The Modeling of Non-invasive Measurement of Fruit Sweetness using Near-Infrared (NIR) Spectroscopy Method

Abstract The sugary content in fruit requires knowledge for both the maker and the consumer. Methods of evaluating the level of sweetness in fruits as previously described, for example through chemical means, is a long and tedious process. This paper demonstrates the benefits of using NIR in determining fruit sweetness as a rapid, non-destructive, and accurate technique compared to the conventional methods. Near infrared spectroscopy (NIR spectroscopy) is a rapid non-destructive technique for measuring the physical and chemical properties of fruit by analyzing the light absorption in the proximal part of infrared spectrum (780-2500 nm). This paper aims to determine the potential of applying food quality NIR spectroscopy spot check, at a wavelength of 940 nm, in the reflectance mode for the evaluation of the sweetness of apples. When light from a 940 nm LED is directed to illuminate the surface of the fruit, some of it is absorbed while the rest is reflected. The reflected light is detected using a photodiode then converted to an electrical quantity using a conversion circuit after which it is passed to a microcontroller. The study was conducted on 30 different apple samples to develop a model, with sweetness quantified using a Brix refractometer. The obtained results were used to construct a linear regression model, showing the correlation between the measured voltage values and the Brix values. The model yielded a correlation coefficient (R²) of 0.6261, demonstrating the feasibility of using the 940 nm near-infrared wavelength to non-invasively measure fruit sweetness. These preliminary results help us understand the experimental design and develop techniques to measure fruit sweetness using light at other wavelengths in the future.