Image analysis and application systems in quality evaluation and prediction for meat and live meat animals

摘要

Image analysis approaches, both in frequency domain and in intensity domain, were used to extract tissue features for the evaluation and prediction of intramuscular fat (marbling) for live meat animals and carcasses. The validation testing of these models showed very good prediction accuracy. The root mean square errors are 1.10 to 1.14 for live animals and .93 to 1.12 for carcasses on 64 to 70 percent of the population. About 83 to 90 percent of the predicted residual errors were less than 1.5%. A software package-Quality Evaluation and Prediction-and on-line application systems are developed to transfer the research results into practical on-line applications. The results indicate that the image analysis technologies, the QEP software package, and the on-line application systems are valuable tools in meat quality evaluation and prediction for livestock producers and meat industries.