Kuanglin Chao

Machine vision technology for agricultural applications

Current applications of machine vision in agriculture are briefly reviewed. The requirements and recent developments of hardware and software for machine vision systems are discussed, with emphases on multispectral and hyperspectral imaging for modern food inspection. Examples of applications for detection of disease, defects, and contamination on poultry carcasses and apples are also given. Future trends of machine vision technology applications are discussed.

MULTISPECTRAL DETECTION OF FECAL CONTAMINATION ON APPLES BASED ON HYPERSPECTRAL IMAGERY: PART I. APPLICATION OF VISIBLE AND NEAR–INFRARED REFLECTANCE IMAGING

Fecal contamination of apples is an important food safety issue. To develop automated methods to detect such contamination, a recently developed hyperspectral imaging system with a range of 450 to 851 nm was used to examine reflectance images of experimentally contaminated apples. Fresh feces from dairy cows were applied simultaneously as a thick patch and as a thin, transparent (not readily visible to the human eye), smear to four cultivars of apples (Red Delicious, Gala, Fuji, and Golden Delicious). To address differences in coloration due to environmental growth conditions, apples were selected to represent the range of green to red colorations. Hyperspectral images of the apples and fecal contamination sites were evaluated using principal component analysis with the goal of identifying two to four wavelengths that could potentially be used in an on-line multispectral imaging system. Results indicate that contamination could be identified using either three wavelengths in the green, red, and NIR regions, or using two wavelengths at the extremes of the NIR region under investigation. The three wavelengths in the visible and near-infrared regions offer the advantage that the acquired images could also be used commercially for color sorting. However, detection using the two NIR wavelengths was found to be less sensitive to variations in apple coloration. For both sets of wavelengths, thick contamination could easily be detected using a simple threshold unique to each cultivar. In contrast, results suggest that more computationally complex analyses, such as combining threshold detection with morphological filtering, would be necessary to detect thin contamination spots using reflectance imaging techniques.

Simultaneous detection of multiple adulterants in dry milk using macro-scale Raman chemical imaging.

The potential of Raman chemical imaging for simultaneously detecting multiple adulterants in milk powder was investigated. Potential chemical adulterants, including ammonium sulphate, dicyandiamide, melamine, and urea, were mixed together into skim dry milk in the concentration range of 0.1-5.0% for each adulterant. Using a 785-nm laser, a Raman imaging system acquired hyperspectral images in the wavenumber range of 102-2538 cm(-1) for a 25 × 25 mm(2) area of each mixture sample, with a spatial resolution of 0.25 mm. Self-modelling mixture analysis (SMA) was used to extract pure component spectra, by which the four types of the adulterants were identified at all concentration levels based on their spectral information divergence values to the reference spectra. Raman chemical images were created using the contribution images from SMA, and their use to effectively visualise identification and spatial distribution of the multiple adulterant particles in the dry milk was demonstrated.

MULTISPECTRAL DETECTION OF FECAL CONTAMINATION ON APPLES BASED ON HYPERSPECTRAL IMAGERY: PART II. APPLICATION OF HYPERSPECTRAL FLUORESCENCE IMAGING

Pathogenic E. coli contamination in unpasteurized apple juice or cider is thought to originate from animal feces, and fecal contamination of apples has been recognized by the FDA as an important health issue. In a companion article, reflectance imaging techniques were shown inadequate for the detection of thin smears of feces applied to apples. The objective of this study was to evaluate the use of fluorescence imaging techniques to detect fecal contamination on apple surfaces. A hyperspectral imaging system based on a spectrograph, camera, and UV light source was used to obtain hyperspectral images of Red Delicious, Fuji, Golden Delicious, and Gala apples. Fresh dairy feces were applied to each apple as both a thick patch and as a thin smear. Results indicate that multispectral fluorescence techniques can be used to effectively detect fecal contamination on apple surfaces. Both principal component analysis and examination of emission maxima identified the same four multispectral bands (450, 530, 685, and 735 nm) as being the optimal bands to allow discrimination of contaminated apple surfaces. Furthermore, the simple two-band ratio (e.g., 685 to 450 nm) reduced the variation in normal apple surfaces while accentuating differences between contaminated and uncontaminated areas. Because of the limited sample size, delineation of an optimal detection scheme is beyond the scope of the current study. However, the results suggest that use of multispectral fluorescence techniques for detection of fecal contamination on apples in a commercial setting may be feasible.

