Experimental Study of Light Propagation in Apple Tissues Using a Multispectral Imaging System

Askoura, Mohamed Lamine; Vaudelle, Fabrice; L’Huillier, Jean-Pierre

doi:10.3390/photonics3030050

Cited by 15 publications

(10 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As with TRS, the diffusion theory has been widely used to estimate the BOP from SRS data. Although the assumption that light propagation in biological tissues is dominated by scattering is often violated, many researchers have used this approach to determine the BOP of agrofood products, such as cucumbers (Lu et al, 2011), apples (Askoura et al, 2016), citrus fruit (Lorente et al, 2015), peaches, pears, plums and tomatoes Lu, 2008, Mollazade et al, 2012). Besides this approach, an inverse Monte Carlo technique has been used because of its higher accuracy in describing reflectance at short illumination-detector distances (Baranyai andZude, 2009, Mollazade et al, 2012).…”

Section: Introductionmentioning

confidence: 99%

Time- and spatially-resolved spectroscopy to determine the bulk optical properties of ‘Braeburn’ apples after ripening in shelf life

Vanoli

Beers

Sadar

et al. 2020

Postharvest Biology and Technology

View full text Add to dashboard Cite

Bulk optical properties, in terms of absorption (µ a ) and reduced scattering coefficients (µ′ s ), can be used for the non-destructive monitoring of fruit quality during ripening. In this study, the performance of time-resolved (TRS) and spatially-resolved (SRS) spectroscopy were compared by analyzing 'Braeburn' apples over a 21 d period of ripening. Nine batches of 20 apples each were measured on the blush side by TRS (540-1064 nm) and SRS (550-1000 nm). Every fruit was analyzed for skin color, texture characteristics, relative internal space volume (RISV), total solid soluble and titratable acidity contents. TRS absorption spectra showed two maxima, the highest at 980 nm (water) and the second at 670 nm (chlorophyll), while in SRS spectra the main peak was measured at 550 nm (anthocyanins) followed by that at 670 nm. The values of µ a 580 SRS and of µ a 670 SRS were much higher than those measured at the same wavelengths by TRS suggesting that TRS and SRS actually explore the apple tissue (skin and/or flesh) in a different way. The values of µ a 980 TRS were higher than those of µ a 980 SRS , probably due to the fact that water content was lower in the skin (mostly probed by SRS) than in the flesh (mostly probed by TRS). No significant correlations were found between µ a 580 SRS and µ a 580 TRS and between µ a 980 SRS and µ a 980 TRS but a low positive relationship was observed between µ a 670 TRS and µ a 670 SRS . On the contrary, high correlations were found between µ a 670 SRS and the spectral index I AD (index of absorbance difference) related to chlorophyll in the skin and between µ a 580 SRS and the spectral index ARI (anthocyanin reflectance index), related to anthocyanin content in the peel, suggesting that µ a 580 SRS is linked to the development of the red color in the peel. Both µ a 670 TRS and µ a 670 SRS decreased during fruit ripening, indicating a decline in chlorophyll in the flesh and skin, respectively. During the shelf life period, apples became soft and mealy, as mechanical and acoustic parameters decreased and RISV increased. Fruit softening was accompanied by increasing values of both µ′ sTRS and µ′ sSRS . The µ′ sTRS and µ′ sSRS were positively related to each other, were positively correlated to RISV and negatively related to mechanical and acoustic parameters. Both the TRS and SRS technique were able to follow ripening processes in 'Braeburn' apples during the shelf life period, as absorption phenomena were related to changes in pigments present in the fruit flesh and skin, while scattering events mirrored changes in the flesh texture.

show abstract

Section: Introductionmentioning

confidence: 99%

Time- and spatially-resolved spectroscopy to determine the bulk optical properties of ‘Braeburn’ apples after ripening in shelf life

Vanoli

Beers

Sadar

et al. 2020

Postharvest Biology and Technology

View full text Add to dashboard Cite

show abstract

“…[43], the modified Adding-Doubling method can be applied to estimate optical coefficients of skin and flesh tissues. A thin sample (∼100 μm) extracted from a peel of a Granny Smith cultivar can be seen as a homogeneous layer [45] from which a scattering phase function can be evaluated. With a scattering coefficient of 40 mm −1 (average value of the apple skins in Ref.…”

Section: Application On Apple Tissuesmentioning

confidence: 99%

Light scattering in thin turbid tissue including macroscopic porosities: A study based on a Monte Carlo model

