Comparison of various wavelet texture features to predict beef palatability

The wavelet transform can be used to characterise the surface texture of beef images in a more efficient manner than classical algorithms such as co-occurrence and run lengths. Features extracted from wavelet decompositions have been used to develop predictive models of important palatability attributes. A variety of common wavelet transforms were considered (biorthogonal, reverse biorthogonal, discrete Meyer, Daubechie, symmetric modified Daubechie and Coifman modified Daubechie) to search for the most useful texture features. A classic run length and co-occurrence algorithm was used for comparison. Using the same data analysis methods for each wavelet type, predictive models of beef acceptability, tenderness, juiciness, flavour and hardness were developed. Genetic algorithms succeeded in finding more accurate models than stepwise and manual elimination except for hardness. An accurate model of flavour (r² = 0.84) was computed. A good model of overall acceptability (r² = 0.79) was computed that fell just short of an important benchmark of accuracy. An encouraging model of juiciness (r² = 0.71) was computed showing that with additional palatability information juiciness might be accurately modelled. Tenderness proved difficult to model with only the classic model satisfying stability criteria and a poorer result (r² = 0.64) meaning substantial additional palatability information is required for accurate modelling. Hardness was particularly difficult to model. The biorthogonal wavelet produced the best model for three palatability measurements but the symmetric modified Daubechie wavelet produced the best model of overall acceptability and thus must be viewed as the most useful wavelet type.     

Prediction of beef palatability from colour,marbling and surface texture features of longissimus dorsi

The palatability of beef has been investigated with digital imaging systems on numerous previous occasions. In the current study, a novel approach was applied using high magnification imaging to develop surface texture features and an alternative colour space greyscale to express muscle surface texture. An automatic segmentation method was applied to develop colour and marbling features and best regression model subsets were selected automatically with genetic algorithms. Results indicated that accurate modelling of beef acceptability with regression models was possible with r² up to 0.95. Modelling of acceptability using high magnification images proved more successful than modelling with low magnification images. Linear models performed well compared to non-linear models. Other sensory measurements particularly TPA hardness were more difficult to model, although an accurate model of juiciness was developed. Addition of non-linear terms did not give large improvements except for juiciness.     

Correlation of consumer assessment of longissimus dorsi beef palatability with image colour,marbling and surface texture features

A new study was conducted to apply computer vision methods successfully developed using trained sensory panel palatability data to new samples with consumer panel palatability data. The computer vision methodology utilized the traditional approach of using beef muscle colour, marbling and surface texture as palatability indicators. These features were linked to corresponding consumer panel palatability data with the traditional approach of partial least squares regression (PLSR). Best subsets were selected by genetic algorithms. Results indicate that accurate modelling of likeability with regression models was possible (r² = 0.86). Modelling of other important palatability attributes proved encouraging (tenderness r² = 0.76, juiciness r² = 0.69, flavour r² = 0.78). Therefore, the current study provides a basis for further expanding computer vision methodology to correlate with consumer panel palatability data.     

Prediction of beef eating quality from colour, marbling and wavelet texture features

Beef longissimus dorsi colour, marbling fat and surface texture are long established properties that are used in some countries by expert graders to classify beef carcasses, with subjective and inconsistent decision. As a computer vision system can deliver objective and consistent decisions rapidly and is capable of handling a greater variety of image features, attempts have been made to develop computerised predictions of eating quality based on these and other properties but have failed to adequately model the variation in eating quality. Therefore, in this study, examination of the ribeye at high magnification and consideration of a broad range of colour and marbling fat features was used to attempt to provide better information on beef eating quality. Wavelets were used to describe the image texture of the beef surface at high magnification rather than classical methods such as run lengths, difference histograms and co-occurrence matrices. Sensory panel and Instron analyses were performed on duplicate steaks to measure the quality of the beef. Using the classical statistical method of partial least squares regression (PLSR) it was possible to model a very high proportion of the variation in eating quality (r² = 0.88 for sensory overall acceptability and r² = 0.85 for 7-day WBS). Addition of non-linear texture terms to the models gave some improvements.     

Identification of important image features for pork and turkey ham classification using colour and wavelet texture features and genetic selection

A method to discriminate between various grades of pork and turkey ham was developed using colour and wavelet texture features. Image analysis methods originally developed for predicting the palatability of beef were applied to rapidly identify the ham grade. With high quality digital images of 50–94 slices per ham it was possible to identify the greyscale that best expressed the differences between the various ham grades. The best 10 discriminating image features were then found with a genetic algorithm. Using the best 10 image features, simple linear discriminant analysis models produced 100% correct classifications for both pork and turkey on both calibration and validation sets.     

