Sukwon Kang

Enhanced Raman spectroscopic discrimination of the geographical origins of rice samples via transmission spectral collection through packed grains.

Transmission Raman spectroscopy has been effectively utilized for the discrimination of rice samples according to geographical origin. Since the constituents of rice are heterogeneously distributed and/or localized in a grain, the collection of Raman spectra providing a better compositional representation of packed rice grains is an essential requirement for accurate analysis. The optimal packing thickness yielding the most reproducible transmission spectra was initially determined. Internal propagation of radiation was more sensitively influenced by random packing when a packing was thinner; while, a thicker packing largely attenuated transmitting Raman signal and eventually degraded the signal-to-noise ratio of collected spectra. At the determined packing thickness, transmission spectra of all rice samples were collected, and discrimination into two different geographical origins was performed using principal component analysis (PCA) combined with linear discriminant analysis (LDA). For comparison, back-scattering Raman spectra of the same samples were also collected. The discrimination accuracy was improved when Raman spectra collected directly through the packed rice grains were used. Since the constituents of rice were not homogeneously distributed in a grain as confirmed using Raman microscopy, the transmission measurement enabling transversal sampling across a packing of rice grains was better for compositional representation of individual grains in the packing and able to recognize minute spectral differences between two groups, ultimately leading to more accurate discrimination of geographical origin.

Hyperspectral Imaging for Detecting Defect on Apples

The reflectance spectra from the hyperspectral images of apples were used to find the optimal wavelengths to discriminate the defect region from the normal region. The optimal wavelength was selected from the correlation analysis between the band ratio or wavelength difference and target regions. The spectral images of selected wavelength were used to validate the correlation analysis. The correlation coefficient value using the band ratio at correlation analysis was 0.91 and the using the difference was 0.79. Thus, the correlation analysis method is feasible to select the wavelength to discriminate defects from normal regions.

Multispectral fluorescence lifetime imaging of feces-contaminated apples by time-resolved laser-induced fluorescence imaging system with tunable excitation wavelengths

We recently developed a time-resolved multispectral laser-induced fluorescence (LIF) imaging system capable of tunable wavelengths in the visible region for sample excitation and nanosecond-scale characterizations of fluorescence responses (lifetime imaging). Time-dependent fluorescence decay characteristics and fluorescence lifetime imaging of apples artificially contaminated with a range of diluted cow feces were investigated at 670 and 685 nm emission bands obtained by 418, 530, and 630 nm excitations. The results demonstrated that a 670 nm emission with a 418 nm excitation provided the greatest difference in time-dependent fluorescence responses between the apples and feces-treated spots. The versatilities of the time-resolved LIF imaging system, including fluorescence lifetime imaging of a relatively large biological object in a multispectral excitation-emission wavelength domain, were demonstrated.

Non-Destructive Quality Evaluation of Pepper (Capsicum annuum L.) Seeds Using LED-Induced Hyperspectral Reflectance Imaging

In this study, we developed a viability evaluation method for pepper (Capsicum annuum L.) seeds based on hyperspectral reflectance imaging. The reflectance spectra of pepper seeds in the 400–700 nm range are collected from hyperspectral reflectance images obtained using blue, green, and red LED illumination. A partial least squares–discriminant analysis (PLS-DA) model is developed to classify viable and non-viable seeds. Four spectral ranges generated with four types of LEDs (blue, green, red, and RGB), which were pretreated using various methods, are investigated to develop the classification models. The optimal PLS-DA model based on the standard normal variate for RGB LED illumination (400–700 nm) yields discrimination accuracies of 96.7% and 99.4% for viable seeds and nonviable seeds, respectively. The use of images based on the PLS-DA model with the first-order derivative of a 31.5-nm gap for red LED illumination (600–700 nm) yields 100% discrimination accuracy for both viable and nonviable seeds. The results indicate that a hyperspectral imaging technique based on LED light can be potentially applied to high-quality pepper seed sorting.

Fluorescence Characteristics of Wholesome and Unwholesome Chicken Carcasses

Each chicken carcass intended for U.S. consumers is mandated to be inspected by Food Safety and Inspection Service (FSIS) inspectors for its wholesomeness at the processing plants. Fluorescence responses of wholesome and unwholesome chicken carcasses were characterized and further evaluated for potential on-line applications for detection and classification of wholesome and unwholesome chicken carcasses. For this study, unwholesome chicken carcasses included cadaver and those with disease conditions such as airsacculitis and septicemia. Fluorescence characteristics from the epidermal layers in the breast areas of chicken carcasses were dynamic in nature. Emission peaks and ridges (maxima) were observed at 386, 444, 472, 512, and 554 nm and valleys (minima) were observed at 410, 460, 484, and 538 nm. One of the major factors affecting the line shapes of fluorescence responses from chicken carcass skin layers was absorption by hemoglobin. With the use of the normalized ratio spectra (NRS) approach, oxyhemoglobin was shown to be a major constituent in chicken carcasses affecting the fluorescence emission line shapes. Subtle line shape changes in the NRS also provided a qualitative means by which to assess the minute differences in oxy- and deoxyhemoglobin compositions perturbed by poultry diseases such as septicemia and airsacculitis. With the use of simple fluorescence band ratios as a multivariate model, wholesome and unwholesome chicken carcasses were correctly classified with 97.1% and 94.8% accuracies, respectively. On-line implementation of fluorescence techniques for the assessment of chicken carcass wholesomeness appears promising.

