Near-infrared hyperspectral imaging for predicting colour, pH and tenderness of fresh beef
Highlights
► We select feature wavelengths for developing multispectral imaging system. ► We use NIR hyperspectral imaging to predict fresh beef quality. ► We visualize colour and pH distributions in beef sample.
Introduction
In the past few decades the meat industry has immensely thrived as demands for superior food quality continues to grow on both international and domestic markets (Desmond et al., 2000, McDonald et al., 2001, McDonald and Sun, 2001). Interests in meat quality are driven by the need to supply the consumer with a consistent high quality product at an affordable price (Sepúlveda et al., 2011, ElMasry and Sun, 2010). To fulfil consumer’s satisfaction it is very important to provide meat products that can better meet market requirements and to refocus the meat industry on the customer’s needs because it directly impacts on its profitability (Troy and Kerry, 2010). To realize these needs, it is a crucial element within the meat industry to accurately assess meat quality attributes by improving modern techniques for quality evaluation (Herrero, 2008).
In the production of meat products, the great variability in raw meat often leads to highly variable products being marketed without a controlled level of quality. This variability originates from differences in both animal production and meat processing, which imposes great pressure on the food manufacturing industry to guarantee the quality of meat (Cozzolino et al., 2002). The problem is aggravated when the industry is unable to satisfactorily characterize this level of quality and cannot therefore market their products with a certified quality level (Damez and Clerjon, 2008). Hence, inferior quality meats are essential to be identified on the slaughter line, so that they can be handled and marketed separately from high quality meats (Byrne et al., 1998). Assortment of intrinsic and extrinsic quality cues of fresh meat are usually used by consumers for choosing and purchasing high-quality meat products at the points of sale and consumption. Healthy appearance, colour, visible drip, visible fat and marbling and tenderness are among the main quality attributes usually sought by the consumers.
Traditional methods for assessing meat quality attributes are time consuming, destructive and are associated with inconsistency and variability. Instrumental techniques for rapid screening of meat properties to improve control and classification of the product in mechanised processes are of great interest for both the industry and the consumers. By these techniques, the slaughter industry can gain knowledge regarding the meat quality early after sticking in order to facilitate sorting before further use. One of these techniques is the image scanning system (Belk et al., 2000, Vote et al., 2003, Hopkins et al., 2004) developed as a non-invasive method operating at normal abattoir chain speeds for automatic inspection of carcasses. However, this system has some limitations and needs therefore to be augmented with other suitable systems to measure meat eating quality traits (Cubero et al., 2011).
On the other hand, among the numerous techniques which have been proposed for meat quality evaluation on the fresh intact product, near infrared (NIR) spectroscopy proved a great potential for continuous monitoring and controlling of process and product quality in food processing industry (Huang et al., 2008). Since NIR is a rapid method, it is considered as a suitable tool to implement frequent quality control during the entire meat processing chain (Konstantinos and Athanasia, 2004). Furthermore NIR technology can provide complete information on the chemical constituents in a sample scanned, it is thus a convenient tool for characterising foods (Andrés et al., 2007, Wu et al., 2008). In contrast to conventional methods for the determination of meat quality, spectral techniques enable rapid, simple and simultaneous assessment of numerous meat properties without sample preparation, resulting in possibly replacing expensive and slow reference methods (Heigl et al., 2009, Prieto et al., 2009, Prevolnik et al., 2010). Spectral techniques in the visible and near-infrared ranges have already found considerable applications in food and meat products (Osborne et al., 1997). Also, common applications with meats include the quantitative prediction of chemical composition such as fat, water and protein (Tøgersen et al., 1999) as well as for predicting physical characteristics such as colour (Liu et al., 2003), pH (Andersen et al., 1999), tenderness (Shackelford et al., 2005, Rust et al., 2008, Bowling et al., 2009), drip loss and water holding capacity (Prevolnik et al., 2010) and other physical, sensory and technological characteristics (Geesink et al., 2003, Leroy et al., 2003, Liu et al., 2003, Cozzolino and Murray, 2004;). However, spectroscopy measurements suffer critically from the small sample area (limited spatial information) which cannot be representative of such a heterogeneous material as meat (Brøndum et al., 2000).
