This is the authors' abstract. We don't add key points for this paper.
On-line pork carcass grading with the Autofom ultrasound system is described. The system consists of 16 ultrasound transducers positioned in a frame. The carcass is measured fully automatically at 3,200 positions to a depth of approximately 12 cm with a depth resolution of .19 mm. The ultrasound data form a three-dimensional ultrasound image, which is processed for noise reduction, orientation detection, and extraction of 127 features describing the carcass composition. The image features are used in a multivariate regression model, that is used for on-line predictions. On-line tests performed at line speeds up to 1,150 carcasses/h provide predictions of the meat percentage with an accuracy of 1.58 to 1.95%. Good predictions of the fat thickness and primal meat cuts have also been made.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Brøndum et al. (1998) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: