SLAM 3D Digital Terrain Mapping with SqueezeNet Driven by Road Traffic Data

In order to improve the efficiency of dynamic visualization large-scale road traffic data in a web environment, this paper proposes SLAM 3D digital terrain map method using SqueezeNet environment. We propose hierarchical organization network, taking into account attributes and drawing roads at different view heights; multiroad merging based on line segment indexing feature WebGL (web graphics library) technology, optimizing scene by combining rejection multithreading technology. The prototype system was developed case studies were carried out national network as an example. experimental results show that frame rate environment is above 40 frames/second, which 20–30 frames/second higher than Baidu’s ECharts GL method.