Raman Chemical Imaging System for Food Safety and Quality Inspection

Raman chemical imaging combines Raman spectroscopy and digital imaging to visualize the composition and structure of a target, and it offers great potential for food safety and quality research. In this study, a laboratory-based Raman chemical imaging platform was designed and developed. The system utilizes a 785 nm spectrum-stabilized laser as an excitation source to generate Raman scattering. The detection module consists of a fiber optic probe, a dispersive Raman imaging spectrometer, and a high-performance spectroscopic CCD camera. The imaging system works in a point-scanning mode. A Raman spectrum is obtained at a time for individual positions in the scene. The specimens are carried by a two-axis motorized positioning table. Hyperspectral image data are accumulated as the samples are moved along two spatial dimensions. The parameterization and data-transfer interface software was developed using LabVIEW. Spectral and spatial calibration procedures are presented. The system covers a Raman shift range of 102.2 to 2538.1 cm-1 with a spectral resolution of 3.7 cm-1, and an area of 127 × 127 mm2 with a spatial resolution as high as 0.1 mm. Performance of the system was demonstrated by an example application on detection of melamine in dry milk. Melamine was mixed into dry milk with concentrations (w/w) ranging from 0.2% to 10.0%. The system was able to create Raman chemical images that can be used to visualize quantity and spatial distribution of melamine particles in the mixtures. The developed system is versatile and can be used for safety and quality inspection of food and agricultural products.

Hyperspectral-Multispectral Line-Scan Imaging System for Automated Poultry Carcass Inspection Applications for Food Safety

A hyperspectral-multispectral line-scan imaging system was developed for differentiation of wholesome and systemically diseased chickens. In-plant testing was conducted for chickens on a commercial evisceration line moving at a speed of 70 birds per minute. Hyperspectral image data was acquired for a calibration data set of 543 wholesome and 64 systemically diseased birds and for a testing data set of 381 wholesome and 100 systemically diseased birds. The calibration data set was used to develop the parameters of the imaging system for conducting multispectral inspection based on fuzzy logic detection algorithms using selected key wavelengths. Using a threshold of 0.4 for fuzzy output decision values, multispectral classification was able to achieve 90.6% accuracy for wholesome birds and 93.8% accuracy for systemically diseased birds in the calibration data set and 97.6% accuracy for wholesome birds and 96.0% accuracy for systemically diseased birds in the testing data set. By adjusting the classification threshold, 100% accuracy was achieved for systemically diseased birds with a decrease in accuracy for wholesome birds to 88.7%. This adjustment shows that the system can be feasibly adapted as needed for implementation for specific purposes, such as paw harvesting operations or prescreening for food safety inspection. This line-scan imaging system is ideal for directly implementing multispectral classification methods developed from hyperspectral image analysis.

On-line inspection of poultry carcasses by a dual-camera system

Abstract The Instrumentation and Sensing Laboratory (ISL) has developed a multi-spectral imaging system for on-line inspection of poultry carcasses. The ISL design is based on two principles: (1) wholesome and unwholesome birds have different chemical compositions of tissues and may have different skin color, and (2) unwholesome carcasses may have physical abnormalities which can be detected by computerized imaging. On-line trials of the multi-spectral chicken carcass inspection system were conducted during a 14-day period in a poultry-processing plant in New Holland, Pennsylvania, where spectral images of 13,132 wholesome and 1459 unwholesome chicken carcasses were measured. For off-line model development, the accuracies for classification of wholesome and unwholesome carcasses were 95% and 88%. On-line testing of the neural network classification models with combination of the filter information was performed. The inspection system gave accuracies of 94% and 87% for wholesome and unwholesome carcasses, respectively. This accuracy was consistent with the results obtained previously on laboratory studies. Thus, the inspection system shows promise for separation of unwholesome chicken carcasses from wholesome carcasses in poultry processing lines.

Potential of surface-enhanced Raman spectroscopy for the rapid identification of Escherichia coli and Listeria monocytogenes cultures on silver colloidal nanoparticles.

Surface-enhanced Raman (SERS) spectra of various batches of bacteria adsorbed on silver colloidal nanoparticles were collected to explore the potential of the SERS technique for rapid and routine identification of E. coli and L. monocytogenes cultures. Relative standard deviation (RSD) of SERS spectra from silver colloidal suspensions and ratios of SERS peaks from small molecules (K3PO4) were used to evaluate the reproducibility, stability, and binding effectiveness of citrate-reduced silver colloids over batch and storage processes. The results suggested consistent reproducibility of silver colloids over batch process and also stability and consistent binding effectiveness over an eight-week storage period. A variety of mixtures of E. coli/L. monocytogenes cultures with different colloidal batches revealed that, despite large variations in relative intensities and positions of SERS active bands, characteristic and unique bands at 712 and 390 cm−1 were consistently observed and were the strongest in E. coli and L. monocytogenes cultures, respectively. Two specific bands were used to develop simple algorithms in the evaluation of binding effectiveness of silver colloids over storage and further to identify E. coli and L. monocytogenes cultures with a 100% success. A single spectrum acquisition took 5∼6 min, and a minimum of 25 μL silver colloid was directly mixed with 25 μL volume of incubated bacterial culture. The short acquisition time and small volume of incubated bacterial culture make silver colloidal nanoparticle based SERS spectroscopy ideal for potential use in the routine and rapid screening of E. coli and L. monocytogenes cultures on large scales. This is the first report of the development of simple and universal algorithms for bacterial identification from the respective exclusive SERS peaks.

