Image analysis application for automatic quantification of intramuscular connective tissue in meat

The application of computer vision systems in meat science and industry - A review

Computer vision systems (CVS) are applied to macro- and microscopic digital photographs captured using digital cameras, ultrasound scanners, computer tomography, and wide-angle imaging cameras. Diverse image acquisition devices make it technically feasible to obtain information about both the external features and internal structures of targeted objects. Attributes measured in CVS can be used to evaluate meat quality. CVS are also used in research related to assessing the composition of animal carcasses, which might help determine the impact of cross-breeding or rearing systems on the quality of meat. The results obtained by the CVS technique also contribute to assessing the impact of technological treatments on the quality of raw and cooked meat. CVS have many positive attributes including objectivity, non-invasiveness, speed, and low cost of analysis and systems are under constant development an improvement. The present review covers computer vision system techniques, stages of measurements, and possibilities for using these to assess carcass and meat quality.

A Computer Vision-Based Method for Classification of Red Meat Quality After Nitrosamine Appendage

Nitrosamine is a carcinogenic chemical used as a preservative in red meat whose identification is an ordeal. This paper presents a computer vision-based non-destructive method for identifying quality disparities between preservative treated and untreated (control) red meat. To access the discrepancy in the quality of red meat, both traditional machine learning and deep learning-based methods have been used. Support vector machine (SVM) classifier and artificial neural network (ANN) models have been used to detect the presence of nitrosamine in test samples. The paper also made use of different pre-trained deep convolutional neural networks (DCNN) with transfer learning approach such as ResNet-34, ResNet-50, ResNet-101, VGG-16, VGG-19, AlexNet and MobileNetv2 to examine the presence of nitrosamine in the food samples. While the ANN classifier performed better in comparison to the SVM classifier, the highest testing accuracy and F1-score were obtained using the deep learning model, ResNet-101 with 95.45% and 96.54%, respectively. The experimental results demonstrate an improved performance in comparison to the existing methods; indicating the feasibility of the proposed work for food quality control in real-time applications.

Characterization of Longissimus thoracis, Semitendinosus and Masseter muscles and relationships with technological quality in pigs. 1. Microscopic analysis of muscles

Three porcine muscles (Longissimus thoracis, Semitendinosus, Masseter), known to have large differences in biochemical and histological traits, were fully characterized and the link between muscle structure and quality evaluated. The oxidative Masseter had more pigment, higher content of metmyoglobin, haem iron, protein and collagen, and was redder with higher fibre numbers, fibre circularity, pH and water holding capacity than the glycolytic Longissimus. Fibre type distribution showed predominance of type IIB in Longissimus and Semitendinosus white, type I in Semitendinosus red and IIA in Masseter. Type I fibres were larger than type IIB and IIA in Semitendinosus and Masseter, respectively, but not in the Longissimus, indicating that fibre size is muscle dependent. Muscle redness was positively correlated with type I fibre traits, haem iron and metmyoglobin, and negatively associated with type II fibre characteristics, non-haem iron and oxymyoglobin. Expressible juice had positive correlation with fibre size and negative with fibre number and connective tissue.

Relationships between structural characteristics of bovine intramuscular connective tissue assessed by image analysis and collagen and proteoglycan content

Three muscles (Longissimus thoracis, Semimembranosus, Biceps femoris) of 40 young bulls from 3 breeds were used to quantify structural characteristics of bovine connective tissue by image analysis, with both macroscopic and microscopic approaches. Collagen and proteoglycan content was also investigated. Perimysium occupied a greater area (8 vs 6%), and was wider (42 vs 2 μm) and shorter per unit area (1.9 vs 30 mm mm(-2)) than endomysium. Perimysium and endomysium from Longissimus were thinner, less ramified than in Biceps. Longissimus showed less total collagen and cross-linking and more proteoglycans (P<0.0001) than Biceps muscle. Blond d'Aquitaine perimysium occupied less area, was more ramified and muscles contained less collagen, cross-linking and more proteoglycans than Angus (P<0.001). Limousin was intermediate. High proteoglycan content in muscle containing less total collagen suggested a complementarity between these molecules. They might influence mechanical properties of intramuscular connective tissue. This was especially true given that proteoglycans and total collagen were negatively and positively linked with structural parameters, respectively.