This paper reports the chemometric analysis of near-infrared spectra drawn from hyperspectral images to develop, evaluate, and compare statistical models for the detection of beef in fish meal. There were 40 pure-fish meal samples, 15 pure-beef meal samples, and 127 fish/beef mixture meal samples prepared for hyperspectral line-scan imaging by a machine vision system. Spectral data for 3600 pixels per sample, in which individual spectra was obtain, were retrieved from the region of interest (ROI) in every sample image. The spectral data spanning 969 nm to 1551 nm (across 176 spectral bands) were analyzed. Statistical models were built using the principal component analysis (PCA) and the partial least squares regression (PLSR) methods. The models were created and developed using the spectral data from the purefish meal and pure-beef meal samples, and were tested and evaluated using the data from the ROI in the mixture meal samples. The results showed that, with a ROI as large as 3600 pixels to cover sufficient area of a mixture meal sample, the success detection rate of beef in fish meal could be satisfactory 99.2% by PCA and 98.4% by PLSR.
This paper reports the development of a multispectral algorithm, using the line-scan hyperspectral imaging system, to detect fecal contamination on leafy greens. Fresh bovine feces were applied to the surfaces of washed loose baby spinach leaves. A hyperspectral line-scan imaging system was used to acquire hyperspectral fluorescence images of the contaminated leaves. Hyperspectral image analysis resulted in the selection of the 666 nm and 688 nm wavebands for a multispectral algorithm to rapidly detect feces on leafy greens, by use of the ratio of fluorescence intensities measured at those two wavebands (666 nm over 688 nm). The algorithm successfully distinguished most of the lowly diluted fecal spots (0.05 g feces/ml water and 0.025 g feces/ml water) and some of the highly diluted spots (0.0125 g feces/ml water and 0.00625 g feces/ml water) from the clean spinach leaves. The results showed the potential of the multispectral algorithm with line-scan imaging system for application to automated food processing lines for food safety inspection of leafy green vegetables.
Fecal contamination of produce is a food safety issue associated with pathogens such as Escherichia coli that can easily
pollute agricultural products via animal and human fecal matters. Outbreaks of foodborne illnesses associated with
consuming raw fruits and vegetables have occurred more frequently in recent years in the United States. Among fruits,
strawberry is one high-potential vector of fecal contamination and foodborne illnesses since the fruit is often consumed
raw and with minimal processing. In the present study, line-scan LED-induced fluorescence imaging techniques were
applied for inspection of fecal material on strawberries, and the spectral characteristics and specific wavebands of
strawberries were determined by detection algorithms. The results would improve the safety and quality of produce
consumed by the public.
In this research, a multispectral algorithm derived from hyperspectral line-scan fluorescence imaging under violet LED
excitation was developed for the detection of frass contamination on mature tomatoes. The algorithm utilized the
fluorescence intensities at two wavebands, 664 nm and 690 nm, for computation of the simple ratio function for effective
detection of frass contamination. The contamination spots were created on the tomato surfaces using four concentrations
of aqueous frass dilutions. The algorithms could detect more than 99% of the 0.2 g/ml and 0.1 g/ml frass contamination
spots and successfully differentiated these spots from clean tomato surfaces. The results demonstrated that the simple
multispectral fluorescence imaging algorithms based on violet LED excitation can be appropriate to detect frass on
tomatoes in high-speed post-harvest processing lines.
Korla fragrant pears are small oval pears characterized by light green skin, crisp texture, and a pleasant perfume for
which they are named. Anatomically, the calyx of a fragrant pear may be either persistent or deciduous; the deciduouscalyx
fruits are considered more desirable due to taste and texture attributes. Chinese packaging standards require that
packed cases of fragrant pears contain 5% or less of the persistent-calyx type. Near-infrared hyperspectral imaging was
investigated as a potential means for automated sorting of pears according to calyx type. Hyperspectral images spanning
the 992-1681 nm region were acquired using an EMCCD-based laboratory line-scan imaging system. Analysis of the
hyperspectral images was performed to select wavebands useful for identifying persistent-calyx fruits and for identifying
deciduous-calyx fruits. Based on the selected wavebands, an image-processing algorithm was developed that targets
automated classification of Korla fragrant pears into the two categories for packaging purposes.
