Hyperspectral Image Recovery Using a Color Camera for Detecting Colonies of Foodborne Pathogens on Agar Plate

Hyperspectral imaging often requires a special camera system to obtain spectral images. The cost for acquisition and process of hyperspectral images is usually much higher than color images. On the contrary, typical consumer-grade digital color cameras are much cheaper to obtain and process spatially high-resolution images than hyperspectral cameras. This paper is concerned with the development of a hyperspectral image recovery technique that can reconstruct hyperspectral images only from color images obtained by a digital color camera. A sparse representation and least squares regression-based classification of foodborne pathogens on agar plates are presented. The target pathogen bacteria were the six representative non-O157 Shiga toxin-producing Escherichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145). The wavelength range for the spectral recovery was from 400 to 1000 nm. Unlike many other studies using color charts with known and noise-free spectra for training their spectral recovery models, we directly used the hyperspectral and color images of real scenes for training the spectral recovery models. Both hyperspectral and color images were calibrated to percent reflectance values and then spatially registered. Two spectral recovery models including polynomial multivariate linear regression (MLR) and partial least squares regression (PLSR) were evaluated and compared by cross-validation and independent test. The spectral recovery results showed that the PLSR was more effective than the MLR. The classification accuracy measured with recovered spectra in an independent test set was about 5–10% less than the case of using the true hyperspectral images although the difference in maximum classification accuracy was only about 2%. The results suggested the potential of a cost-effective color imaging system using hyperspectral image classification algorithms for differentiating pathogens in agar plates.

[1]  Izumi Nishidate,et al.  Estimation of Melanin and Hemoglobin Using Spectral Reflectance Images Reconstructed from a Digital RGB Image by the Wiener Estimation Method , 2013, Sensors.

[2]  Kurt C. Lawrence,et al.  Automatic Counting and Classification of Bacterial Colonies Using Hyperspectral Imaging , 2015, Food and Bioprocess Technology.

[3]  Elizabeth A. Peck,et al.  Introduction to Linear Regression Analysis , 2001 .

[4]  Kurt C. Lawrence,et al.  Development of Hyperspectral ImagingTechnique for Salmonella Enteritidis and Typhimurium on Agar Plates , 2014 .

[5]  Javier Hernández-Andrés,et al.  Selecting algorithms, sensors, and linear bases for optimum spectral recovery of skylight. , 2007, Journal of the Optical Society of America. A, Optics, image science, and vision.

[6]  Hokwon A. Cho,et al.  Introduction to Regression Analysis , 2004 .

[7]  L. Maloney,et al.  Color constancy: a method for recovering surface spectral reflectance. , 1986, Journal of the Optical Society of America. A, Optics and image science.

[8]  P. Fratamico,et al.  Detection of Shiga toxin-producing Escherichia coli in ground beef using the GeneDisc real-time PCR system , 2012, Front. Cell. Inf. Microbio..

[9]  Paul M. Hubel,et al.  A Comparison of Methods of Sensor Spectral Sensitivity Estimation , 1994, Color Imaging Conference.

[10]  Amir Averbuch,et al.  Compressed sensing snapshot spectral imaging by a regular digital camera with an added optical diffuser. , 2016, Applied optics.

[11]  Pierre Vandergheynst,et al.  Hyperspectral image compressed sensing via low-rank and joint-sparse matrix recovery , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[12]  Shun-ichi Amari,et al.  The AIC Criterion and Symmetrizing the Kullback–Leibler Divergence , 2007, IEEE Transactions on Neural Networks.

[13]  L. Pardo.  Statistical Inference Based on Divergence Measures , 2005 .

[14]  Seung-Chul Yoon,et al.  Detection by hyperspectral imaging of shiga toxin-producing Escherichia coli serogroups O26, O45, O103, O111, O121, and O145 on rainbow agar. , 2013, Journal of food protection.

[15]  Nils Lid Hjort,et al.  Model Selection and Model Averaging , 2001 .

[16]  Douglas M. Hawkins,et al.  The Problem of Overfitting , 2004, J. Chem. Inf. Model..

[17]  Kurt C. Lawrence,et al.  Hyperspectral imaging using RGB color for foodborne pathogen detection , 2015, J. Electronic Imaging.

[18]  Kevin W Eliceiri,et al.  NIH Image to ImageJ: 25 years of image analysis , 2012, Nature Methods.

[19]  K. Lawrence,et al.  Hyperspectral Reflectance Imaging for Detecting a Foodborne Pathogen: Campylobacter , 2009 .

[20]  Haris Vikalo,et al.  Accelerated orthogonal least-squares for large-scale sparse reconstruction , 2018, Digit. Signal Process..

