Application of Satellite Image Segmentation for Urban Planning Optimization
Abstract— This article presents research results of a convolution neural network for building detection on high-resolution aerial images of Planet database. Jaccard index was used for analysis of the quality of machine learning algorithm. This index of similarity compares results of algorithms with real masks. The masks were sliced on smaller parts together with images before training of developed model. The convolution neural network was launched on NVIDIA DGX-1 supercomputer, which was provided by AIcenter of P.G Demidov Yaroslavl State University. The problem of building detection on satellite images can be put into practice for urban planning, building control, search of the best locations for outlets etc.
Index Terms— machine learning, aerial image segmentation, building detection.
P.G. Demidov Yaroslavl State University, RUSSIA
Vladimir Khryashchev, Anna Ostrovskaya, Alexander Semenov
People’s Friendship University of Russia, RUSSIA
Cite: Vladimir Khryashchev, Leonid Ivanovsky, Anna Ostrovskaya, Alexander Semenov, "Application of Satellite Image Segmentation for Urban Planning Optimization," Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering, pp. 171-175, Hong Kong, 15-17 June, 2019.