The aim of the present work is to classify pork and turkey hams with the identification of the best colour and textural features extracted from digital images. Image processing techniques were developed to classify and characterize three qualities (high, medium and premium) of pork and turkey hams from their image features. Images from cooked pork and turkey hams were acquired and evaluated to classify them on the basis of their best colour and textural features. Different colour and textural features (a total of 26 colour features and 40 textural features) were extracted and the best features which might have significant effect to classify the three types of hams from each other were obtained from Mahalanobis distances observed between groups and inter-correlations among the features. In such a way, two best colour features (mean of S, std. deviation of b*; from HSV and L*a*b*colour space, respectively) and three textural features (entropy of b*, contrast of H, entropy of R; from L*a*b*, HSV and RGB colour space, respectively) for pork, and three colour features (mean of R, mean of H, std. deviation of a*; from RGB, L*a*b*, and HSV colour space, respectively) and two textural features (contrast of B, contrast of L; from RGB and L*a*b* colour space, respectively ) for turkey hams were selected as features with the highest discriminant power. The Linear Discriminant Analysis (LDA) was used to predict the classification accuracy. High classification performances were reached for both types of hams (>99.5% for pork and >90.5% for turkey) using the best selected features or combinations of them.