Extracting 3D Information Of Urban Features Based On Laser Altimetry Data

Urban areas are gathering more and more people.How to quickly obtain urban construction and ecological environment information is very important for the future development of the city.In the research fields of population and resource distribution,urban expansion trend and the impact of carbon cycle,the acquisition of urban related information is also of great value.The Geoscience Laser Altimetry System(GLAS)equipped with Ice Cloud and Land Elevation Satellite(ICESat)was launched in January 2003.GLAS is an active spaceborne lidar altimeter system,which has been running intermittently for seven years in orbit.It is the first lidar system in the world for global observation(Li DAR).In this paper,the GLAS data corresponding to urban areas and the actual measurement data are compared and analyzed.The relationship between the echo signal and the corresponding ground object environment is mainly analyzed.The wavelet denoising algorithm under the parameter optimization in the smoothing filtering of the full waveform data and the effective peak correction algorithm in the decomposition of the full waveform data are proposed.It not only enriches the research content of full waveform data,but also broadens the research field of full waveform data.The main research contents and conclusions of this paper are as follows:1.The related contents of full waveform data processing of spaceborne lidar are combed and introduced from four aspects,including the general situation of ICESat /GLAS data,the preprocessing of full waveform data,the decomposition method of waveform,and the extraction of some characteristic parameters of waveform.The preprocessing of full waveform data mainly includes five steps: voltage conversion,waveform regularization,data decompression,background noise estimation and smoothing filtering.The waveform decomposition methods are mainly Gaussian decomposition,wavelet decomposition,waveform inversion and waveform deconvolution.2.Aiming at the problem of parameter selection of wavelet de-noising for full waveform data,the wavelet de-noising algorithm based on parameter optimization is used to study full waveform data,and good filtering effect of original echo data is obtained.For the wavelet denoising algorithm under parameter optimization,firstly,the control parameter types of all inputs are set according to the waveform characteristics.Then,the echo waveform is reconstructed by referring to the wavelet formula and the one-dimensional denoising function.Then,the distribution of better control parameters is calculated under the preset conditions.Finally,the optimal control parameter combination is extracted according to the SNR.3.Aiming at the error and accuracy problems in the full waveform data processing,a full waveform data processing method based on the effective peak correction algorithm is proposed,which improves the measurement accuracy of the corresponding ground objects.For the effective peak correction algorithm,firstly,the data segmentation method is used to estimate the mean and variance of noise in the process of data preprocessing.Then,the corresponding filtering width is used for Gaussian filtering in different signal segments.Then,the effective peak information is extracted by combining the noise threshold and the filtered waveform.Finally,the Gaussian component parameters are corrected according to the peak information.4.In the application of the above algorithm in the experimental analysis,the filtering effect of each parameter combination,the filtering effect of each filtering method and the height measurement results of all methods are compared and analyzed.The research shows that the effective peak correction algorithm can accurately complete the height measurement of ground objects.Compared with the traditional denoising algorithm,the improved wavelet denoising algorithm has better filtering effect and can also be applied in the height measurement of ground objects.