Measuring Sugar Levels in Tomatoes Using Advanced Imaging Technology

Jim Crocker
15th August, 2024

Measuring Sugar Levels in Tomatoes Using Advanced Imaging Technology

Image Source: Natural Science News, 2024

Key Findings

  • The study from Ningxia University developed a non-destructive method to predict tomato flavor using hyperspectral imaging and neural networks
  • Higher nitrogen levels in the soil resulted in lower soluble solids content (SSC) in tomatoes, affecting their flavor
  • The prediction models, particularly CARS-CNN and IRIV-PCNN, showed high accuracy, making them reliable tools for monitoring tomato quality
Tomatoes are a popular fruit known for their sweet and sour flavor, which is largely influenced by their soluble solids content (SSC). SSC is a crucial indicator of tomato flavor and nutritional value. The recent study from Ningxia University[1] explored how different nitrogen treatments affect the SSC of tomatoes using hyperspectral imaging (HSI) combined with neural network prediction models. The study aimed to develop a non-destructive method for predicting the SSC of tomatoes grown under varying nitrogen conditions. Nitrogen is a vital nutrient for plant growth, but its form and concentration can significantly impact the physiological processes and flavor of tomatoes. Traditional methods of measuring SSC, such as colorimetric enzymatic measurements and high-performance liquid chromatography (HPLC), are accurate but time-consuming and destructive[2]. Therefore, a rapid and non-destructive alternative is highly desirable. In this study, the researchers employed HSI, a technique that captures and processes information across the electromagnetic spectrum. This method, combined with advanced neural network models, can predict the SSC of tomatoes without damaging them. The researchers used two specific methods for feature extraction: Competitive Adaptive Reweighed Sampling (CARS) and Iterative Retained Information Variable (IRIV). These methods help identify the most relevant wavelengths from the HSI data, which are then used to build predictive models. The study constructed and optimized custom convolutional neural networks (CNNs) to predict the SSC of tomatoes. The results showed that the SSC was negatively correlated with nitrogen fertilizer concentration. This means that higher nitrogen levels in the soil resulted in lower SSC in the tomatoes. The prediction models, particularly the CARS-CNN and IRIV-parallel convolutional neural networks (PCNN), demonstrated good accuracy, with residual predictive deviation (RPD) values of 1.64 and 1.66, respectively. An RPD value above 1.6 indicates a robust predictive model. This study builds on previous research that utilized advanced imaging and neural network techniques to assess fruit quality. For instance, earlier work on cherry tomatoes used hyperspectral imaging combined with a convolutional neural network with Transformer to predict SSC and pH[3]. The findings from both studies underscore the effectiveness of integrating spectral information with neural networks to enhance the accuracy of quality predictions in fruits. Moreover, the study's approach aligns with the broader trend of using non-destructive methods for assessing fruit quality. For example, ATR-FTIR spectroscopy has been used to predict internal quality properties like sweetness and acidity in peach fruits[2]. These methods offer rapid and reliable alternatives to traditional techniques, making them valuable tools for both researchers and producers. The implications of this study are significant for tomato cultivation and quality control. By providing a non-destructive method to predict SSC, the research offers a practical solution for monitoring and optimizing tomato flavor based on nitrogen treatments. This can help farmers adjust their fertilization practices to produce tomatoes with the desired flavor profile, ultimately benefiting consumers. In conclusion, the study from Ningxia University demonstrates the potential of using hyperspectral imaging and neural network models to predict the SSC of tomatoes under different nitrogen treatments. By leveraging advanced feature extraction methods and custom CNNs, the research provides a valuable tool for non-destructive quality assessment in tomato production. This approach not only enhances our understanding of how nitrogen affects tomato flavor but also offers practical applications for improving fruit quality in the agricultural industry.

VegetablesAgricultureBiochem

References

Main Study

1) Determination of soluble solids content in tomatoes with different nitrogen levels based on hyperspectral imaging technique.

Published 13th August, 2024

https://doi.org/10.1111/1750-3841.17264

Related Studies

2) Determination of the composition in sugars and organic acids in peach using mid infrared spectroscopy: comparison of prediction results according to data sets and different reference methods.

https://doi.org/10.1021/ac402428s

3) Hyperspectral Imaging Using a Convolutional Neural Network with Transformer for the Soluble Solid Content and pH Prediction of Cherry Tomatoes.

https://doi.org/10.3390/foods13020251


Related Articles

An unhandled error has occurred. Reload 🗙