Font Size: a A A

Remote Sensing Recognition Of Urban Black And Odorous Water Bodies Based On GF-2 Images

Posted on:2019-03-20Degree:MasterType:Thesis
Country:ChinaCandidate:S WenFull Text:PDF
GTID:2371330548496198Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
The so-called "black and odorous" is an extreme phenomenon of organic pollution of water body,which is caused by the water body's anoxia and organic matter corruption.It has become a common environmental pollution problem in many large and medium-sized cities in our country.At present,the researches on urban black and odorous water bodies mostly focus on the formation mechanism,the methods and indexes of water quality evaluation and ecological restoration,but the research on its recognition algorithm is still lacking.This study focused on the black and odorous water bodies in the main city of Nanjing.Based on the field measured water body data and image data,the reflectance spectrum characteristics of urban black and odorous water bodies were analyzed.According to the setting of GF-2 sensor band,the recognition algorithm based on the characteristic band and the saturation and chromaticity value were constructed respectively.The recognition accuracy,stability and applicability of different algorithms were evaluated,and selected the optimal recognition algorithm for GF-2 images.Applying this algorithm to GF-2 satellite images of 2015,2016 and 2017,the spatial distribution of black and odorous water bodies in different phases of Nanjing City was obtained.Then the spatial and temporal distribution characteristics of urban black and odorous water bodies were analyzed.The main conclusions of this study are as follows:(1)Reflectance spectra characteristics of urban black and odorous water bodiesCompared with urban normal water bodies,the urban black and odorous water bodies had the lowest remote sensing reflectance,which was lower than 0.025sr-1;and raised slowly in the range of 400nm-500nm with the lowest slope.The spectrum of black and odorous water bodies appeared low peak at the range of 550nm?580nm and had no obvious absorption valley and reflection peak at 620nm and 700nm.(2)Remote sensing recognition algorithm for urban black and odorous water bodiesBased on the analysis of the spectral characteristics of urban black and odorous water bodies,and the band setting of GF-2 image,the recognition algorithms were proposed.These algorithms were single-band threshold based on reflectance of green band,DBWI(Difference of Black-odorous Water Index)based on difference of blue band and green band,SB WI(Slope of Black-odorous Water Index)based on the product of slope of the spectrum between blue band,green band and red band,NDBWI(Normalized Difference Black-odorous Water Index)based on ratio of green band and red band,and chromaticity and saturation value.According to the calculation of modeling samples,the discriminant threshold of each algorithm for urban black and odorous water bodies and normal water bodies were determined.Using actual ground samples and satellite synchronous sampling points to evaluate the accuracy of each algorithm.Among them,the NDBWI was suitable for the recognition of urban black and odorous water bodies and normal water bodies and could be used as the optimal recognition algorithm in this study.(3)Spatial and temporal distribution characteristics of black and odorous water bodies in NanjingBased on the recognition results of GF-2 remote sensing images on April 21,2015,November 3,2016 and October 9,2017,11 black and odorous rivers were identified in Nanjing in 2015,with length and area of 19.19 Km and 0.313Km2 respectively.Nine black and odorous rivers were identified in 2016,with length and area of 14.79Km and 0.244Km2 respectively.Four black and odorous rivers were identified in 2017 with length and area of 2.12Km and 0.088 Km2 respectively.From the perspective of spatial distribution,the black and odorous water bodies in urban areas were mainly distributed in central urban areas(Gulou District,Jianye District,Xuanwu District and Qinhuai District).Among them,the black-odorous phenomenon in Gulou District and Jianye District was the most serious.The densely populated areas,the construction sites,the factories and broken nose were the high-incidence areas of black-odorous phenomenon.Moreover,the location of the black and odorous rivers was not fixed,which was easily affected by the surrounding environment.From 2015 to 2017,the quantity,length and area of black and odorous water bodies in Nanjing showed a downward trend.The number of black and odorous rivers was the largest in 2015,and the black-odorous phenomenon was the most obvious in the rivers in Nanjing.After one year's management,the black-odorous phenomenon still existed in many river sections in Nanjing Municipality in 2016.In 2017,the overall water quality has been greatly improved.The black-odorous phenomenon in rivers in Qinhuai District and Jianye District has basically disappeared,and the water quality in many river sections has not been deteriorated.
Keywords/Search Tags:Urban black and odorous water bodies, Nanjing, Spectral characteristics, Recognition algorithm, GF-2 image
PDF Full Text Request
Related items