On-line Ham Grading using pattern recognition models based on available data in commercial pig slaughterhouses

Abstract

The thickness of the subcutaneous fat in hams is one of the most important factors for the dry-curing process and largely determines its final quality. This parameter is usually measured in slaughterhouses by a manual metrical measure to classify hams. The aim of the present study was to propose an automatic classification method based on data obtained from a carcass automatic classification equipment (AutoFom) and intrinsic data of the pigs (sex, breed, and weight) to simulate the manual classification system. The evaluated classification algorithms were decision tree, support vector machines (SVM), k-nearest neighbour and discriminant analysis. A total of 4000 hams selected by breed and sex were classified as thin (0-10 mm), standard (11-15 mm), semi-fat (16-20 mm) and fat (>20 mm). The most reliable model, with a percentage of success of 73%, was SVM with Gaussian kernel, including all data available. These results suggest that the proposed classification method can be a useful online tool in slaughterhouses to classify hams.

Automatic ham classification method based on support vector machine model increases accuracy and benefits compared to manual classification

“…The objective of this statistical evaluation was to compare the classification obtained with the manual classification (HC_M) and the automatic classification performed by the SVM, respectively. SVM refers to Gaussian Medium algorithm (HC_SVM), the best model obtained in Masferrer et al (2018), to classify hams according to SFT. In order to evaluate the prediction of the ham class, the SFT of the finished hams was specifically measured for this study, in order to obtain the Golden Standard Ham Class (HC_GS) as described in the previous section.…”

“…Then pigs were eviscerated and automatically split according to standard commercial procedures using a robot. After that, the two half-carcasses were weighted and an experienced operator visually determined the sex of the pig (female, entire male or castrated male) and classified the half carcass as described in Masferrer et al (2018), in order to normalize the classification. The left half carcass was always used to avoid possible errors produced by the robot cut deviation.…”

“…With the measures obtained with Autofom III and including information about sex, breed and warm carcass weight (see Table.1), the ham class predicted by SVM algorithm (HC_SVM) was obtained (Masferrer et al, 2018).…”

“…It is possible to measure SFT once the green ham has been shaped, however it would be interesting to estimate SFT measure on-line in order to classify the ham before being processed (Masferrer et al 2018). Optimization of the industrial processes in slaughterhouses is essential to become more competitive.…”

“…The objective of this study was to compare a SVM classification algorithm with a manual classification system, commonly used in commercial slaughterhouses, in order to classify hams according to their SFT. The SVM algorithm employs a middle Gaussian core, the best model obtained in Masferrer et al (2018), which is trained with intrinsic data of the pigs (sex, weight and breed) and data predicted by AutoFom-III (Frontmatec Smørum A/S, Herlev, Denmark) (Brøndum, Egebo, Agerskov, & Busk, 1998). Furthermore, it is also an objective of this work to evaluate the economic effect of the accuracy on ham classification method for the meat industry.…”