<p>Model-based methods utilizing finite mixtures of probability distributions are extensively employed for classifying observations. These techniques determine the probability of assigning each observation to a specific group. One notable application of these methods is image segmentation, where various elements of an image are separated and categorized, facilitating object identification within the image. Typically, machine learning algorithms are used for this purpose. In this paper, we present finite mixtures of skewed distributions and evaluate the accuracy of these classification methods in image segmentation, specifically comparing them to traditional model-based classification methods and machine learning algorithms.</p>