Body Composition Evaluation

Neural Network Application for Classifying Beef Intramuscular Fat Percentage

  • Nam-Deuk Kim (Iowa State University)
  • Viren R. Amin (Iowa State University)
  • Doyle E. Wilson (Iowa State University)
  • Gene H. Rouse (Iowa State University)


In the previous report, we have presented statistical pattern recognition and classification techniques to preclassify the ultrasonic images into the low- or high- IFAT groups (less than 8% and more than 8%). The classification tree was used in the previous report, and it provided overall classification accuracy of 90% for low- and high- IFAT groups of images. Here, we are presenting artificial neural network (ANN) as a pattern recognition tool to get better classification accuracy. ANNs provide a nonparametric approach for the nonlinear estimation of data. These models are trained to mimic the desired behavior using example data from the actual problem. The ANN model provided classification accuracy of 95% for 653 sample images.

Keywords: ASL R1438

How to Cite:

Kim, N., Amin, V. R., Wilson, D. E. & Rouse, G. H., (1998) “Neural Network Application for Classifying Beef Intramuscular Fat Percentage”, Iowa State University Animal Industry Report 1(1).

Download pdf



Published on
01 Jan 1998
Peer Reviewed