Measurement of meat color using a computer vision system

The limits of the colorimeter and a technique of image analysis in evaluating the color of beef, pork, and chicken were investigated. The Minolta CR-400 colorimeter and a computer vision system (CVS) were employed to measure colorimetric characteristics. To evaluate the chromatic fidelity of the image of the sample displayed on the monitor, a similarity test was carried out using a trained panel. The panelists found the digital images of the samples visualized on the monitor very similar to the actual ones (P < 0.001). During the first similarity test the panelists observed at the same time both the actual meat sample and the sample image on the monitor in order to evaluate the similarity between them (test A). Moreover, the panelists were asked to evaluate the similarity between two colors, both generated by the software Adobe Photoshop CS3 one using the L*, a* and b* values read by the colorimeter and the other obtained using the CVS (test B); which of the two colors was more similar to the sample visualized on the monitor was also assessed (test C). The panelists found the digital images very similar to the actual samples (P < 0.001). As to the similarity (test B) between the CVS- and colorimeter-based colors the panelists found significant differences between them (P < 0.001). Test C showed that the color of the sample on the monitor was more similar to the CVS generated color than to the colorimeter generated color. The differences between the values of the L*, a*, b*, hue angle and chroma obtained with the CVS and the colorimeter were statistically significant (P < 0.05–0.001). These results showed that the colorimeter did not generate coordinates corresponding to the true color of meat. Instead, the CVS method seemed to give valid measurements that reproduced a color very similar to the real one.     

Predicting meat yields and commercial meat cuts from carcasses of young bulls of Spanish breeds by the SEUROP method and an image analysis system

The SEUROP system is currently in use for carcass classification in Europe. Image analysis and other new technologies are being developed to enhance and supplement this classification system. After slaughtering, 91 carcasses of local Spanish beef breeds were weighed and classified according to the SEUROP system. Two digital photographs (a side and a dorsal view) were taken of the left carcass sides, and a total of 33 morphometric measurements (lengths, perimeters, areas) were made. Commercial butchering of these carcasses took place 24 h postmortem, and the different cuts were grouped according to four commercial meat cut quality categories: extra, first, second, and third. Multiple regression analysis of carcass weight and the SEUROP conformation score (x variables) on meat yield and the four commercial cut quality category yields (y variables) was performed as a measure of the accuracy of the SEUROP system. Stepwise regression analysis of carcass weight and the 33 morphometric image analysis measurements (x variables) and meat yield and yields of the four commercial cut quality categories (y variables) was carried out. Higher accuracy was achieved using image analysis than using only the current SEUROP conformation score. The regression coefficient values were between R² = 0.66 and R² = 0.93 (P < 0.001) for the SEUROP system and between R² = 0.81 and R² = 0.94 (P < 0.001) for the image analysis method. These results suggest that the image analysis method should be helpful as a means of supplementing and enhancing the SEUROP system for grading beef carcasses.     

Comparison of the predictive power of beef surface wavelet texture features at high and low magnification

Beef longissimus dorsi surface texture is an indicator used in predicting beef palatability by expert graders. Computer vision systems have previously used imaging at normal view to develop surface texture features with some success. Good models of beef overall acceptability using imaging at high magnification have been recently developed. As a comparison the same surface texture features were computed from the corresponding images at normal view and used to model overall acceptability. Both sets of texture features were also combined with muscle colour and marbling features and used to model overall acceptability. Models using texture features alone were more successful at normal modality. However colour and marbling features combined much better with texture features at high modality to yield the most accurate model of overall acceptability (r² = 0.93). Accurate Partial Least Squares Regression (PLSR) models were computed at both modalities with and without inclusion of colour and marbling features. Addition of squared terms to the models failed to improve accuracy.     

基于形态学特征的对虾完整性识别方法构建

目的 针对对虾加工过程中缺损对虾混入完整对虾从而降低对虾产品外观品质的问题, 构建基于形态学特征的对虾完整性识别方法。方法 首先, 使用灰度差异法处理对虾图像, 经过连通域、中值滤波等形态学操作后, 得到较为完整的感兴趣区域图像, 再对其采取二值化、轮廓化等操作; 然后, 对轮廓提取骨架线, 并求轮廓内最大内切圆直径以得到长宽比特征, 并求其圆度特征; 最后, 将以上2个特征作为判别对虾完整性的核心指标, 构建融合特征判别算法。结果 本研究所提算法应用于1063幅生鲜虾测试集图像中识别准确率达到99.25%, 相比于传统曲率法, 识别准确率提升了6.48%, 识别时间降低了1598.6 ms。结论 该方法具有较大优势和应用前景, 为开发大规模南美白对虾在线品质的无损检测装备提供关键技术。     

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.     

