An investigation into the use of video image analysis (VIA) and visible-near infrared (NIR) spectroscopy for carcase evaluation : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Animal Science at Massey University, Palmerston North, New Zealand

In order for the meat industry to move towards a carcase payment system that is more consumer-focused, there is a need to identify carcases that have a higher yield of superior eating quality meat. Through a series of experiments, this thesis investigates the relationships between video image analysis (VIA) variables and saleable meat yield (SMY%) of high-value cuts in beef carcases, and also the relationships between visible-near infrared (NIR) spectra and instrumental meat quality parameters in beef, lamb and venison of various breeds and genders. Results showed that VIA could effectively replace the visual classifier for classifying beef carcases according to the EUROP carcase classification system, and that both visual and VIA systems showed some promise for predicting the yield of high-value sirloin yield through the EUROP-grid information. Both VIA and visual systems could only account for approximately 57% of the variation in sirloin SMY%, but the relationship between SMY% and other possible VIA outputs such as lengths, widths and volumes remains largely uncharacterized. Instrumental measures of meat quality (shear force, pH and colour) of M. longissimus thoracis et lumborum (LTL) from 234 beef carcases and 208 Texel lambs showed that gender had a larger effect on meat quality than breed. Data from these two experiments was used to determine the relationship between NIR spectra and instrumental meat quality parameters in beef and lamb LTL. NIR showed promise for identifying beef with high ultimate pH values and lamb with high intramuscular fat percentages, but the prediction of shear force using NIR spectra in both beef and lamb was less accurate. The effects on meat quality of sex, breed, chilled aging and location within venison M. Longissimus lumborum, for samples from 79 farmed deer showed that all factors influenced venison meat quality, with aging time and gender having the largest effects. The relationships between NIR spectra and venison meat quality indicated that NIR spectra could be used to identify samples with high ultimate pH and high shear force values