Effects of optical variables in a single integrating sphere system on estimation of scattering properties of turbid media

Authors: HU, DONG - Zhejiang A & F University; Lu, Renfu; HUANG, YUPING - Nanjing Forestry University; YING, YIBIN - Zhejiang University; FU, XIAPING - Zhejiang Sci-Tech University

Submitted to: Biosystems Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/19/2020
Publication Date: 6/1/2020
Citation: Hu, D., Lu, R., Huang, Y., Ying, Y., Fu, X. 2020. Effects of optical variables in a single integrating sphere system on estimation of scattering properties of turbid media. Biosystems Engineering. 194:82-98. https://doi.org/10.1016/j.biosystemseng.2020.03.012.
DOI: https://doi.org/10.1016/j.biosystemseng.2020.03.012

Interpretive Summary: In recent years, there is increased interest in measuring optical absorption and scattering properties as a means for assessing composition, quality and condition of food and agricultural products. However, accurate measurements of these optical parameters are challenging owing to the need of using sophisticated instrumentation and a complicated mathematical approach, and the lack of a good reference measurement technique. This research was intended to determine the effect of different optical variables used in the single integrating sphere (SIS) technique, a widely used reference method, on the measurement of optical absorption and scattering parameters. Thirty-six samples, 12 each of pure absorption, pure scattering and mixed (combined absorbing and scattering materials) samples, were created, plus two solid samples of known optical properties. Experimental measurements were taken for these samples using both the SIS technique and another standard technique (i.e., collimated transmittance) to measure the absorption and scattering parameters of the 36 samples. Results showed that the SIS technique overall performed satisfactorily in terms of measurement accuracy, repeatability and sensitivity. Among all optical variables investigated, six variables (i.e., light beam size and its incidence position, refractive indexes of sample and the sample holder or cuvette, sphere wall reflectivity, and specular reflectance) were found to have a significant effect on the measurement of the optical scattering parameter and, hence, they should be carefully considered when using the SIS technique. The research provides guidance on how to properly use the SIS technique to ensure accurate measurements of the optical scattering parameter.

Technical Abstract: This research was aimed at determining the effect of optical variables for an improved single integrating sphere system, coupled with the inverse adding-doubling method, on estimating the reduced scattering coefficient of turbid media over 500-1,000 nm. Integrating sphere and collimated transmittance measurements were performed on three groups of 36 liquid samples (i.e., 12 pure absorbing samples, 12 pure scattering samples and 12 mixed samples) covering a wide range of absorption coefficient and reduced scattering coefficient, together with two commercial solid samples. The results demonstrated that the system had good accuracy, with average relative errors for estimating the absorption coefficient (pure absorption) and reduced scattering coefficient (pure scattering) of liquid samples, being 15.0% and 4.7% over 550-900 nm, compared with the collimated transmittance method and empirical equation, respectively. It also had good reproducibility for the absorption and reduced scattering coefficients with the coefficients of variation being lower than 4%. Sensitivity analysis revealed that the detectable value of absorption was around 0.02 (1/cm). Entrance and detector port diameters, standard reflectivity, and anisotropy factor had a negligible effect (less than 2.2%). However, refractive indexes of the sample and cuvette, sphere wall reflectivity, beam diameter and its position, and specular reflectance had a relatively large effect. Hence, these variables should be carefully chosen or accurately measured before optical property measurements are carried out. Moreover, a smaller diameter light beam, with the illumination incident on the sample center perpendicularly, should be used in the measurement. The specular reflectance from the sample can be, however, ignored when estimating the reduced scattering coefficient due to the difficulty of separating it from the diffuse reflectance.