Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Food Quality Laboratory » Research » Publications at this Location » Publication #378411

Research Project: New Sensors and Methods for Phenotypic Analysis of Small Grains

Location: Food Quality Laboratory

Title: Effect of curvature on hyperspectral reflectance images of cereal seed-sized objects

item Delwiche, Stephen - Steve
item Kim, Moon

Submitted to: Biosystems Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/17/2020
Publication Date: 12/15/2020
Citation: Delwiche, S.R., Baek, I., Kim, M.S. 2021. Effect of curvature on hyperspectral reflectance images of cereal seed-sized objects. Biosystems Engineering. 202: 55-65.

Interpretive Summary: Hyperspectral imaging (HSI), a technology that is becoming increasingly popular for inspection of agricultural commodities, combines the features of digital imaging and ultraviolet/visible/near-infrared spectroscopy. This means that physical features such as size and shape, as well as chemical features derived from spectroscopy are available for determining aspects of food quality and safety of seeds, produce, and meat. A longstanding challenge to HSI is on how the typically rounded features of agricultural products are addressed when attempting to study local features within each item. This need is especially acute when the items are the size of cereal seeds such as wheat, in which the seed's curvature is large with respect to pixel size. The current study addressed the development of a correction function for the intensity of captured radiation (light) from a rounded surface, based on assumptions of simple geometrical shapes (cylinders and prolate spheroids, i.e., football-shaped objects) and well-known theories of the angular dependency of light reflected from a matte surface. It was demonstrated that the mathematical correction functions work well for the central regions of rounded surfaces, specifically, those closest to the camera lens, and diminish in accuracy for pixels nearest to what are perceived as the edges in the camera's two-dimensional representation of curved surfaces. This work will directly benefit the developers of hyperspectral image and digital image software of agricultural products, and eventually benefit the cereals industry through the application of improved inspection systems for quality and safety.

Technical Abstract: Hyperspectral imaging for quality and safety inspection of agricultural products is frequently confronted with the challenge of handling rounded surfaces. This challenge is especially noted when the objects as small as cereal grains for which curvature with respect to the spatial dimension of a pixel is large. Diffusely reflected light from regions near the edges of a uniformly illuminated object the size of a wheat kernel will have its intensity reduced greatly, historically modeled by the Lambert Cosine Law, eponymously named after Johann Lambert of the sixteenth century, who theorized that the fraction of energy reflected from a spot on a matte surface is related to the declination angle of the observer. The current study was performed to compare the predicted Lambertian response of reflected light to actual measurement on curved surfaces of mathematically definable shapes, namely cylinders and prolate ellipsoids, with the latter introduced as an approximation to a wheat kernel. Carbon black-doped and sintered PTFE (0.16-0.99 nominal reflectance) cylinders of three diameters, with the smallest on par with the minor dimension of a wheat kernel, were scanned using a benchtop hyperspectral imaging system, as were dull gray painted prolate spheroids of equivalent size and white wheat kernels. The analysis consisted of comparing the measured reflectances from individual pixels along the curved surface with those determined according to the Lambert Cosine Law. The findings indicate that central pixels responded well to correction, with a greater departure from theory for pixels closest to the edges.