Font Size: a A A

Research On Key Technologies And System Implementation Of Woodland Change Survey

Posted on:2020-03-12Degree:MasterType:Thesis
Country:ChinaCandidate:K X ZhangFull Text:PDF
GTID:2393330626965511Subject:Cartography and Geographic Information System
Abstract/Summary:
Since the 18 th National Congress of the Communist Party of China,in order to strengthen the construction of ecological civilization,starting from 2017,China requires that forest land change surveys and forest land map databases be updated every county in order to effectively monitor and investigate the occurrence of illegal occupation of forest land.Forest land change survey requirements are based on Internet + high-scoring remote sensing,satellite positioning,and GIS technology to improve the efficiency,accuracy and reliability of forest land change surveys.The main causes of forest land change include: afforestation,logging,forest fires,forest pests,and forest land acquisition.The actual operation process of the forest land change investigation includes three major aspects: the use of two periods of high-score remote sensing images to artificially interpret the suspected spots on the forest land,sending people to the site to verify the suspected spots,and updating the database of a forest land map based on the verification results.According to the survey of the actual work of the forest land change for three consecutive years nationwide,problems such as high labor intensity,tight work tasks,and need to improve reliability and accuracy have been exposed.How to use the Internet + highscoring remote sensing,satellite positioning,and GIS software development to improve the efficiency,accuracy,and reliability of forest land change investigations has become a hot topic for forestry science and technology workers.This article focuses on the key technical issues involved in forest land change investigations,and completes the following aspects of research and development:(1)Based on the two-phase high-scoring remote sensing images,an algorithm capable of automatically extracting suspected spots that reflect the dynamic changes of forest land is researched and implemented in software.Comparing the suspected speckles automatically extracted by the algorithm and the artificially interpreted suspected speckles,the accuracy can reach about 80%,which can greatly reduce the workload of indoor manual interpretation.(2)Propose the operation mode of forest land change investigation based on the Internet + integration of internal and external industries,participate in the development of an online version of the forest land change investigation software platform,and take the province as the unit to investigate suspected patches and two high-scoring remote sensing images required for forest land change investigation Data such as forest land planning and utilization are placed on the central database server of each province,and they are pushed to users in cities and counties throughout the province by using the network.Realize mission distribution through the network,ensure uniformity and standards of survey data across the province,progress management of survey missions,and online retrieval of survey results.(3)Participate in the development of mobile survey software.Use a tablet computer to log in to the server with a user name and password,and download the survey data pushed by the provincial monitoring center.Adopt satellite dynamic navigation and positioning to quickly reach the spot of suspected spots and conduct verification and investigation of forest land dynamic changes,including the collection of graphics and attribute data,photographs and evidence,trajectory records,survey data real-time return and storage.The software can greatly improve the accuracy and efficiency of field inspections and ensure the reliability and accuracy of survey data.Using the research and development results of this article,networked forest land change survey experiments were conducted in Sichuan Province in 2017 and 2018.The results show that the image detection algorithm provided in this article can reduce the indoor manual interpretation workload by about 60%,using both internal and external It also integrates software platform and digital operation mode,which can effectively improve the efficiency,accuracy and reliability of forest land change investigation.The research and development results of this paper have important reference value for the investigation of normalized forest land change and the update of a map database of forest land in China.
Keywords/Search Tags:Woodland change investigation, suspected spot, change detection, NDVI
Related items