Development of a computer vision system to measure the color of potato chips

The objective of this research was to design and implement an inexpensive computer vision system for measuring the color of a highly heterogeneous food material not only in shape as well in color such as potato chips in L^*a^*b^* units from RGB images. The system was composed of (i) a digital color camera for acquiring the images in a digital format, (ii) a computer for storage the images, (c) image analysis routines integrated into a software programmed in Matlab that converts the color RGB of the food image into L^*a^*b^* units. In this way the color of potato chips can be calculated in L^*a^*b^* units over representative areas and in a reproducible way. The kinetics of color changes in potato slices during frying at four temperatures was followed using the implemented computer vision system (CVS). Color values in L^*a^*b^* units were recorded at different sampling times during frying at the four oil temperatures using the total color change parameter (ΔE). Chips fried at higher temperatures get darker as expected and showed by the CVS. The implemented computer vision system can be used to study as well foods different from potato chips by selecting their proper settings for image acquisition and digital image processing.     

Diagnostic model for assessing traceability system performance in fish processing plants

This paper introduces a diagnostic tool that can be used by fish processing companies to evaluate their own traceability systems in a systematic manner. The paper begins with discussions on the rationale of traceability systems in food manufacturing companies, followed by a detailed analysis of the most important indicators in the designing and executing traceability systems. The diagnostic tool is presented in four grids through which fish companies can evaluate their own developed traceability system. The paper argues that if a company operates at a higher level of contextual factors, then design and execution of traceability system needs to be at a higher level as well so as to achieve a higher level of traceability system performance. The paper concludes that companies that are able to systematically assess their own developed traceability systems are able to determine food safety problems well in advance, and thereby take appropriate corrective actions.     

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.     

Robust colour calibration of an imaging system using a colour space transform and advanced regression modelling

A new algorithm for the conversion of device dependent RGB colour data into device independent L*a*b* colour data without introducing noticeable error has been developed. By combining a linear colour space transform and advanced multiple regression methodologies it was possible to predict L*a*b* colour data with less than 2.2 colour units of error (CIE 1976). By transforming the red, green and blue colour components into new variables that better reflect the structure of the L*a*b* colour space, a low colour calibration error was immediately achieved (ΔE(CAL) = 14.1). Application of a range of regression models on the data further reduced the colour calibration error substantially (multilinear regression ΔE(CAL) = 5.4; response surface ΔE(CAL) = 2.9; PLSR ΔE(CAL) = 2.6; LASSO regression ΔE(CAL) = 2.1). Only the PLSR models deteriorated substantially under cross validation. The algorithm is adaptable and can be easily recalibrated to any working computer vision system. The algorithm was tested on a typical working laboratory computer vision system and delivered only a very marginal loss of colour information ΔE(CAL) = 2.35. Colour features derived on this system were able to safely discriminate between three classes of ham with 100% correct classification whereas colour features measured on a conventional colourimeter were not.     

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.     

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.     

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.     

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.     

Investigation of image registration method for the multi-directional image fusion of woven fabrics

Abstract

Traditional single-side scanning or single-vision image acquisition methods have the limitation of incomplete information caused by the existence of blind spots. To collect the complete texture information of fabric images, a new multi-vision image acquisition and the related fusion method is developed to solve this problem. However, linear addition of image sequences acquired from multiple directions cannot achieve a good result of image fusion, it is necessary to conduct the image fusion based on the image registration between images at pixel level. Therefore, a new multi-directional digital image acquisition system for woven fabrics is established in this article; one set of image fusion algorithm based on image registration is proposed for the image enhancement of fabric. Fabric texture images are digitized by means of multi-directional vision imaging instead of unidirectional imaging, the structural information of fabric texture could be enhanced using image registration and fusion technology and the indexing and localization of texture corresponding points could be controlled using matching points or control points. Our experimental results show that the proposed method could be used to merge the effective information from the multi-directional vision images completely, it has the potential application for the rendering of woven fabrics using image driven virtual reality enhancement.     

A DNA method for qualitative identification of plant raw materials in feedstuff

This research developed a simple and not expensive DNA method for the qualitative identification of plant raw materials used as feed mixtures. Specific simple sequence tagged (STS) markers were developed to detect faba bean (Lectin A gene), field pea (Convicilin A gene), grain sorghum (UDP-glucosyltransferase gene) and barley (Hordoindoline-a gene), whereas identification of durum and common wheat (lipid transfer protein gene), soybean (Gly m Bd 30K allergen gene) and maize (invertase gene) was carried out using markers available from the literature. Cross-reactivity of the primer pairs was also checked against oat, rye, kidney bean and lentil. The method was effectively applied to the analysis of flour mixtures and extruded feedstuff. It could be included in traceability and certification of animal feeding systems within high quality animal production chains which are strictly related to the production area by the valorisation of locally grown raw materials.     

