Article

Hyperspectral Imaging for Defect Detection of Pickling Cucumbers

01/2010; DOI: 10.1016/B978-0-12-374753-2.10014-0
Source: OAI
1 Follower
 · 
118 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Near infrared spectroscopy is a quick and non-destructive method to assess the major components (e.g. protein, fat, moisture) in food. Other derived properties, like the chemical changes of aged beef might also be detected by spectral measurement. However, because of the non-homogeneous structure of beef, this method needs destructed (ground) samples to measure. Hyperspectral imaging offers a new way for studying non-homogeneous surfaces, but this measurement technique is much more influenced by the surrounding conditions. Proper calibration method of the sensor and particularly well-defined conditions are needed to reduce the noise and stabilize the measurement. Software environment of push-broom hyperspectral system was developed for controlling sensors and Y-table stepping motors. It supports the setting of optimal AD parameters, two-point spectral calibration, moves the table and acquires frames as to avoid spatial distortion in the hypercube. The hypercube retrieved is saved in ENVI file format for later image processing (ENVI) and statistical analysis (R Project). This measurement system was tested by inspecting ageing state of beef. The ageing state of beef sirloin was estimated by NIR spectral properties as a test application of the system. All the samples were measured by conventional spectrophotometer and hyperspectral system as well to compare their capabilities. The reproducibility of meat measurement was checked first by inspecting samples stored in different conditions (fresh cut, O2- and vacuum-packaging) and left on open air during the 7 times 10 minutes of measurement. The spectra were imported into MS Excel sheet to process and visualize the changes by the time. Aged beef samples were measured then in the duration of 4 times 7 days to prove whether the ageing state can be estimated by the reflected spectra. The location of the samples within sirloin steak was also registered to detect possible influence on their spectra. R Project algorithm analyzed the absorption spectra fitting PLS regression model for predicting ageing time. Although the hyperspectral measurements are usually less accurate than the instrumental spectral ones, the segmentation of different tissues (meat/fat) was expected to improve the efficiency of prediction. ENVI classification method segmented the areas of meat and fat tissues on hyperspectral frames and extracted the average spectra of different tissues. PLS model was built by R Project algorithm on the base of these spectra as well. The measurement was reproducible since the changes of intensity was predictable and the variance of given samples were bigger than the change of the time-average. In case of the ageing experiment, the intensity was characteristic mostly to the origin of sample and less to the ageing time. The spectra were normalized further to focus only to the spectral properties. The NIR spectra were proper to calibrate PLS model for prediction of ageing time with R2= 0.98 (factors= 17). The hyperspectral data, preprocessed by ENVI algorithms, resulted the average spectra of pure flash. Although the hyperspectral acquisition has generally much more noise, these spectra were also enough to calibrate PLS model for ageing time prediction with R2= 0.98 (factors= 15). A non-contact multispectral system can use its significant wavelengths in an industrial application.
    Chinese-European Cooperation For a Long-term Sustainability, Budapest, Hungary; 11/2011
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In recent years, hyperspectral imaging has gained a wide recognition. as a non-destructive and fast quality and safety analysis and assessment method for a wide range of food products. As the second part of this review, applications in quality and safety determination for food products are presented to illustrate the capability of this technique in the food industry for classification and grading, defect and disease detection, distribution visualization of chemical attributes, and evaluations of overall quality of meat, fish, fruits, vegetables, and other food products. The state of the art of hyperspectral imaging for each of the categories was summarized in the aspects of the investigated quality and safety attributes, the used systems (wavelength range, acquisition mode), the data analysis methods (feature extraction, multivariate calibration, variables selection), and the performance (correlation, error, visualization). With its success in different applications of food quality and safety analysis and assessment, it is evident that hyperspectral imaging can automate a variety of routine inspection tasks. Industrial relevance: It is anticipated that real-time food monitoring systems with this technique can be expected to meet the requirements of the modern industrial control and sorting systems in the near future.
    Innovative Food Science & Emerging Technologies 07/2013; 19:15-28. DOI:10.1016/j.ifset.2013.04.016 · 2.25 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This study was aimed at exploring the feasibility of detecting and quantifying melamine, and the structural analogue cyanuric acid, contamination in soybean meal, using line-scan near infrared (NIR) hyperspectral imaging spectroscopy (HIS). Soybean meal is one of the main ingredients used in the feed industry because it offers a complete protein profile. Each year, demand increases for soybean products and soya oil, the consumption of which is directly boosted by Chinese consumers’ growing wealth, and for soybean meal, which is indirectly affected by the growing demand for meat. Recent cases of deliberate melamine contamination of soybean meal have been reported. This study focuses on the development of a methodology based on NIR–HIS for the acquisition, treatment and interpretation of images and spectra, as well as for the detection and quantification of melamine and cyanuric acid contamination in soybean meal. A total of 40 commercial soybean meal samples were collected, and 17 adulterated samples were prepared by adding different amounts of melamine/cyanuric acid to the samples, with concentrations varying between 0.5% and 5%. The spectral data were collected using line-scan NIR–HIS, and a qualitative model was created based on a principal-component analysis (PCA), whereas partial least-squares discriminant analysis was used to obtain a discrimination model and a semi-quantitative prediction of the content of contaminant. This study has permitted the detection of low levels of melamine and also revealed some limitations for the feasibility of quantifying melamine in soybean meal.
    Journal of Near Infrared Spectroscopy 01/2014; 22(2):103. DOI:10.1255/jnirs.1109 · 1.48 Impact Factor

Preview

Download
2 Downloads
Available from