In this work, we propose to use deep learning to segment an image based on its color and content. We start by reviewing previously developed content-color-dependent screening (CCDS) presented in [1] [2]. The goal of CCDS is to apply different color assignments for the two or more regular or irregular halftones within the image depending on the local color and content of the image. If the image content locally contains high variance of color and texture, the artifacts due to halftoning will not be as visible as the artifacts in smooth areas of the image [1]. Therefore, the objective of CCDS was to detect the smooth areas of the image and apply the best possible color assignments in those areas. In order to detect smooth areas, the image segmentation algorithm involving the retrieval of the cluster-map and the segmented edge-map was proposed. The main disadvantage of the current approach is that for any given image, the result is highly dependent on the initial parameters, such as the number of clusters, low and high thresholds for edge detection, bilateral filter parameters and others. In this work, we propose to use the wellknown U-net architecture to detect the smooth areas of the image, and then apply the well-known K-means algorithm to cluster the image based on color. The U-net is a type of a convolutional neural network (CNN) designed for quick, precise image segmentation, and it is used to predict a label for every single pixel [3]. The architecture of the U-net is suitable for this work because it consists of a contracting path to capture context and a symmetric expanding path that enables precise localization [3]. We believe that using the U-net to detect smooth areas of the image greatly improves the current approach and provides better results.