Non-destructive and Rapid Prediction of Moisture Content in Red Pepper (Capsicum annuum L.) Powder Using Near-infrared Spectroscopy and a Partial Least Squares Regression Model

National Academy of Agricultural Science, Rural Development Administration, 150 Suinro, Gwonseon-gu, Suwon, Gyeonggi-do 441-100, Korea Environmental Microbial and Food Safety Laboratory, Beltsville Agricultural Research Center, Agricultural Research Service, United States Department of Agriculture, 10300 Baltimore Ave, Beltsville, MD 20705 Received: July 23rd 2014; Revised: July 29th 2014; Accepted: August 9th 2014 Purpose: The aim of this study was to develop a technique for the non-destructive and rapid prediction of the moisture content in red pepper powder using near-infrared (NIR) spectroscopy and a partial least squares regression (PLSR) model. Methods: Three red pepper powder products were separated into three groups based on their particle sizes using a standard sieve. Each product was prepared, and the expected moisture content range was divided into six or seven levels from 3 to 21% wb with 3% wb intervals. The NIR reflectance spectra acquired in the wavelength range from 1,100 to 2,300 nm were used for the development of prediction models of the moisture content in red pepper powder. Results: The values of RV2, SEP, and RPD for the best PLSR model to predict the moisture content in red pepper powders of varying particle sizes below 1.4 mm were 0.990, ±0.487% wb, and 10.00, respectively. Conclusions: These results demonstrated that NIR spectroscopy and a PLSR model could be useful techniques for measuring rapidly and non-destructively the moisture content in red pepper powder.

A Near-Infrared Sensing Technique for Measuring the Quality of Potatoes

Near–infrared (NIR) spectroscopy is a promising technique for nondestructive sensing of agricultural product for multiple quality attributes. The objective of this research was to explore a NIR sensing technique in interaction mode for rapid acquisition of spectral information to predict the quality parameters of potatoes. The percentage of dry matter and specific gravity are important quality parameters for assessing the potential of potato tubers to produce acceptable yields of processed products with good textural and color quality attributes. In this research, the NIR calibration for dry matter and specific gravity was developed with an on-line system to predict and classify the specific gravity and dry matter of potatoes. The model predicted specific gravity of intact potatoes with 87% accuracy compared with the measured data, and the Standard Error of Correlation (SEC) was ±0.0022. From the cross validation result, the correlation coefficient was 0.85 and the Standard Error of Prediction (SEP) was ±0.0024. For the dry matter of intact potatoes, the correlation coefficient of calibration was 0.85 and the Standard Error of Correlation (SEC) was ± 0.60%. The correlation coefficient of prediction was 0.82 and the Standard Error of Prediction (SEP) was ±0.66%. Thus, the developed model and the on-line VIS/NIR transmittance system were feasible to classify the potato by the specific gravity and dry matter.

Determination of Germination Quality of Cucumber (Cucumis Sativus) Seed by LED-Induced Hyperspectral Reflectance Imaging

Purpose: We developed a viability evaluation method for cucumber (Cucumis sativus) seed using hyperspectral reflectance imaging. Methods: Reflectance spectra of cucumber seeds in the 400 to 1000 nm range were collected from hyperspectral reflectance images obtained using blue, green, and red LED illumination. A partial least squares–discriminant analysis (PLS-DA) was developed to predict viable and non-viable seeds. Various ranges of spectra induced by four types of LEDs (Blue, Green, Red, and RGB) were investigated to develop the classification models. Results: PLS-DA models for spectra in the 600 to 700 nm range showed 98.5% discrimination accuracy for both viable and non-viable seeds. Using images based on the PLS-DA model, the discrimination accuracy for viable and non-viable seeds was 100% and 99%, respectively Conclusions: Hyperspectral reflectance images made using LED light can be used to select high quality cucumber seeds.

Classification of fecal contamination on leafy greens by hyperspectral imaging

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.

Evaluation of Antibody Immobilization Methods for Detection of Salmonella using Impedimetric Biosensor

Conventional methods for pathogen detection and identification are labor-intensive and take several days to complete. Recently developed biosensors have shown potential for the rapid detection of foodborne pathogens. In this study, an impedimetric biosensor was developed for rapid detection of Salmonella typhimurium. To develop the biosensor, an interdigitated microelectrode (IME) was fabricated by using semiconductor fabrication process. Anti-Salmonella antibodies were immobilized based on either avidin-biotin binding or self assembled monolayer (SAM) on the surface of the IME to form an active sensing layer. To evaluate effect of antibody immobilization methods on sensitivity of the sensor, detection limit of the biosensor was analyzed with Salmonella samples innoculated in phosphate buffered saline (PBS) or food extract. The impedimetric biosensor based on SAM immobilization method produced better detection limit. The biosensor could detect 107 CFU/mL of Salmonella in pork meat extract. This method may provide a simple, rapid, and sensitive method to detect foodborne pathogens.