Stone walls are structures present in the landscape of Denmark and are protected not only for their cultural and historical significance, but also for their vital role in supporting local biodiversity. A considerable number of stone walls structures have either disappeared, suffered substantial damage, or had segments removed. Additionally, as it stands today, the registry of these structures, managed by each municipality respectively, is not up-to-date and lacks completeness. Recent developments in Machine Learning and CNN’s, and the increasing availability of LiDAR-based (Light Detection and Ranging) terrain models have enabled new methodologies in the extraction and mapping of terrain features and structures. With this in mind, our study aims to analyse the publicly available terrain data derived from the Danish LiDAR data (40 cm resolution), using a U-Net-like CNN model, in order to assess the stone walls dataset, and provide for an update of the registry. The study was focused on the Danish municipality of Ærø. Good results were seen using the Digital Terrain Model (DTM) alone, however better results were obtained when adding Height Above Terrain (HAT) and an additional DTM layer with a Sobel filter applied. Using a pixel-wise evaluation, there was an overall agreement of 93% between ground truth and prediction of stone walls in a validation area, and 88% overall agreement for the whole predicted area. Good generalizability was found when externally validating the model on new data, showing positive results for either the existent stone walls, as well as predicting new potential ones, upon visualization. The method performed best in open areas, however positive results were also seen in forested areas, although denser areas and urban areas presented as challenging. Given the inexistence of a reference dataset or other studies on this specific matter, the evaluation of our study was heavily based on the stone walls registry itself, and visual inspection of the predictions and on the ground. Further improvements can come from the inclusion of aerial imagery and other relevant data, as well as further optimization of the CNN model. This application demonstrates the potential of automatization the process of identification and update of the stone walls’ registry in Denmark, of great relevance to the local governments. We suggest that a Decision Support System be developed to allow municipalities access to the results of this method.