As an extension of both spectroscopy and imaging techniques, hyperspectral imaging technique has been emerged to integrate both techniques in one system to provide spectral and spatial backgrounds simultaneously. In recent years there have been growing interests in this technology from researchers around the world for non-destructive analysis in many research and industrial sectors (ElMasry et al., 2007, ElMasry et al., 2009, Cluff et al., 2008, Naganathan et al., 2008a, Naganathan et al., 2008b, Menesatti et al., 2009, Lorente et al., 2011a). The use of hyperspectral imaging has approached to be a viable alternative to the conventional imaging and spectroscopy in a wide range of applications. Therefore, developing quality evaluation system based on hyperspectral imaging technology to assess meat quality attributes and to ensure its authentication would thus bring economical benefits to the meat industry by increasing consumer confidence in the quality of the meat products. Although this technology has not yet been sufficiently exploited in meat processing lines and for quality assessment, its potential is very promising. Indeed, it is important to emphasize that the current ‘hyperspectral’ imaging systems are a laboratory-based system, which is not yet ready for implementation in meat processing lines due to its high dimensionality as well as time constraints for image acquisition and subsequent image analyses. Therefore, the challenge is to seek the most sensitive wavebands to predict the essential meat quality attributes, leading to the development of an optimized ‘multispectral’ imaging system that could be implemented directly in industrial applications (ElMasry and Sun, 2010, Lorente et al., 2011b).
Therefore, the prime focus of this study was to develop a NIR hyperspectral imaging system (900–1700 nm) for objective prediction of some quality attributes (colour, pH and tenderness) of intact fresh beef. Specific objectives were to:
- •
establish a NIR hyperspectral imaging system in the NIR spectral range of 900–1700 nm;
- •
extract spectral information, build multivariate analysis models and identify the sensitive wavelengths most related to colour, pH and tenderness prediction; and
- •
develop image processing algorithms for visualizing variations in quality attributes in beef samples.
Section snippets
Preparation of beef samples
A total of 27 bulls from three different breeds (Holstein–Friesian, Jersey × Holstein–Friesian and Norwegian Red × Holstein–Friesian) with nine bulls from each breed were slaughtered at a commercial slaughterhouse (Meadow Meats, Rathdowney, Co. Laois, Ireland). The mean hot carcass weight was 300.60 ± 34.36 kg (ranged from 242.0–386.6 kg). At 24 h post-mortem, three muscles (M. longissimus dorsi (LD), M. semitendinosus (ST) and M. psoas major (PM)) were dissected from each carcass and then sliced to
Spectral profiles
The acquired hyperspectral image (hypercube) consists of a series of 237 contiguous sub-images; each one represents the intensity and spatial distribution of the tested beef sample from 910 to 1700 nm. All individual sub-images could be easily picked up from the ‘hypercube’ at any wavelength(s) to display the sample at this wavelength. Generally speaking, the hyperspectral image described as I(x, y, λ) can be viewed either as a separate spatial sub-image I(x, y) at each wavelength (λ) as shown in
Conclusions
This study was carried out to evaluate the feasibility of using a hyperspectral imaging system in the NIR spectral region (900–1700 nm) for rapid prediction of some quality attributes in intact fresh beef. The results discussed in this paper indicated that it was possible to utilize this non-destructive technique to simultaneously predict both intrinsic quality (tenderness and pH) and extrinsic quality (colour). By means of PLSR, correlations were established between the NIR absorbance spectra
Acknowledgements
The authors gratefully acknowledge the financial support provided by the Irish Government Department of Agriculture, Fisheries and Food under the Food Institutional Research Measure (FIRM) programme.
References (62)
- et al.
Chemical and discriminant analysis of bovine meat by near infrared reflectance spectroscopy (NIRS)
Meat Science
(2003) - et al.
Prediction of sensory characteristics of lamb meat samples by near infrared reflectance spectroscopy
Meat Science
(2007) - et al.