[21]  Norimichi Tsumura,et al.  Spectral color imaging system for estimating spectral reflectance of paint , 2007 .

[22]  N. M. Faber,et al.  How to avoid over-fitting in multivariate calibration--the conventional validation approach and an alternative. , 2007, Analytica chimica acta.

[23]  M. Heyndrickx,et al.  Comparison of Six Chromogenic Agar Media for the Isolation of a Broad Variety of Non-O157 Shigatoxin-Producing Escherichia coli (STEC) Serogroups , 2015, International journal of environmental research and public health.

[24]  Dong Liang,et al.  Study on the key technology of reconstruction spectral reflectance based on the algorithm of compressive sensing , 2015 .

[25]  Yuji Sakatoku,et al.  Reconstruction of Hyperspectral Image based on Regression Analysis - Optimum Regression Model and Channel Selection , 2009, IMAGAPP.

[26]  S. Sutton.  Accuracy of Plate Counts , 2011 .

[27]  Hui-Liang Shen,et al.  Estimating reflectance from multispectral camera responses based on partial least-squares regression , 2010, J. Electronic Imaging.

[28]  Xiaowei An,et al.  Structural Regression Model Based Inverse Sparse Representation for Tracking Objects , 2018, IEEE Access.

[29]  C. Ripamonti,et al.  Computational Colour Science Using MATLAB , 2004 .

[30]  Kurt C. Lawrence,et al.  Differentiation of big-six non-O157 Shiga-toxin producing Escherichia coli (STEC) on spread plates of mixed cultures using hyperspectral imaging , 2013, Journal of Food Measurement and Characterization.

[31]  Stanley Osher,et al.  L1 unmixing and its application to hyperspectral image enhancement , 2009, Defense + Commercial Sensing.

[32]  Leihong Zhang,et al.  Study on the effects of sample selection on spectral reflectance reconstruction based on the algorithm of compressive sensing , 2015 .

[33]  Richard G. Baraniuk,et al.  Sparsity and Structure in Hyperspectral Imaging : Sensing, Reconstruction, and Target Detection , 2014, IEEE Signal Processing Magazine.

[34]  E. Candès,et al.  Compressive fluorescence microscopy for biological and hyperspectral imaging , 2012, Proceedings of the National Academy of Sciences.

[35]  H Haneishi,et al.  System design for accurately estimating the spectral reflectance of art paintings. , 2000, Applied optics.

[36]  Stephen Westland,et al.  Computational Colour Science using MATLAB®: Westland/Computational Colour Science using MATLAB® , 2012 .

[37]  P. Fratamico,et al.  Isolation of Shiga toxin-producing Escherichia coli serogroups O26, O45, O103, O111, O121, and O145 from ground beef using modified rainbow agar and post-immunomagnetic separation acid treatment. , 2012, Journal of food protection.

[38]  Dakai Liang,et al.  A high resolution spectrum reconstruction algorithm using compressive sensing theory , 2015, International Conference on Optical and Photonic Engineering.

[39]  S. Wold,et al.  PLS-regression: a basic tool of chemometrics , 2001 .

[40]  S. Yoon.  Foodborne Pathogen Detection , 2015 .

[41]  Roy S. Berns,et al.  Spectral Estimation Using Trichromatic Digital Cameras , 1999 .

[42]  Kurt C. Lawrence,et al.  Detection of Campylobacter colonies using hyperspectral imaging , 2010 .

[43]  M. Hauta-Kasari,et al.  Wiener estimation method in estimating of spectral reflectance from RGB images , 2007, Pattern Recognition and Image Analysis.

[44]  Jussi Parkkinen,et al.  Temporal clustering of minced meat by RGB- and spectral imaging , 2012 .

[45]  Yasuyuki Matsushita,et al.  High-resolution hyperspectral imaging via matrix factorization , 2011, CVPR 2011.

[46]  Alberto Signoroni,et al.  Hyperspectral image analysis for rapid and accurate discrimination of bacterial infections: A benchmark study , 2017, Comput. Biol. Medicine.

[47]  Vasilis Valdramidis,et al.  Recent applications of hyperspectral imaging in microbiology. , 2015, Talanta.

[48]  Glenn Healey,et al.  Using reflectance models for color scanner calibration. , 2002, Journal of the Optical Society of America. A, Optics, image science, and vision.

[49]  Kinjiro Amano,et al.  Recovering spectral data from natural scenes with an RGB digital camera and colored filters , 2007 .

[50]  Kurt C. Lawrence,et al.  Hyperspectral Imaging for Differentiating Colonies of Non-0157 Shiga-Toxin Producing Escherichia Coli (STEC) Serogroups on Spread Plates of Pure Cultures , 2013 .