Determining the position of an object in outdoor environments is an essential requirement for autonomous systems such as unmanned aerial vehicles (UAVs) and mobile robots. In this context, this study introduces a simple yet effective method for estimating the distance and Global Positioning System (GPS) coordinates of an object using only monocular camera images as input. The core of the proposed method lies in combining the YOLOv8 object detection model with the pinhole camera geometry and a reliable intrinsic calibration process, enabling accurate mapping between image pixels and real‐world positions in 3D space. Based on the estimated distance information, the system infers the global position of the object by transforming to the ENU coordinate system and interpolating the GPS coordinates using the camera’s position and viewing direction. The method is designed for platforms with limited computational resources, allowing deployment on embedded devices without the need for dedicated depth sensors. Experimental results show that the average distance estimation error is 2.02 m, corresponding to a percentage error of 5.87%, while the average GPS localization error reaches 2.60 m within a range of 4.5–40 m. With stable performance and a lightweight computational structure, the proposed method demonstrates strong potential for practical applications in low‐cost intelligent localization tasks.
Han et al. (Thu,) studied this question.