Amylose, a key component of starch, significantly influences the taste and flavor of fresh corn. Traditionally, determining amylose content has been a time-consuming and labor-intensive process, involving chemical detection methods that often destroy the samples. This posed a challenge for modern agricultural production and food processing, where rapid and non-destructive detection methods are crucial.
The research team addressed this challenge by developing a non-destructive detection model based on near-infrared spectroscopy. “We aimed to create a method that not only saves time and effort but also preserves the integrity of the samples,” explained Xue Zhicheng, the lead author of the study. The team collected diffuse reflectance spectral data from the middle area of complete corn ears using a laboratory-built near-infrared spectroscopic detection system. They then established a standard database by determining the physical and chemical values of amylose content in the samples.
To optimize the model’s performance, the researchers compared five mainstream spectral pretreatment methods: standard normal variable (SNV) transform, multiplicative scatter correction (MSC), SavitZky-Golay smoothing (SGS), first-order derivative (FD), and detrending (DT). They found that the “SNV-CARS-PLSR” model, which integrated SNV preprocessing with competitive adaptive reweighted sampling (CARS) feature extraction, exhibited superior performance. This model achieved a calibration coefficient of determination (R2C) of 0.826 and a prediction coefficient of determination (R2P) of 0.820, demonstrating a 14.0% improvement in R2P compared to the full-band PLSR model with SNV preprocessing alone.
The CARS algorithm played a pivotal role in this achievement by identifying key feature wavelengths strongly correlated with amylose content. “Through its adaptive weighting and iterative optimization process, CARS successfully extracted 22 characteristic wavelengths from the original 157 wavelength points in the full spectrum,” said Zhang Yongli, a co-author of the study. This selective extraction process effectively eliminated redundant spectral information and noise interference, significantly improving the model’s predictive accuracy.
The implications of this research are far-reaching. By providing a rapid and non-destructive method for detecting amylose content, the study offers valuable technical support for quality control in the agricultural and food processing industries. This innovation could lead to more efficient production processes, reduced waste, and improved product quality, ultimately benefiting consumers and the industry as a whole.
As the world continues to grapple with the challenges of food security and sustainability, advancements in agricultural technology are more critical than ever. The research conducted by Xue Zhicheng and his team represents a significant step forward in this field, offering a glimpse into the future of smart agriculture. With further development and application, this technology could reshape the way we approach food production and quality control, paving the way for a more sustainable and efficient agricultural sector.