An important, and not yet solved problem in meat industry is the issue of estimating the intramuscular fat level content in the carcass. Solution of the problem of identification of quantity of the intramuscular fat, on the basis of Information in ultrasound images taken on lamb's carcasses or even living animal, is ofessential utilitarian importance. The amount of intramuscular fat (known as marbling) has significant impact on market value and meat's culinary usefulness. Previoitsly used methods for marbling classification in carcasses based on an analysis of animal 's age, weight and gender, or had invasive nature. These methods were estimated as unreliable and inefficient. There have been noticed growing explorers' interest in drawing conclusions based on information of data coded in a graphic form. The neuronal identification of pictorial data, with special emphasis on both quantitative and qualitative analysis, is more frequently utilized to gain and deepen the empirical data knowledge. Extraction and then classification of selected picture features, such as color or surface structure, enables one to create computer tools in order to identify these objects presented as, for example, digital pictures. Thispaper presents an attempt to create noninvasive method to classify marbling, based on ultrasound images, computer image analysis and artificial neural networks.