Preset position deviation correction method for pan-tilt cameras in substation intelligent inspection systems

With the widespread deployment of the “unattended and centrally monitored” operational mode in substations, the preset positions of pan-tilt cameras have become increasingly crucial for intelligent inspection tasks. However, due to mechanical wear, environmental disturbances, and imaging-related coupling errors, preset positions are prone to cumulative deviations, resulting in field-of-view misalignment, degraded recognition accuracy, and reduced performance of image–model collaborative functions. To address these issues, this paper proposes an intelligent preset-position correction method tailored for substation scenarios. The proposed approach utilizes the positional variations of targets between the reference image and the deviated image, constructing a line-of-sight vector analysis based on the pinhole imaging model to estimate the horizontal and vertical deviation angles of the camera. A step-angle tuning strategy and a dynamic compensation mechanism for cumulative errors are further designed to enable precise calculation of the deviation and adaptive adjustment of pan-tilt control parameters. Field experiments conducted in real substations validate the feasibility of the proposed algorithm. The results demonstrate that the proposed intelligent correction method exhibits strong robustness, engineering applicability, and scalability, providing an efficient and practical solution for automatic preset-position verification in substation intelligent inspection systems.