[{"data":1,"prerenderedAt":200},["ShallowReactive",2],{"content-query-AXuAMYU5YU":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"body":13,"_type":194,"_id":195,"_source":196,"_file":197,"_stem":198,"_extension":199},"/news/en/1576","en",false,"","LuoJiaNET Intelligent Remote Sensing Framework Released in the MindSpore Community","We sincerely invite experts and enthusiasts in the remote sensing field to try this new framework and leave your valuable suggestions.","2022-05-31","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/06/27/b14698fe8230483387043a773d42f2b3.png","news",{"type":14,"children":15,"toc":191},"root",[16,24,30,38,43,48,57,65,70,78,83,88,93,98,106,111,116,121,129,134,145,150,159,164,173,182],{"type":17,"tag":18,"props":19,"children":21},"element","h1",{"id":20},"luojianet-intelligent-remote-sensing-framework-released-in-the-mindspore-community",[22],{"type":23,"value":8},"text",{"type":17,"tag":25,"props":26,"children":27},"p",{},[28],{"type":23,"value":29},"LuoJiaNET, jointly built by the MindSpore team and Wuhan University, was released in the MindSpore community. We sincerely invite experts and enthusiasts in the remote sensing field to try this new framework and leave your valuable suggestions.",{"type":17,"tag":25,"props":31,"children":32},{},[33],{"type":17,"tag":34,"props":35,"children":37},"img",{"alt":7,"src":36},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/06/25/34ac83c1690b423ba9e4132ba75e0d0f.png",[],{"type":17,"tag":25,"props":39,"children":40},{},[41],{"type":23,"value":42},"Although new applications are being continuously developed in the prospering remote sensing industry, the use of general frameworks has become the bottleneck in development because they can only process few types of small-sized images with limited scales and channels.",{"type":17,"tag":25,"props":44,"children":45},{},[46],{"type":23,"value":47},"To break this bottleneck, Wuhan University and the MindSpore team built LuoJiaNET. Powered by Ascend AI technologies, it is the world's first AI framework dedicated to remote sensing image interpretation. The joint team also released LuoJiaSET, the industry's largest remote sensing image dataset. As a dedicated intelligent framework, LuoJiaNET incorporates efficient and accurate sample labeling tools, sample libraries, and corresponding basic models. LuoJiaNET facilitates the development of remote sensing applications and enables intelligent remote sensing technologies to be widely used in the natural resources, maritime, agriculture, forest, and emergency response industries.",{"type":17,"tag":25,"props":49,"children":50},{},[51],{"type":17,"tag":52,"props":53,"children":54},"strong",{},[55],{"type":23,"value":56},"Feature Highlights of LuoJiaNET",{"type":17,"tag":25,"props":58,"children":59},{},[60],{"type":17,"tag":52,"props":61,"children":62},{},[63],{"type":23,"value":64},"1. Full-Stack Deep Learning Architecture",{"type":17,"tag":25,"props":66,"children":67},{},[68],{"type":23,"value":69},"Remote sensing applications need to complete tasks at multiple levels (scenario-, target-, and pixel-levels). Based on this characteristic, the team centers on the computational graph and incorporates the features of remote sensing mechanism models and geoscience knowledge graphs to build the domain specific AI framework LuoJiaNET based on MindSpore.",{"type":17,"tag":25,"props":71,"children":72},{},[73],{"type":17,"tag":52,"props":74,"children":75},{},[76],{"type":23,"value":77},"2. Multi-Dimensional Remote Sensing Characteristics",{"type":17,"tag":25,"props":79,"children":80},{},[81],{"type":23,"value":82},"In addition to the features of efficient memory expansion and adaptive selection of scales and channels, LuoJiaNET is designed with characteristics of remote sensing data in mind, for example, the multi-dimensional spatial and temporal spectrums:",{"type":17,"tag":25,"props":84,"children":85},{},[86],{"type":23,"value":87},"LuoJiaNET uses equivalent decomposition of operators for distributed computing, which allows this network to directly process large-sized images;",{"type":17,"tag":25,"props":89,"children":90},{},[91],{"type":23,"value":92},"The adaptive data channel selection module enables LuoJiaNET to classify hyperspectral remote sensing images directly;",{"type":17,"tag":25,"props":94,"children":95},{},[96],{"type":23,"value":97},"Multiple remote sensing knowledge models, including vegetation, drought, building, and topography indexes, significantly improve the classification performance of LuoJiaNET.",{"type":17,"tag":25,"props":99,"children":100},{},[101],{"type":17,"tag":52,"props":102,"children":103},{},[104],{"type":23,"value":105},"3. A Unified Classification System",{"type":17,"tag":25,"props":107,"children":108},{},[109],{"type":23,"value":110},"Existing sample libraries face the problems of inconsistent classification systems, limited types of sensors, and weak generalization capability of models. To address these problems, the joint team released LuoJiaSET, a large remote sensing image sample dataset that complies with the Open Geospatial Consortium (OGC) standards.",{"type":17,"tag":25,"props":112,"children":113},{},[114],{"type":23,"value":115},"LuoJiaSET is built on a unified classification system with specifications on the collection, content, and processing of samples. This system improves the efficiency in collection, creation, management, sharing, and application of remote sensing image sample libraries of multiple levels and types.",{"type":17,"tag":25,"props":117,"children":118},{},[119],{"type":23,"value":120},"You are welcome to try out LuoJiaNET online in the MindSpore community.",{"type":17,"tag":25,"props":122,"children":123},{},[124],{"type":17,"tag":52,"props":125,"children":126},{},[127],{"type":23,"value":128},"Related links",{"type":17,"tag":25,"props":130,"children":131},{},[132],{"type":23,"value":133},"1. LuoJiaSET Intelligent Remote Sensing Service Platform:",{"type":17,"tag":25,"props":135,"children":136},{},[137],{"type":17,"tag":138,"props":139,"children":143},"a",{"href":140,"rel":141},"http://58.48.42.237/luojiaSet",[142],"nofollow",[144],{"type":23,"value":140},{"type":17,"tag":25,"props":146,"children":147},{},[148],{"type":23,"value":149},"2. LuoJiaNET Intelligent Remote Sensing Service Platform",{"type":17,"tag":25,"props":151,"children":152},{},[153],{"type":17,"tag":138,"props":154,"children":157},{"href":155,"rel":156},"http://58.48.42.237/luojiaNet",[142],[158],{"type":23,"value":155},{"type":17,"tag":25,"props":160,"children":161},{},[162],{"type":23,"value":163},"3. LuoJiaNET source code:",{"type":17,"tag":25,"props":165,"children":166},{},[167],{"type":17,"tag":138,"props":168,"children":171},{"href":169,"rel":170},"https://github.com/WHULuoJiaTeam/luojianet",[142],[172],{"type":23,"value":169},{"type":17,"tag":25,"props":174,"children":175},{},[176],{"type":17,"tag":138,"props":177,"children":180},{"href":178,"rel":179},"https://gitlab.com/whuluojiateam/luojianet",[142],[181],{"type":23,"value":178},{"type":17,"tag":25,"props":183,"children":184},{},[185],{"type":17,"tag":138,"props":186,"children":189},{"href":187,"rel":188},"https://bitbucket.org/luojiateam/luojianet",[142],[190],{"type":23,"value":187},{"title":7,"searchDepth":192,"depth":192,"links":193},4,[],"markdown","content:news:en:1576.md","content","news/en/1576.md","news/en/1576","md",1776506043605]