会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 2. 发明专利
    • IDENTIFICATION OF FOREST FIRE SUSCEPTIBILITY USING GIS AND NEURAL NETWORK APPROACHES
    • AU2021106836A4
    • 2021-11-18
    • AU2021106836
    • 2021-08-24
    • GUDIKANDHULA NARASIMHA RAOPEDDADA JAGADEESWARA RAODUVVURU RAJESHKONDAPALLI BEULAHKODAMALA PRATHYUSHABANGARU BALAKRISHNA
    • GUDIKANDHULA NARASIMHA RAOPEDDADA JAGADEESWARA RAODUVVURU RAJESHKONDAPALLI BEULAHKODAMALA PRATHYUSHABANGARU BALAKRISHNA
    • G06N5/04G06N3/04G06N20/20G08B17/00
    • A system for identification of forest fire susceptibility using GIS and neural network approaches is implemented based on a spatial prediction of vulnerable zones of forest fires. The fire behaviours are modelled by texture analysis using computer vision systems. The Central Server receives fire-affected regions from the volunteer's smartphone. The geotagged photos would provide fired location coordinates, and the photo can be rotated and used in different angles to locate exact fire locations based on Google Earth API. Object detection algorithms compute the position vector of a moving object. Antennas or Satellite systems will grasp the information from the fire regions then the data will be analyzed in GIS for use in the rescue operation and sends SMS alerts to local people about forest fires and the same will be disseminated to the APSDRF / NDRF team. The proposed invention builds up a smarter system that is based on semantic neural networking to focused on the burned areas. SELECT STUDY AREA Data collection (Topographic data, Supporting Infrastructures, Vegeation cover data, Remote sensing date, Climatic data, -Z ~fo Demographic data, Forest fire events) Derived Gire-related factors In a GI S Burntarea map (Slope, Aspect, Road density, Viewsheds from fire S utasin rap watchtowers, Land cover, Vegetation index - NDVI, ANN WEIGHTS sngGSPrecipitation, Population density) ASSIGNMENT I L E ThematicI oaesgnration and datafromG;Srelated categories inASCII formal Bumnt areasr Categodies Select tralning, layer In raste layers In raster i testing and format forrnat validation subsets network architecture PROBABILISTIC RATINGS usingwMATLAB software PROCEDURE weightsGeneration rating Chang thematic layers (r) weights N RMSE goal using GIs increase ANN size y | Run test and i Forest Fire validation subsets Data Susceptibility Index Integration N RSgolusing G IS FFSI =Irtwt Met? Roselect seotbs triU n SEMANTIC WEB Collect Weight Interpretation more data according to Guha et al. (2005) ----- methodology FOREST-FIRE SUSCEPTIBILITY INDEX F~inalisatIon of weights for thematic layers (wh )