| Coastline is an important reference for the study of sea area and land change.Therefore,automatic extraction and change analysis of coastline have important research significance.With the rapid economic development of offshore and land areas,the area of coastal aquaculture,saltpans,ports,seawalls and salt marshes is increasing.Effective technology desired to extract coastline features from remote sensing images to assist monitoring and management of the vast along the coastline.Because different coasts(such as silty,sandy,bedrock,biological coasts and artificial coastlines)have different attributes and characteristics,and the types and extent of coastlines are constantly changing,automatic extraction and update of coastlines are instantly needed to solve the problem.The currently work takes the coastline of Qingdao,Shandong as an example.The main research content includes the following two aspects:(1)Automatic extraction of Qingdao coastline.Focus on the characteristics,environment and dynamic changes along the coastline to accurately describe and regularly monitor the Qingdao coastline in eastern China.In the current research collected ETM+ and OLI data on the continental coastline in 2000,2010 and 2019,and used RS and GIS technology to explore the characteristics and temporal and spatial variation of the shoreline in the past 19 years.For this constantly changing coastline,it is a major challenge to understand the changes of the coastline through appropriate technology.In the current research used three statistical methods to measure coastline changes: Linear Regression rate,End Point rate,and Net shoreline movement.Among them,the maximum speed reached by the End Point rate,Net shoreline movement and Linear Regression rate were 266.07m/yr,2391.85 m and 124.47m/yr,respectively.The maximum accumulation of End Point rate,Net shoreline movement,and Linear Regression rate are-142.55m/yr,-1234.59 m and-63.22m/yr,respectively.For automatic coastline extraction,use different techniques,but the most suitable method is the canny edge algorithm method,which can generate results that are more accurate.Then compared the accuracy of this model with other generic edge detection algorithms(such as Prewitt,Sobel,Robert edge detection methods,thresholds edge detection method,and region growth edge detection method),and found that our model is superior in accurately identifying coastlines.(2)Analysis of changes in Qingdao coastline.The results show that in the past two decades,the length of Qingdao’s coastline has continued to increase,with a total increase of 18.14 km.However,different natural and artificial coastlines have undergone major changes.The deterioration of the coastline caused by man-made has gradually increased from 53.63% in2000 to 68.40% in 2019,while the length of the natural coastline has been greatly reduced.The Qingdao shoreline has geomorphological abundances of bedrocks,salt fields,and seawall,as well as aquaculture,sandy beaches,muddy,ports,buildings,and estuarine features.Higher elevation and angle bedrocks are frequently the source of sediment and primary contributors to sediment deposition.The areas where the coastline has undergone significant changes are concentrated in the Jiaozhou Bay area,and major changes have taken place in the west and east of the Qingdao coast.Due to the combined influence of natural and human factors,the coastline has greatly expanded the extent to the seaside.The main factor of coastal evolution is the construction of coastal engineering.The driving force that triggers coastline changes shows obvious temporal heterogeneity.Therefore,this study introduces the monitoring process of coastal areas,and uses geospatial technology to analyze the changes in the coastline,which will be very useful for the planning and management of coastal areas in Qingdao coastal areas. |