Quality control of Toxoplasma gondii in meat packages: standardization of an ELISA test and its use for detection in rabbit meat cuts

Toxoplasma gondii causes severe disease both to man and livestock and its detection in meat after slaughtering requires PCR or biological tests. Meat packages contain retained exudate that could be used for serology due to its blood content. Similar studies reported false negative assays in those tests. We standardized an anti-T. gondii IgG ELISA in muscle juices from experimentally infected rabbits, with blood content determination by cyanhemoglobin spectrophotometry. IgG titers and immunoblotting profiles were similar in blood, serum or meat juice, after blood content correction. These assays were adequate regardless of the storage time up to 120 days or freeze-thaw cycles, without false negative results. We also found 1.35% (1/74) positive sample in commercial Brazilian rabbit meat cuts, by this assay. The blood content determination shows ELISA of meat juice may be useful for quality control for toxoplasmosis monitoring.     

Real-time multispectral imaging system for online poultry fecal inspection using unified modeling language

A prototype real-time multispectral imaging system for fecal and ingesta contaminant detection on broiler carcasses has been developed. The prototype system includes a common aperture camera with three optical trim filters (517, 565 and 802-nm wavelength), which were selected by visible/NIR spectroscopy and validated by a hyperspectral imaging system with decision tree algorithm. The on-line testing results showed that the multispectral imaging technique can be used effectively for detecting feces (from duodenum, ceca, and colon) and ingesta on the surface of poultry carcasses with a processing speed of 140 birds per minute. This paper demonstrated both multispectral imaging hardware and real-time image processing software. For the software development, the Unified Modeling Language (UML) design approach was used for on-line application. The UML models included class, object, activity, sequence, and collaboration diagram. User interface model included 17 inputs and 6 outputs. A window-based real-time image processing software composed of 11 components, which represented class, architecture, and activity. Both hardware and software for a real-time fecal detection were tested at the pilot-scale poultry processing plant. The run-time of the software including online calibration was fast enough to inspect carcasses on-line with an industry requirement. Based on the preliminary test at the pilot-scale processing line, the system was able to acquire poultry images in real-time. According to the test results, the imaging system is reliable for the harsh environments and UML-based image processing software is flexible and easy to be updated when additional parameters are needed for in-plant trials. Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.     

Modeling cooling of ready-to-eat meats by 3D finite element analysis: Validation in meat processing facilities

A three-dimensional (3D) finite element model for simulating heat transfer during cooling of irregular-shaped ready-to-eat meat products was developed and validated. The heat transfer model considered conduction as the governing equation, subject to convection, radiation and moisture evaporation boundary conditions. A 3D finite element algorithm developed in Java™ was used to solve the model. The algorithm generated solutions for meshes containing 4-node tetrahedral volume elements and 3-node triangular boundary elements. Product geometries were generated from CT-scan images of the meat products. The model was adapted to receive input parameters that can be easily provided by a meat processors including air relative humidity, air temperature, air velocity, type of casing, duration of water shower, product weight, and estimated core temperature of product prior to entering the cooling chamber. Model validation was conducted in four commercial facilities, under normal processing conditions. Temperatures predicted by the model were in agreement with observed values. Average root-mean-square error (RMSE) was 1.19 ± 0.54 °C for core temperatures, 1.73 ± 0.48 °C for temperatures 0.05 m from core to surface, and 2.01 ± 1.01 °C for surface temperatures. The developed heat transfer model was integrated with predictive microbiology models through a food safety website: numodels4safety.unl.edu. The integration can be useful for estimating the severity of cooling deviations and resulting microbiological safety caused by unexpected cooling process disruptions.     

Textural analysis of hyperspectral images for improving contaminant detection accuracy

Previous studies demonstrated a hyperspectral imaging system has a potential for poultry fecal contaminant detection by measuring reflectance intensity. The simple image ratio at 565 and 517 nm images with optimal thresholding was able to detect fecal contaminants on broiler carcasses with high accuracy. However, differentiating false positives from real contaminants, especially cecal feces were challenging. Further image processing such as textural analysis in the spatial domain was able to reduce false positive errors. In this study, textural analysis of hyperspectral images was conducted to improve detection accuracy by reducing false positives. Specifically, textural analysis with co-occurrence matrix of hyperspectral images performed well to identify “true” contamination. In addition, co-occurrence matrix textural features including average, variance, entropy, contrast, correlation, moment of poultry hyperspectral images were investigated for selecting optimal features to represent contamination. Image pre-processing with co-occurrence textural analysis, specifically mean of co-occurrence textural feature from the matrix (0° angle and distance equals to one) followed by image ratio was able to improve fecal detection accuracy without additional optical filters that increase cost for system hardware of multispectral imaging system for on-line application. Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.