Potential of Raman Spectroscopy and Imaging Methods for Rapid and Routine Screening of the Presence of Melamine in Animal Feed and Foods

The impact of melamine-contaminated animal feed ingredients on food safety has become a major public concern ever since melamine was identified as the organic compound responsible for the deaths of a significant number of cats and dogs in 2007 by way of adulterated pet food. Melamine, a common industrial chemical often added to resins to improve flame resistance and proposed as an alternative form of fertilizer-N for plant growth, was found to be intentionally added to animal feed in amounts ranging from 0.2% to 8% of total mass as a way to boost the products’ apparent protein content. It was also used as a binder in the production of pelleted feed for animals. In addition to melamine, a small amount of cyanuric acid, ammeline, and ammelide were also detected in pet feed and in the tissues and urine of dead pets that had consumed the contaminated food. Even though it is possible that cyanuric acid, ammeline, and ammelide were added, their presence more likely resulted from the degraded derivatives of melamine. There is a great concern that melamine will again enter the food chain and be consumed by humans and animals. As part of the Food Protection Plan, US federal agencies such as the FDA and FSIS and other organizations have established GCMS and LC-MS/MS procedures for the analysis of melamine in food/feed commodities. Although they can detect melamine contaminants in trace amount, these time-consuming laboratory procedures require chemical solvents for the extraction steps and depend on expensive mass spectrometry. Rapid, nondestructive, and routine methods for the specific detection of melamine in raw feed materials are increasingly important, not only for public health concerns but also for melamine screening to prevent protein fraud. Undoubtedly, the well-defined mass spectroscopic technique is preferred due to its low detection limit and capability for structural elucidation; however, since adulteration of raw materials by melamine usually occurs in higher concentrations in order to affect protein content, the high sensitivity of the mass spectroscopic technique may not be necessary. In addition, mass spectrometry might not be sufficiently rapid to screen for the presence of melamine in a large number of food/feed materials from very different sources, because the identification process includes sample-specific extraction procedures, which are labor-intensive and time-consuming. Fast melamine screening requires minimal sample preparation (e.g., no extraction or centrifugation), routine analysis of a number of samples without reagents, minimal procedural steps, and easy operation and interpretation of results. The Raman technique, which has been used to obtain structural information on melamine, is an alternative approach that can be applied to solid materials with no sample pretreatment. In addition, the use of the Fourier transform (FT) methodology and a 1064 nm excitation laser in the near-infrared (NIR) region provides precise wavenumber measurement and good-quality Raman spectra by reducing the interference from fluorescence and photodecomposition of colorants present in food and feed. Raman studies of melamine and melamine-modified resins have been reported in the literature, and the results have revealed the feasibility of the Raman technique for the structural characterization of melamine state in resins. However, there have been few reports on Raman investigation and identification of melamine in complex food and feed systems. The objectives of this study were (1) to identify the characteristic Raman bands in melamine-contaminated wheat flour, corn gluten, and soybean meal mixtures; and (2) to develop simple and universal ratio algorithms for qualitative and quantitative analysis of melamine in mixtures. The ultimate goal is to develop both Raman spectroscopy and Raman chemical imaging methods for rapid, accurate, nondestructive, specific, and routine screening of the presence of melamine in food and feed for public/animal safety and security.

Detection of melamine in milk powders using near-infrared hyperspectral imaging combined with regression coefficient of partial least square regression model

Illegal use of nitrogen-rich melamine (C3H6N6) to boost perceived protein content of food products such as milk, infant formula, frozen yogurt, pet food, biscuits, and coffee drinks has caused serious food safety problems. Conventional methods to detect melamine in foods, such as Enzyme-linked immunosorbent assay (ELISA), High-performance liquid chromatography (HPLC), and Gas chromatography-mass spectrometry (GC-MS), are sensitive but they are time-consuming, expensive, and labor-intensive. In this research, near-infrared (NIR) hyperspectral imaging technique combined with regression coefficient of partial least squares regression (PLSR) model was used to detect melamine particles in milk powders easily and quickly. NIR hyperspectral reflectance imaging data in the spectral range of 990-1700nm were acquired from melamine-milk powder mixture samples prepared at various concentrations ranging from 0.02% to 1%. PLSR models were developed to correlate the spectral data (independent variables) with melamine concentration (dependent variables) in melamine-milk powder mixture samples. PLSR models applying various pretreatment methods were used to reconstruct the two-dimensional PLS images. PLS images were converted to the binary images to detect the suspected melamine pixels in milk powder. As the melamine concentration was increased, the numbers of suspected melamine pixels of binary images were also increased. These results suggested that NIR hyperspectral imaging technique and the PLSR model can be regarded as an effective tool to detect melamine particles in milk powders.