The physical and mechanical properties of baby spinach were investigated, including density, Young's modulus, fracture
strength, and friction coefficient. The average apparent density of baby spinach leaves was 0.5666 g/mm3. The tensile
tests were performed using parallel, perpendicular, and diagonal directions with respect to the midrib of each leaf. The
test results showed that the mechanical properties of spinach are anisotropic. For the parallel, diagonal, and
perpendicular test directions, the average values for the Young's modulus values were found to be 2.137MPa, 1.0841
MPa, and 0.3914 MPa, respectively, and the average fracture strength values were 0.2429 MPa, 0.1396 MPa, and 0.1113
MPa, respectively. The static and kinetic friction coefficient between the baby spinach and conveyor belt were
researched, whose test results showed that the average coefficients of kinetic and maximum static friction between the
adaxial (front side) spinach leaf surface and conveyor belt were 1.2737 and 1.3635, respectively, and between the
abaxial (back side) spinach leaf surface and conveyor belt were 1.1780 and 1.2451 respectively. These works provide the
basis for future development of a whole-surface online imaging inspection system that can be used by the commercial
vegetable processing industry to reduce food safety risks.
This research developed and evaluated the multispectral algorithms derived from hyperspectral line-scan fluorescence
imaging under violet LED excitation for detection of fecal contamination on Golden Delicious apples. The algorithms
utilized the fluorescence intensities at four wavebands, 680 nm, 684 nm, 720 nm, and 780 nm, for computation of simple
functions for effective detection of contamination spots created on the apple surfaces using four concentrations of
aqueous fecal dilutions. The algorithms detected more than 99% of the fecal spots. The effective detection of feces
showed that a simple multispectral fluorescence imaging algorithm based on violet LED excitation may be appropriate
to detect fecal contamination on fast-speed apple processing lines.
This paper reported the development of hyperspectral fluorescence imaging system using ultraviolet-A excitation (320-400 nm) for detection of bovine fecal contaminants on the abaxial and adaxial surfaces of romaine lettuce and baby
spinach leaves. Six spots of fecal contamination were applied to each of 40 lettuce and 40 spinach leaves. In this study,
the wavebands at 666 nm and 680 nm were selected by the correlation analysis. The two-band ratio, 666 nm / 680 nm, of
fluorescence intensity was used to differentiate the contaminated spots from uncontaminated leaf area. The proposed
method could accurately detect all of the contaminated spots.
An online line-scan imaging system was developed for differentiation of wholesome and systemically diseased chickens. The hyperspectral imaging system used in this research can be directly converted to multispectral operation and would provide the ideal implementation of essential features for data-efficient high-speed multispectral classification algorithms. The imaging system consisted of an electron-multiplying charge-coupled-device (EMCCD) camera and an imaging spectrograph for line-scan images. The system scanned the surfaces of chicken carcasses on an eviscerating line at a poultry processing plant in December 2005. A method was created to recognize birds entering and exiting the field of view, and to locate a Region of Interest on the chicken images from which useful spectra were extracted for analysis. From analysis of the difference spectra between wholesome and systemically diseased chickens, four wavelengths of 468 nm, 501 nm, 582 nm and 629 nm were selected as key wavelengths for differentiation. The method of locating the Region of Interest will also have practical application in multispectral operation of the line-scan imaging system for online chicken inspection. This line-scan imaging system makes possible the implementation of multispectral inspection using the key wavelengths determined in this study with minimal software adaptations and without the need for cross-system calibration.
During in-plant testing of a hyperspectral line-scan imaging system, images were acquired of wholesome and
systemically diseased chickens on a commercial processing line moving at a speed 70 birds per minute. A fuzzy logic
based algorithm using four key wavelengths, 468 nm, 501 nm, 582 nm, 629 nm, was developed using image data from
the validation set of images of 543 wholesome and 66 systemically diseased chickens. A classification method using the
fuzzy logic based algorithm was then tested on the testing set of images of 457 wholesome and 37 systemically diseased
chickens, as well as 80 systemically diseased chickens that were imaged off-shift during breaks between normal
processing shifts of the chicken plant. The classification method correctly identified 89.7% of wholesome chicken
images and 98.5% of systemically diseased chicken images in the validation set. For the testing data set, the method
correctly classified 96.7 % of 457 wholesome chicken images and 100% of 37 systemically diseased chicken images.
