A Combined Convolutional Neural Network for Urban Land-Use Classification with GIS Data

Abstract

The classification of urban land-use information has become the underlying database for a variety of applications including urban planning and administration. The lack of datasets and changeable semantics of land-use make deep learning methods suffer from low precision, which prevent improvements in the effectiveness of using AI methods for applications. In this paper, we first used GIS data to produce a well-tagged and high-resolution urban land-use image dataset. Then, we proposed a combined convolutional neural network named DUA-Net for complex and diverse urban land-use classification. The DUA-Net combined U-Net and Densely connected Atrous Spatial Pyramid Pooling (DenseASPP) to extract Remote Sensing Imagers (RSIs) features in parallel. Then, channel attention was used to efficiently fuse the multi-source semantic information from the output of the double-layer network to learn the association between different land-use types. Finally, land-use classification of high-resolution urban RSIs was achieved. Experiments were performed on the dataset of this paper, the publicly available Vaihingen dataset and Potsdam dataset with overall accuracy levels reaching 75.90%, 89.71% and 89.91%, respectively. The results indicated that the complex land-use types with heterogeneous features were more difficult to extract than the single-feature land-cover types. The proposed DUA-Net method proved suitable for high-precision urban land-use classification, which will be of great value for urban planning and national land resource surveying.

A Spatiotemporal Hierarchical Analysis Method for Urban Traffic Congestion Optimization Based on Calculation of Road Carrying Capacity in Spatial Grids

“…Grid-based data management and analysis methods are commonly used in resource management, analysis applications of geographic information systems and geospatial information system platforms, and management and analysis applications of road traffic. On the one hand, grid models are used to establish resource and scene organization based on geospatial partitioning; an example is the organization method of two-dimensional base maps in combination models of two-dimensional image data or three-dimensional city models [24]. On the other hand, in national land, planning, and transportation applications, grid models are also used for regional resource analysis and planning, such as grid-based traffic flow analyses, vehicle-to-grid networks [25], vehicle-to-grid layout-based sustainable urban networks [26], and vehicle-to-grid power-grid service methods [27], which are used for electric vehicle location planning.…”

A Review and Meta-analysis of Semantic Segmentation Models in Land Use/ Land Cover Mapping

“…Publications in urban studies also cover 6% in land survey management [35] and 25% in urban classification and detection [36][37][38][39][40] particularly in building applications and for Urban Land-Use Classification [41].…”

Man-made object segmentation around reservoirs by an end-to-end two-phase deep learning-based workflow

“…Residential land, industrial land, traffic land, woodland, and unused land are five defined classes by Ref. 18 for collected RGB images with 0.5 m resolution. To segment images, they proposed a workflow in which the images are fed to two networks in parallel.…”