Skip to main content
ARS Home » Southeast Area » Athens, Georgia » U.S. National Poultry Research Center » Quality and Safety Assessment Research Unit » Research » Publications at this Location » Publication #315161

Title: Hyperspectral imaging using a color camera and its application for pathogen detection

item Yoon, Seung-Chul
item Shin, Tae-Sung
item Heitschmidt, Gerald - Jerry
item Lawrence, Kurt
item Park, Bosoon
item Gamble, Gary

Submitted to: Proceedings for Electronic Imaging Meeting
Publication Type: Proceedings
Publication Acceptance Date: 3/3/2015
Publication Date: 3/13/2015
Citation: Yoon, S.C., Shin, T., Heitschmidt, G.W., Lawrence, K.C., Park, B., Gamble, G.R. 2015. Hyperspectral imaging using a color camera and its application for pathogen detection. Proceedings for Electronic Imaging Meeting.

Interpretive Summary: A study was conducted in order to develop a hyperspectral image recovery (reconstruction) technique using a RGB color camera for detecting colonies of non-O157 E. coli pathogens on agar media. Hyperspectral imaging is an optical imaging technique to capture both spatial and spectral information. The purpose of the study was to evaluate whether a digital RGB color camera could be used to mathematically reconstruct hyperspectral images in the visible and near-infrared spectral range between 400 and 1,000 nm and even to predict the types of the pathogens with an image classification algorithm that requires the full spectra. Polynomial regression methods including multivariate least-squares and partial least squares were used for the recovery of reflectance spectra from RGB color images. The preliminary results from 3 replicated experiments showed that partial least-squares regression (PLSR) was more effective than the standard multivariate least-squares regression (MLR). The classification accuracy was about 90%, when measured with an independent test set of images. These results suggest that a color camera is potentially feasible as a cost-effective and rapid imaging tool to predict hyperspectral images and classify the test pathogen samples. However, there are still challenges, such as improving robustness and accuracy of spectral recovery, before this technology becomes commercially viable.

Technical Abstract: This paper reports the results of a feasibility study for the development of a hyperspectral image recovery (reconstruction) technique using a RGB color camera and regression analysis in order to detect and classify colonies of foodborne pathogens. The target bacterial pathogens were the six representative non-O157 Shiga-toxin producing Escherichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145) grown in Petri dishes of Rainbow agar. The purpose of the feasibility study was to evaluate whether a DSLR camera (Nikon D700) could be used to predict hyperspectral images in the wavelength range from 400 to 1,000 nm and even to predict the types of pathogens using a hyperspectral STEC classification algorithm that was previously developed. Unlike many other studies using color charts with known and noise-free spectra for training reconstruction models, this work used hyperspectral and color images, separately measured by a hyperspectral imaging spectrometer and the DSLR color camera. The color images were calibrated (i.e. normalized) to relative reflectance, subsampled and spatially registered to match with counterpart pixels in hyperspectral images that were also calibrated to relative reflectance. Polynomial multivariate least-squares regression (PMLR) was previously developed with simulated color images. In this study, partial least squares regression (PLSR) was also evaluated as a spectral recovery technique to minimize multicollinearity and overfitting. The two spectral recovery models (PMLR and PLSR) and their parameters were evaluated by cross-validation. The QR decomposition was used to find a numerically more stable solution of the regression equation. The preliminary results showed that PLSR was more effective especially with higher order polynomial regressions than PMLR. The best classification accuracy measured with an independent test set was about 90%. The results suggest the potential of cost-effective color imaging using hyperspectral image classification algorithms for rapidly differentiating pathogens in agar plates.