Title: Hyperspectral image reconstruction using RGB color for foodborne pathogen detection on agar plates Authors
Submitted to: Proceedings of SPIE
Publication Type: Proceedings
Publication Acceptance Date: January 13, 2014
Publication Date: N/A
Interpretive Summary: A feasibility study was conducted to investigate the potential of using color imaging with a hyperspectral image classification model developed by hyperspectral imaging for detecting pathogens on agar plates. A reason to study a color imaging technique that can reconstruct hyperspectral images with hundreds of wavelengths was the benefit of a low cost of digital color cameras compared to the high cost of hyperspectral imaging equipment. Hence, the objective was to investigate statistical regression methods for recovering hyperspectral images only from RGB color images and to compare the performance of the hyperspectral image classification model using reconstructed hyperspectral images with the performance achieved with the original hyperspectral images. Reflectance spectra reconstructed in the visible spectral range from 400 to 700 nm were better fit to the original spectra than the ones in the longer wavelengths from 700 to 1,000 nm. The R-squared value, a statistical measure for goodness-of-fit of a regression model, was 0.98 (1 means a perfect fit.) in the visible spectral region. The overall classification accuracy of the hyperspectral image classification model in identifying the types of the big six non-O157 STEC serogroups was up to 92% (99% with the original hyperspectral images). Thus, the results of the study suggested the potential of color imaging in reconstructing hyperspectral images and applying them for hyperspectral imaging applications such as pathogen detection and classification.
Technical Abstract: This paper reports the latest development of a color vision technique for detecting colonies of foodborne pathogens grown on agar plates with a hyperspectral image classification model that was developed using full hyperspectral data. The hyperspectral classification model depended on reflectance spectra measured in the visible and near-infrared spectral range from 400 and 1,000 nm (473 narrow spectral bands). Multivariate regression methods were used to estimate and predict hyperspectral data from RGB color values. The six representative non-O157 Shiga-toxin producing Eschetichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145) were grown on Rainbow agar plates. A line-scan pushbroom hyperspectral image sensor was used to scan 36 agar plates grown with pure STEC colonies at each plate. The 36 hyperspectral images of the agar plates were divided in half to create training and test sets. The mean R-squared value for hyperspectral image estimation was about 0.98 in the spectral range between 400 and 700 nm for linear, quadratic and cubic polynomial regression models and the detection accuracy of the hyperspectral image classification model with the principal component analysis and k-nearest neighbors for the test set was up to 92% (99% with the original hyperspectral images). Thus, the results of the study suggested that color-based detection may be viable as a multispectral imaging solution without much loss of prediction accuracy compared to hyperspectral imaging.