A Spatiotemporal Atmospheric Refraction Correction Method for Improving the Geolocation Accuracy of High-Resolution Remote Sensing Images

Atmospheric refraction is one of the most significant factors that affect geolocation accuracy high-resolution remote sensing images. However, current atmospheric correction methods based on empirical data neglect spatiotemporal variation pressure, temperature, and humidity atmosphere, inevitably resulting in poor geometric positioning accuracy. Therefore, terms problems mentioned above, this study proposed a method (SARCM) global measured to avoid uncertainty traditional models. Initially, atmosphere was stratified into 42 layers according their pressure property, each layer divided 1,042,560 grid cells with intervals 0.25 longitude latitude. Then, refractive index imaging region accurately calculated using high-precision Ciddor formula, result interpolated three splines. Subsequently, rigorous model, line-of-sight pixel viewing zenith angle outside WGS84 were derived provide input for correction. Finally, coordinates ground control points corrected Snell’s law. The experimental results showed SARCM could effectively improve image large angle, especially, improvement percentage 34.2426° x-direction 99.5%. Moreover, better than primary state-of-the-art methods.