Optical measurements of pH in meat Meat Science
(1999) - et al.
Using reflectance spectroscopy to predict beef tenderness
Meat Science
(2009) - et al.
Prediction of water-holding capacity and composition of porcine meat by comparative spectroscopy
Meat Science
(2000) - et al.
Non-destructive prediction of selected quality attributes of beef by near-infrared reflectance spectroscopy between 750 and 1098 nm
Meat Science
(1998) - et al.
Theory and application of near infrared reflectance spectroscopy in determination of food quality
Trends in Food Science & Technology
(2007) - et al.
Identification of animal meat muscles by visible and near infrared reflectance spectroscopy
LWT-Food Science and Technology
(2004) - et al.
Meat quality assessment using biophysical methods related to meat structure
Meat Science
(2008) - et al.
Effect of rapid and conventional cooling methods on the quality of cooked ham joints
Meat Science
(2000)
Hyperspectral imaging for non-destructive determination of some quality attributes for strawberry
Journal of Food Engineering
Detecting chilling injury in red delicious apple using hyperspectral imaging and neural networks
Postharvest Biology and Technology
Prediction of pork quality attributes from near infrared reflectance spectra
Meat Science
Near-infrared reflection spectroscopy and partial least squares regression for determining the total carbon coverage of silica packings for liquid chromatography
Vibrational Spectroscopy
Raman spectroscopy a promising technique for quality assessment of meat and fish: a review
Food Chemistry
Prediction of sensory characteristics of beef by near-infrared spectroscopy
Meat Science
Video image analysis in the Australian meat industry – precision and accuracy of predicting lean meat yield in lamb carcasses
Meat Science
Near infrared spectroscopy for on/in-line monitoring of quality in foods and beverages: a review
Journal of Food Engineering
Prediction of color, texture, and sensory characteristics of beef steaks by visible and near infrared reflectance spectroscopy: a feasibility study
Meat Science
Effect of evacuation rate on the vacuum cooling process of a cooked beef product
Journal of Food Engineering
The effect of injection level on the quality of a rapid vacuum cooled cooked beef product
Journal of Food Engineering
Visible/near-infrared hyperspectral imaging for beef tenderness prediction
Computers and Electronics in Agriculture
Predicting pork water-holding capacity with NIR spectroscopy in relation to different reference methods
Journal of Food Engineering
Ability of near infrared reflectance spectroscopy (NIRS) to estimate physical parameters of adult steers (oxen) and young cattle meat samples
Meat Science
Potential use of near infrared reflectance spectroscopy (NIRS) for the estimation of chemical composition of oxen meat samples
Meat Science
Potential use of near infrared reflectance spectroscopy (NIRS) for the estimation of chemical composition of oxen meat samples
Meat Science
Application of near infrared reflectance spectroscopy to predict meat and meat products quality: a review
Meat Science
Prediction of drip-loss, pH, and color for pork using a hyperspectral imaging technique
Meat Science
Prediction of beef quality attributes from early post mortem near infrared reflectance spectra
Food Chemistry
Changes in meat quality characteristics of bovine meat during the first 6 days post-mortem
Meat Science
Aspects of quality related to the consumption and production of lamb meat consumers versus producers
Meat Science
Cited by (381)
Exploring the potential of hyperspectral imaging for microbial assessment of meat: A review
2024, Spectrochimica Acta - Part A: Molecular and Biomolecular SpectroscopyBeef marbling assessment by structured-illumination reflectance imaging with deep learning
2024, Journal of Food EngineeringHyperspectral imaging–based assessment of fresh meat quality: Progress and applications
2024, Microchemical JournalComputer vision in livestock management and production
2024, Engineering Applications in Livestock ProductionQuantification of polycyclic aromatic hydrocarbons in roasted Tan lamb using fluorescence hyperspectral imaging technology
2023, Journal of Food Composition and Analysis
- 1
Permanent address: Agricultural Engineering Department, Suez Canal University, Ismailia, Egypt.