The 80 images acquired off-shift were also 100% correctly identified.
We have developed nondestructive opto-electronic imaging techniques for rapid assessment of safety and
wholesomeness of foods. A recently developed fast hyperspectral line-scan imaging system integrated with a
commercial apple-sorting machine was evaluated for rapid detection of animal feces matter on apples. Apples
obtained from a local orchard were artificially contaminated with cow feces. For the online trial, hyperspectral
images with 60 spectral channels, reflectance in the visible to near infrared regions and fluorescence emissions with
UV-A excitation, were acquired from apples moving at a processing sorting-line speed of three apples per second.
Reflectance and fluorescence imaging required a passive light source, and each method used independent continuous
wave (CW) light sources. In this paper, integration of the hyperspectral imaging system with the commercial applesorting
machine and preliminary results for detection of fecal contamination on apples, mainly based on the
fluorescence method, are presented.
Several of visible and NIR bands were sought to explore the potential for the classification of fecal / ingesta ("F/I")
objectives from rubber belt and stainless steel ("RB/SS") backgrounds. Spectral features of "F/I" objectives and
"RB/SS" backgrounds showed large differences in both visible and NIR regions, due to the diversity of their chemical
compositions. Such spectral distinctions formed the basis on which to develop simple three-band ratio algorithms for the
classification analysis. Meanwhile, score-score plots from principal component analysis (PCA) indicated the obvious
cluster separation between "F/I" objectives and "RB/SS" backgrounds, but the corresponding loadings did not show any
specific wavelengths for developing effective algorithms. Furthermore, 2-class soft independent modeling of class
analogy (SIMCA) models were developed to compare the correct classifications with those from the ratio algorithms.
Results indicated that using ratio algorithms in the visible or NIR region could separate "F/I" objectives from "RB/SS"
backgrounds with a success rate of over 97%.
A hyperspectral line-scan imaging system for automated inspection of wholesome and diseased chickens was developed and demonstrated. The hyperspectral imaging system consisted of an electron-multiplying charge-coupled-device (EMCCD) camera and an imaging spectrograph. The system used a spectrograph to collect spectral measurements across a pixel-wide vertical linear field of view through which moving chicken carcasses passed. After a series of image calibration procedures, the hyperspectral line-scan images were collected for chickens on a laboratory simulated processing line. From spectral analysis, four key wavebands for differentiating between wholesome and systemically diseased chickens were selected: 413 nm, 472 nm, 515 nm, and 546 nm, and a reference waveband, 622 nm. The ratio of relative reflectance between each key wavelength and the reference wavelength was calculated as an image feature. A fuzzy logic-based algorithm utilizing the key wavebands was developed to identify individual pixels on the chicken surface exhibiting symptoms of systemic disease. Two differentiation methods were built to successfully differentiate 72 systemically diseased chickens from 65 wholesome chickens.
A simple multispectral classification method for the identification of systemically diseased chickens was developed and tested between two different imaging systems. An image processing algorithm was developed to define and locate the region of interest (ROI) as classification areas on the image. The average intensity was calculated for each classification area of the chicken image. A decision tree algorithm was used to determine threshold values for each classification areas. The wavelength of 540 nm was used for image differentiation purpose. There were 164 wholesome and 176 systemically diseased chicken images collected using the first imaging system, and 332 wholesome and 318 systemically diseased chicken images taken by the second imaging system. The differentiation thresholds, generated by the decision tree method, based on the images from the first imaging system were applied to the images from the second imaging system, and vice versa. The accuracy from evaluation was 95.7% for wholesome and 97.7% of systemically diseased chickens for the first image batch, and 99.7% for wholesome and 93.5% for systemically diseased chickens for the second image batch. The result showed that using single wavelength and threshold, this simple classification method can be used in automated on-line applications for chicken inspection.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.