Airborne Full-waveform LiDAR Data Processing And Forest LAI Estimation Study

LiDAR(Light Detection and Ranging)is an active remote sensing technology that can accurately obtain the three-dimensional information of the target.With the development of laser radar system and related technology,more and more airborne LiDAR system has the ability to obtain full waveform data,and further promote the application of laser radar technology in the industry.The full-waveform laser radar system returns the waveform information by digitally sampling the return pulse signal.Therefore,compared with the system that only provides the data of the discrete echo point cloud,the data analysis and processing of the whole waveform laser radar is more controllable and can be Of the industry application requirements for the whole waveform data analysis and processing to obtain more information on the forest structure parameter estimation and other research is very important.The full-waveform file uses the generic LAS format,which is mainly for airborne LiDAR data,which facilitates the storage and exchange of data between different hard and software.It is generally recognized by users and is widely used.LAS 1.3 format.As an important parameter of forest canopy structure,forest leaf area index plays an important role in simulating forest growth,forest and atmospheric energy exchange and fire prediction.This paper firstly introduces the research status of airborne full-wave LiDAR data,and then combines the characteristics of full-waveform data and the necessity of its analysis and processing.Taking birch forest as the research object,through the analysis and analysis of the whole waveform data,Based on the calculation of the traditional laser penetrating index(LPI),the corresponding full-waveform laser penetration index(LPIfi)is calculated by using the complete return pulse energy information of each waveform in combination with the characteristics of full waveform data.(LPIf-mean),which is used to invert the LAI(Leaf Area Index,LAI)of the forest forest,and the other is the mean value of the laser penetrating index The inversion model was validated by the measured LAI data.The results show:(1)Based on the characteristics of full waveform data and the analytical process of LAS1.3 file and the analysis and processing methods,it is possible to obtain accurate information such as amplitude,position and wave width and energy(Gaussian distribution curve and time axis Area of information).And according to the needs of follow-up work,the study area of the Leica ALS60 LAS 1.3 format full waveform file was processed.The results show that there are about 82,000 full-waveform data of 80 square samples with a length of 10 m,of which about 99.93%can be well treated,and the rest is processed according to the local maximum method.In this paper,the Gaussian decomposition method used in this paper can extract the characteristic information of the whole waveform data more accurately.(2)Based on the analysis and calculation of the ground return pulse energy and canopy return pulse energy of the whole-wave laser pulse,the ratio of the reflectance ratio of the ground layer and vegetation canopy of the 80 field samples is basically the same Is 1.2,indicating that the energy value of the ground return pulse is slightly larger than that of the forest canopy,which is basically consistent with the actual situation of the forest land.(3)The LI estimation of forest stands can be improved by LPIf-mean,which is extracted by airborne full-wave laser radar data.The prediction accuracy of airborne full-wave laser radar data and forest LAI model established by LPIf-mean.is R2 = 0.864 and RMSE = 0.139,which indicates that the new parameter can improve the inversion accuracy of forest LAI.(R2 = 0.815,RMSE = 0.105)is higher than that of the discrete echo point cloud(R2 = 0.720,RMSE =0.140),and the accuracy of the forest LAI is reversed by the full-wave laser radar.This,in turn,reflects the advantages of airborne full-wave LiDAR data itself,providing high-quality quantitative data sources and new parameters for high-precision quantitative estimation of forest LAI.The article concludes with a summary of the conclusions and the next step.