Thermographic images for predicting the weight of lamb cuts

Context Accurate prediction of carcass cut weights is essential to ensure standardization and add value to lamb production. Traditional methods of carcass evaluation are often subjective, time-consuming and prone to errors. Aims This study aimed to develop predictive models for the weights of lamb primal cuts using morphometric measurements obtained from surface thermographic imaging (STI) combined with hot carcass weight (HCW). Methods Seventy-two lamb carcasses (8-month-old males) were evaluated. STI images were captured from each carcass and analyzed using ImageJ software to extract morphometric measurements: external carcass length, thoracic depth, leg length and leg width. Carcasses were fully deboned to obtain the weights of the shoulder, flank steak, 13-rib rack, rump cap and topside. Multiple linear regression models were developed using STI variables and HCW. Model performance was assessed through leave-one-out cross-validation with R2, mean absolute error (MAE), root mean square error of cross-validation (RMSECV), mean squared error (MSE), bias and slope. Key results The topside weight model showed the highest accuracy (adjusted R2 = 0.81; MAE = 0.15; RMSECV = 0.195; MSE = 0.038; bias = 0.000; slope = 0.809) using HCW, thoracic depth, external carcass length, and leg width. The shoulder and 13-rib rack weight models (adjusted R2 = 0.69; MAE = 0.120; RMSECV = 0.188; MSE = 0.035; bias = −0.000; slope = 0.678; and adjusted R2 = 0.41; MAE = 0.071; RMSECV = 0.146; MSE = 0.019; bias = −0.000; slope = 0.411, respectively) were predicted using HCW alone. The flank steak and rump cap weight models (adjusted R2 = 0.58; MAE = 0.073; RMSECV = 0.116; MSE = 0.013; bias = −0.000; slope = 0.56; and adjusted R2 = 0.60; MAE = 0.063; RMSECV = 0.086; MSE = 0.007; bias = −0.000; slope = 0.583, respectively) included HCW and leg length. Conclusions Lamb cut weights can be predicted accurately and efficiently using STI-derived morphometric measurements combined with HCW. Implications These findings have significant implications for the meat industry by providing an objective and accurate tool to improve carcass classification, enable fair payment systems and support decision-making in meat processing operations.