Studies On Remote Sensing Image Optimal Object Construction Classification Algorithm Based On Jeffries-Matusita Distance And Its Application

In the past 40 years,with the rapid development of China's economy,the area of coastal cities has been expanding continuously,the construction of ports has been intensified,and the coastline has been continuously extending to the sea due to the reclamation of land from the sea,which has caused changes in the coastline and tidal flat wetland.This paper designed the optimal image object construction algorithm and analyzed the test results of the algorithm.Based on the optimal image object construction algorithm,land use change information from 1987 to 2017 in the surrounding areas of Jiaozhou Bay and Laizhou Bay was extracted from Landsat series images,and analyzed the land use change situation of the surrounding areas of Jiaozhou Bay and Laizhou Bay from 1987 to 2017,and analyzed the change characteristics and rules of tidal flat wetland of Jiaozhou Bay and coastline of Laizhou Bay from the perspective of land use change.Main research contents :(1)The optimal image object construction algorithm corresponding to ground objects in remote sensing images was designed,and the construction effect of the algorithm was analyzed,so as to be used for subsequent classification of remote sensing images.(2)Analyzed the land use change characteristics and land use degree in the surrounding areas of Jiaozhou Bay during1987-2017,and based on land use change over the years,analyzed the change characteristics of tidal flat wetland.(3)Analyzed the land use change characteristics and land use degree in the surrounding areas of Laizhou Bay during 1987-2017,and based on land use change over the years,analyzed the change characteristics of coastline.Main conclusions :(1)The optimal image object construction algorithm proposed in this paper can construct the optimal image object for the vast majority of remote sensing images,and it greatly improved the accuracy of automatic land use classification.(2)From 1987 to 2017,the land use change around Jiaozhou Bay was dominated by the increase of construction land and the decrease of cultivated land.The degree of land use had gotten stronger year by year.The tidal flat wetland of Jiaozhou Bay had decreased year by year,and its upper boundary had advanced towards the interior of Jiaozhou Bay year by year.Reclamation was the main reason for the change of tidal flat wetland in Jiaozhou Bay from 1987 to 2017.ConstructionDistance and Its application land,grassland,bare land and inland water bodies were the main land use types after reclamation.(3)From 1987 to 2017,the land use change around Laizhou Bay was dominated by the increase of aquaculture land,salt field land and construction land.The degree of land use had gotten stronger year by year.From 1987 to 2017,the coastline of Laizhou Bay had both advanced to the sea and moved back to the sea.Reclamation had changed the coastline;Aquaculture land,bare land and construction land were the main land use types that had affected the length of coastline.Innovation points :(1)The optimal image object construction algorithm was designed and it greatly improved the accuracy of automatic land use classification.(2)From the perspective of land use change,the characteristics of tidal flat wetland and coastline in two typical bays were analyzed.