Wei Zhu

Dr. Wei Zhu


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
Verified
Journal Article
Open access
PDF file
WZWei Zhu
Wei Zhu
QGQingsheng Guo
Qingsheng Guo
NYNai Yang
Nai Yang
YTYing Tong
Ying Tong
CZChuanbang Zheng
Chuanbang Zheng
Multi-scale electronic map tiles are important basic geographic information data, and an approach based on deep learning is being used to generate multi-scale map tiles. Although generative adversarial networks (GANs) have demonstrated great potential in single-scale electronic map tile generation, further research concerning multi-scale electronic map tile generation is needed to meet cartographic requirements. We designed a multi-scale electronic map tile generative adversarial network (MsM-GAN), which consisted of several GANs and could generate map tiles at different map scales sequentially. Road network data and building footprint data from OSM (Open Street Map) were used as auxiliary information to provide the MsM-GAN with cartographic knowledge about spatial shapes and spatial relationships when generating electronic map tiles from remote sensing images. The map objects which should be deleted or retained at the next map scale according to cartographic standards are encoded as auxiliary information in the MsM-GAN when generating electronic map tiles at smaller map scales. In addition, in order to ensure the consistency of the features learned by several GANs, the density maps constructed from specific map objects are used as global conditions in the MsM-GAN. A multi-scale map tile dataset was collected from MapWorld, and experiments on this dataset were conducted using the MsM-GAN. The results showed that compared to other image-to-image translation models (Pix2Pix and CycleGAN), the MsM-GAN shows average increases of 10.47% in PSNR and 9.92% in SSIM and has the minimum MSE values at all four map scales. The MsM-GAN also performs better in visual evaluation. In addition, several comparative experiments were completed to verify the effect of the proposed improvements.

Verified
Journal Article
Open access
PDF file
YTYing Tong
Ying Tong
QGQingsheng Guo
Qingsheng Guo
WZWei Zhu
Wei Zhu
CZChuanbang Zheng
Chuanbang Zheng
Typification is a crucial generalization operator for maintaining the structure of building patterns in map generalization. However, in densely built-up areas, the structure of building patterns is complicated, and spatial conflicts exist not only within building patterns but also between them. Applying typification solely to individual patterns fails to resolve conflicts between them and does not adequately preserve their spatial distribution. To address these challenges, this study proposes a new typification method that treats linear building patterns with common buildings as a whole named combined linear patterns, simultaneously resolving the spatial conflict both within and between combined linear patterns. Firstly, linear building patterns and combined linear patterns are recognized, respectively, using visual perception principles and the common elements of linear building patterns. Secondly, spatial conflicts within the combined linear pattern are resolved through the segmental building reconstruction of linear building patterns and a reduction in linear building patterns at junction buildings. Finally, spatial conflicts between combined linear patterns are solved based on the position and number of conflict buildings in the patterns. The proposed method is evaluated using a building dataset from OpenStreetMap. The experimental results indicate that the proposed method can preserve the spatial distribution of building patterns with better legibility than the usual typification method for individual linear building patterns.