In this paper, a photoelectric navigation image super-resolution system based on convolutional neural network is designed to provide reliable navigation information for aircraft flight or landing and satellite images. Super definition visible images can provide reliable information for ground monitors, pilots or drones. The super definition method can reduce the requirements of navigation system on hardware equipment and computing resources, and is suitable for ordinary civil aircraft and small UAV with low precision cameras. In this paper, the convolutional neural network is used as a framework to build an image super-resolution model for near real-time computing.