|Qin, Jianwei - MICHIGAN ST UNIVERSITY|
Submitted to: ASABE Annual International Meeting
Publication Type: Proceedings
Publication Acceptance Date: May 13, 2006
Publication Date: July 9, 2006
Citation: Qin, J., Lu, R. 2006. Measurement of the optical properties of apples using hyperspectral diffuse reflectance imaging. ASABE Annual International Meeting. Paper No. 063037. Interpretive Summary: Scattering and absorption are two basic phenomena as light interacts with turbid food and biological materials. Light absorption is related to the chemical properties of the material whereas light scattering is influenced by density and structural characteristics. Hence, spectral absorption and scattering properties are useful in quality evaluation of agricultural and food products. Several techniques have been developed for measuring the optical properties of turbid biological materials in the biomedical engineering field. The techniques are generally expensive, inefficient or time consuming, and inconvenient in measurement; they are not yet suitable for food and agricultural products. This paper reports on the development of a new technique using hyperspectral imaging, instrumental calibration methods and procedure, and a computational algorithm, for rapid determination of the spectral absorption and scattering properties of turbid foods in the visible and near-infrared (with wavelengths longer than the visible) spectral region. Hyperspectral imaging, which combines the main features of imaging and spectroscopy, acquires from an object both spectral and spatial information. The hyperspectral diffuse reflectance imaging system was tested with three types of simulation samples with known optical properties and was then used to measure the optical properties of 'Golden Delicious' apples. Results showed that the technique could measure the spectral scattering and absorption properties of turbid materials with the average measurement errors of 11% and 18%, respectively, which are comparable to those obtained with other techniques. The optical properties of 'Golden Delicious' apples were determined between 600 nm and 950 nm, and they were related to fruit firmness and soluble solids content, two important quality parameters. The hyperspectral diffuse reflectance imaging technique provides a new means for rapid measurement of the optical properties of fruit and liquid and solid food products, which are largely absent in literature. This can help us better understand the interaction of light with fruit tissue and design more effective optical inspection systems for measuring the chemical composition and structural/physical characteristics of fruit.
Technical Abstract: This paper reports on the development of a novel hyperspectral imaging technique for rapid determination of the absorption and scattering properties of turbid food materials over the visible and near-infrared region of 450-1,000 nm. A hyperspectral imaging system in line scanning mode was tested and validated with three types of simulation samples, and it was then used to measure the optical properties of 'Golden Delicious' apples. An instrumental calibration procedure was developed to compensate for the nonuniform instrument response of the imaging system. A nonlinear curve-fitting algorithm for a steady-state diffusion theory model was proposed to determine absorption and reduced scattering coefficients from the spatially-resolved hyperspectral reflectance profiles. The hyperspectral imaging system provided good measurement of the absorption and reduced scattering coefficients for the three types of simulation samples with the average fitting errors of 18% and 11%, respectively. The optical properties of Golden Delicious apples for the spectral range of 600-950 nm were determined from the hyperspectral reflectance scattering profiles that were corrected for the fruit size effect. Simple regression analysis showed that the absorption coefficient of 690 nm was correlated with the soluble solids content of the fruit (r=-0.77) and the reduced scattering coefficient of 780 nm was correlated with fruit firmness (r=-0.70). Compared to other techniques, the hyperspectral imaging technique is simpler, faster and easier to use, and more importantly it is capable of determining optical properties over a broad spectral range simultaneously. The technique will be especially useful for measuring the optical properties of turbid food and agricultural products.