Classification of tough and tender beef by image texture analysis

Texture features of fresh-beef images were extracted and used to classify steaks into tough and tender groups in terms of cooked-beef tenderness. Crossbred steers varying in quality were processed in a commercial plant and two short loin steaks were sampled from each carcass. One sample was used for imaging and the other was broiled for sensory evaluation of tenderness by a trained panel. The samples were segregated into tough and tender groups according to the sensory scores. A wavelet-based decomposition method was used to extract texture features of fresh-beef images. The texture feature data for 90 sample images were used to train and test sample calssifiers in a rotational leave-one-out scheme. A correct classification rate of 83.3% was obtained in cross validations. While texture features alone may not be sufficient to segregate beef products into many levels of tenderness, they can be significant members in a set of indicators that will lead to adequate tenderness prediction.     

Automatic segmentation of beef longissimus dorsi muscle and marbling by an adaptable algorithm

An algorithm for automatic segmentation of beef longissimus dorsi (LD) muscle and marbling has been developed. The algorithm used simple thresholding to remove the background and then used clustering and thresholding with contrast enhancement via a customised greyscale to remove marbling. It was possible to attain lean muscle free of obvious marbling or background pixels where specular reflection could be effectively mitigated. Features of the automatically derived LD muscle and marbling images were compared to corresponding features of LD muscle and marbling images derived with a segmentation method requiring manual completion. Very strong correlations (up to r = 1) were found between the colour features of both sets of LD muscle images. Strong correlations (up to r = 0.96) were found between the features of both sets of marbling images. The automatic segmentation method has shown its good ability to approximate colour and marbling features. The algorithm has adaptable parameters and can be retailored to suit different image acquisition environments.     

Automatic measurement of pores and porosity in pork ham and their correlations with processing time, water content and texture

Pores formed in pork ham have a significant effect on its quality. However, they are mostly characterised using manual methods with special devices. In this paper, an automatic method for pore characterisation of pork ham was developed using computer vision. To segment pores from images of pork ham, three stages of image processing algorithm were developed, i.e., ham extraction, image enhancement, and pore segmentation. From the segmented pores, the porosity, number of pores, pore size, and size distribution were measured. The statistical analysis showed that 79.81% of pores have area sizes between 6.73×10(-3) and 2.02×10(-1)mm(2). Furthermore, it was found that the total number of pore (TNP) and porosity highly negatively related to the water content of pork ham (P<0.05), and had negative correlations with the cooking and cooling time. However, for texture analysis, positive correlations were found between the pore characterisations and WBS, hardness, cohesion, and chewiness, respectively, while springiness and gumminess were negatively related to TNP and porosity.     

The influence of attachment to beef surfaces on the survival of cells of Salmonella enterica serovar Typhimurium DT104, at different a(w) values and at low storage temperatures

This study investigated the influence of attachment to beef surfaces on the survival, injury and death of stationary phase cells of Salmonella enterica serovar Typhimurium DT104, compared to cells free in solution. The effects on cells are considered at different a(w) values and low temperatures in relation to osmotic and cold temperature shock effects. Attachment of cells to meat surfaces prevented cell injury and death from hyperosmosis and low temperatures, compared to meat solutions. Storage of cells for 72h resulted in higher levels of cell death on cells attached to meat surfaces. The improved survival of cells in solutions was considered to be related to adaptation to osmotic stress as a result of exposure to a previous hyperosmotic shock and the ability of the cells to produce cold shock proteins. Pathogen cell growth at low temperatures is discussed in relation to the presence of low levels of NaCl. Finally the data is discussed in relation to pathogen survival on beef carcass surfaces during refrigeration.     

Comparison of primary predictive models to study the growth of Listeria monocytogenes at low temperatures in liquid cultures and selection of fastest growing ribotypes in meat and turkey product slurries

This study compared the performance of four primary mathematical models to study the growth kinetics of Listeria monocytogenes ribotypes grown at low temperature so as to identify the best predictive model. The parameters of the best-fitting model were used to select the fastest growing strains with the shortest lag time and greatest growth rate. Nineteen food, human and animal L. monocytogenes isolates with distinct ribotype were grown at 4, 8, and 12 degrees C in tryptic soy broth and slurries prepared from cooked uncured sliced turkey breasts (with or without potassium lactate and sodium diacetate, PL/SD) and cooked cured frankfurters (with or without PL/SD). Separate regressions were performed on semi-logarithm growth curves to fit linear (based on Monod) and non-linear (Gompertz, Baranyi-Roberts, and Logistic) equations and performance of each model was evaluated using an F-test. No significant differences were found in the performance of linear and non-linear models, but the Baranyi model had the best fit for most growth curves. The maximum growth rate (MGR) of Listeria strains increased with the temperature. Similarly MGR was found significantly greater when no antimicrobials were present in the formulation of turkey or frankfurter products. The variability in lag times and MGRs in all media as determined by the Baranyi model was not consistent among strains. No single strain consistently had the fastest growth (shortest lag time, fastest MGR, or shortest time to increase 100-fold), but nine strains were identified as fastest growing strains under most growth conditions. The lack of association between serotype and fastest strain was also observed in the slurry media study. The fastest growing strains resulting from this study can be recommended for future use in L. monocytogenes challenge studies in delicatessen meat and poultry food matrices, so as to develop conservative pathogen growth predictions.