基于地面三维激光扫描强度数据的潮滩表层含水量估算

Estimation of Surface Moisture of Tidal Flat based on Intensity Data of Terrestrial Laser Scanner

大面积潮滩表层含水量的测定是潮滩研究中的难题,传统的测量方法难以同时满足高效和精度的要求。地面三维激光扫描技术凭借其高精度、高分辨率以及主动性强等优点,已经高效运用在潮滩地形研究中。但是,对潮滩含水量进行有效分析,仅利用点云的空间几何信息是不够的,还需要对点云的强度数据进行挖掘。地面激光扫描仪提供了包含目标表面光谱反射特性的点云强度数据,利用强度数据可以有效地进行目标表面特性提取。本文提出了一种新的长距离地面激光扫描仪强度数据改正方法,对入射角和距离效应进行有效改正。利用Riegl VZ-4000长距离地面激光扫描仪建立室内含水量模型并对上海市崇明岛一处潮滩进行测试分析,同时收集26个潮滩沉积物样品并利用传统干湿称重法进行含水量验证计算。结果表明:相比于传统技术,利用改正后的激光强度值估算大面积潮滩沉积物表层含水量是一种精确和高效的方法。改正后的激光强度值与潮滩表层含水量存在幂函数关系,相关系数为0.961,估算精度为91.94%。

Estimation of surface soil moisture and distributions play a key role in the ecological, environmental,and topographical investigations for intertidal mudflats, which are characteristically wet and periodically submerged by sea water. Unfortunately, existing techniques of large-scale surface moisture monitoring have measurement limitations, and the efficiency and accuracy of traditional methods are difficult to be ensured simultaneously. Thanks to the advantages of high precision, high resolution, and strong flexibility, Terrestrial Laser Scanners(TLSs) have been widely used to study the geomorphological features of intertidal mudflats.However, merely geometrical information is not enough to derive the physical characteristics of a mudflat, it is also necessary to mine intensity data of points cloud. Aside from the 3D points cloud, TLS can record the intensity value of each point, which contains spectral characteristics of the scanned target and is of vital importance to the improvement of TLS data classification and feature extraction. Most TLSs emit near-infrared lasers that can be strongly absorbed by water. Thus, the intensity values of areas with high water moisture are theoretically smaller than those of the regions with low moisture. In this study, the intensity data of TLSs were corrected for the incidence angle and distance effects, and the corrected intensity data were utilized to quantitatively estimate the surface soil moisture of intertidal mudflats. The Riegl VZ-4000 long-distance terrestrial laser scanner with a near-infrared wavelength of 1550 nm was used to establish the indoor moisture model and to conduct a case study of a tidal flat in the Chongming Island, Shanghai, China. Additionally, 26 sediment samples were also collected by the traditional way of gravimetric measurements to get moisture. Results show that there was a power function relationship between the corrected intensity data and surface moisture, with the correction coefficient being 0.961 and the estimation accuracy reaching 91.94%. Our findings show that,compared with the traditional technique(i.e., gravimetric measurements), the corrected intensity data of longrange TLSs are effective for quick, accurate, and detailed estimation of surface soil moisture over large areas.