Information extracted from aerial photographs has found applications in different areas including urban planning, forest management, disaster relief, and climate modeling. In many cases labeling of information in the photo is still performed by human experts, making the process slow, costly, and error-prone. This paper shows how a convolutional neural network can be used to determine the location of GPS markers in aerial photos. A simple linear iterative clustering (SLIC) method is then applied to improve the classification result. Both binary and multi-label CNN architectures are evaluated and analyzed. The results of the classification and segmentation show that CNNs are a viable tool for extracting the locations of objects from aerial photographs.