Vaudelle

2018

Optics Communications

Self Cite

View full text Add to dashboard Cite

is an open access repository that collects the work of Arts et Métiers ParisTech researchers and makes it freely available over the web where possible. This is an author-deposited version published in: https://sam.ensam.eu Handle ID: .http://hdl.handle.net/10985/17263 To cite this version :Fabrice VAUDELLE -Light scattering in thin turbid tissue including macroscopic porosities: A study based on a Monte Carlo model A Monte Carlo code is built taking into account macroscopic spheroid cavities inside a turbid medium, i.e. in mixing Multi-Layer Monte Carlo (MLMC) and Monte Carlo Ray Tracing (MCRT). That simulates a tissue with a strong and heterogeneous porosity, such as flesh tissues of fruit or bone tissues. This kind of tissue, which has two scales of porosity (microscopic and macroscopic), differs notably of the homogeneous and continuous model used in the usual radiative transfer equation. The influence of the presence of spheroids can be observed on the shape of the effective phase function, on the effect related to the time-resolved diffusion solution or also on the scattering coefficient retrieved by means of the Beer-Lambert relationship. For instance, the reduced scattering coefficients retrieved thanks to time-resolved transmittance from MLMC-MCRT models having a lot of intertwined large cavities show variations coherent with those retrieved from bone tissue. Furthermore, the effect of porosity on optical transmission seems to have a real impact when relative refractive index is close to 1.In this case, the equivalence problem between such porous MLMC-MCRT model and a homogeneous turbid medium, can be discussed at the level of the angular intensity distribution over the plane boundaries. This requires to fit this angular distribution by an Adding-Doubling model using optimized optical depth and scattering phase function. Experimental scattering phase functions obtained from apple tissues are considered in order to test this idea, and then compared with those computed with a MLMC-MCRT model.

show abstract

“…5(b) and (c)) was captured with a CCD (Charge-Coupled-Device) camera. The setup [62] is presented in Fig. 5(a).…”

Section: Transmittance-reflectance Modementioning

confidence: 99%

“…Subsequently, the Henyey-Greenstein function will be only considered for the flesh layer (g=g HG =0.8 or 0.74 for 852 and 784 nm, respectively). Second, the profiles extracted from 21 images, related to a wavelength of 784 nm, are used to build an average profile [62]. The uniform radial distribution for the incident photons was used and the density p MHG (θ) for the skin layer were defined with the previous [64] or new retrieved parameters.…”

Section: Experimental Reflectance Profiles Versus Simulationsmentioning

confidence: 99%

Light source distribution and scattering phase function influence light transport in diffuse multi-layered media

Vaudelle

L’Huillier

Askoura

2017

Optics Communications

Self Cite

View full text Add to dashboard Cite

Red and near-Infrared light is often used as a useful diagnostic and imaging probe for highly scattering media such as biological tissues, fruits and vegetables. Part of diffusively reflected light gives interesting information related to the tissue subsurface, whereas light recorded at further distances may probe deeper into the interrogated turbid tissues. However, modelling diffusive events occurring at short source-detector distances requires to consider both the distribution of the light sources and the scattering phase functions. In this report, a modified Monte Carlo model is used to compute light transport in curved and multi-layered tissue samples which are covered with a thin and highly diffusing tissue layer. Different light source distributions (ballistic, diffuse or Lambertian) are tested with specific scattering phase functions (modified or not modified Henyey-Greenstein, Gegenbauer and Mie) to compute the amount of backscattered and transmitted light in apple and human skin structures. Comparisons between simulation results and experiments carried out with a multi-spectral imaging setup confirm the soundness of the theoretical strategy and may explain the role of the skin on light transport in whole and half-cut apples. Other computational results show that a Lambertian source distribution combined with a Henyey-Greenstein phase function provides a higher photon density in the stratum corneum than in the upper dermis layer. Furthermore, it is also shown that the scattering phase function may affect the shape and the magnitude of the Bidirectional Reflectance Distribution (BRDF) exhibited at the skin surface.

show abstract

Experimental Study of Light Propagation in Apple Tissues Using a Multispectral Imaging System

Cited by 15 publications

References 49 publications

Time- and spatially-resolved spectroscopy to determine the bulk optical properties of ‘Braeburn’ apples after ripening in shelf life

Time- and spatially-resolved spectroscopy to determine the bulk optical properties of ‘Braeburn’ apples after ripening in shelf life

Light scattering in thin turbid tissue including macroscopic porosities: A study based on a Monte Carlo model

Light source distribution and scattering phase function influence light transport in diffuse multi-layered media

Contact Info

Product

Resources

About