论文部分内容阅读
为了监测污水排放后扩散和稀释的过程,可在污水里注入染料,利用考察船追踪其羽状扩散痕迹(以下简称痕迹)。最近几年的实践证明,遥感影像是船只调查的有效辅助手段。这是因为遥感影像可实时显示大范围的痕迹,而船只调查,由于潮汐、风和水流变化的缘故,是不能做到这点的。 本文所描述的调查,旨在通过机载多光谱扫描影像和模型化、反演及精度评价等一套方法,对三种染色的下水道污水浓度作出估算与制图。 航空多光谱扫描数据,经过辐射、大气和辐射纠正后,所得的标准化辐射指数与染料浓度密切相关(r=0.85,相关在99％置信度水平上显著)。将这个关系对所有航空多光谱扫描数据进行反演和外推后得到22个痕迹图。作7组分类时,染料浓度的痕迹图,在95％的置信度水平上,达到49-90％的精度;作2组分类时,在95％的置信度水平上,精度提高到68-100％。
In order to monitor the process of diffusion and dilution after sewage discharge, dye may be injected into the sewage and traces of its feathering spread (hereinafter referred to as traces) may be tracked by a survey vessel. In recent years, practice has proved that remote sensing images are an effective aid to vessel surveys. This is because remote sensing images can show a wide range of real-time traces, and vessel surveys can not do this due to tidal, wind and water changes. The investigation described in this paper aims to estimate and map the three sewers of sewages using a set of methods such as airborne multispectral scanning imaging and modeling, inversion and accuracy assessment. Aeronautical multispectral scanning data, after radiation, atmospheric and radiation corrections, resulted in a close correlation between the normalized radiation index and dye concentration (r = 0.85, correlation being significant at 99% confidence level). Twenty-two traces of this relationship were obtained by inversion and extrapolation of all aeronautical multispectral scan data. For the 7-group classification, the trace of the dye concentration was 49-90% at a 95% confidence level; for the 2-group classification the accuracy was increased to 68-100 at the 95% confidence level %.