Signal simplification and cloud detection with an improved Douglas-Peucker algorithm for single-channel lidar

Abstract

Single-channel lidar is a widely used system in atmospheric aerosol and cloud detection. However, many difficulties remain in the automatic and accurate identification of cloud from backscatter signals. Popular methods have been proposed, but there is still large uncertainty in cloud detection, especially when the signal-to-noise ratio is low. In this study, a signal simplification approach based on an improved Douglas-Peucker algorithm is proposed. The layer base and top are then selected using the simplified signal and the raw range-corrected signal. Finally, we use a peak-to-base ratio function to distinguish a cloud layer from a non-cloud layer. The detection results of our algorithm are remarkably better than those obtained using the differential zero-crossing method.