Beef quality parameters estimation using ultrasound and color images

Area of Science:

• Animal Science
• Food Science
• Computer Vision

Background:

• Beef quality assessment is economically significant and complex.
• Automated estimation of quality parameters in live or post-mortem cattle is highly desired.
• Key quality indicators include rib eye area, intramuscular fat, and backfat thickness.

Purpose of the Study:

• To develop and evaluate algorithms for automatic beef quality estimation.
• To measure rib eye area, intramuscular fat percentage, and backfat thickness using imaging techniques.
• To integrate in vivo (ultrasound) and post mortem (color) image analysis for comprehensive quality assessment.

Main Methods:

• Curve evolution algorithm for rib eye area calculation.
• Support Vector Regression (SVR) model for intramuscular fat percentage estimation.
• Feature extraction from regions of interest in ultrasound and color images.

Main Results:

• Accurate calculation of rib eye area and backfat thickness.
• Promising results in predicting intramuscular fat percentage.
• Validation using expert-acquired images, commercial software estimations, and chemical analysis.

Conclusions:

• The developed algorithms effectively measure rib eye area and backfat thickness.
• The methods show potential for accurate prediction of intramuscular fat.
• Automated imaging analysis offers a viable approach for beef quality assessment.