This paper presents a low-cost and efficient method for 3D virtual urban scene reconstruction based on multi-source remote sensing big data and deep learning. By integrating maps, satellite optical images, and digital terrain model (DTM), the proposed method achieves a reasonable reconstructed 3D model for complex urban. The method consists of two independent convolutional neural networks (CNN) to process the land cover and the building height extraction. The proposed method is then tested on a 100 km 2 scene in San Diego, USA, including about 30 000 buildings. The land cover classification achieves an overall accuracy (OA) of 80.4% for eight types of land as defined in NLCD 2011 datasets. Building height estimation achieves an average error at 1.9 meters on NYC open data, the building footprint. Furthermore, the scene reconstruction including the estimation of both land cover and building height can be finished in 10 min on a single NVidia Titan X GPU.INDEX TERMS Optical image, urban reconstruction, convolutional neural networks.

No abstract

In comparison with the conventional radar altimeter, synthetic aperture radar altimeter (SRAL) is developed with better along‐track spatial resolution and height precision. When SRAL improves the performance of height measurement, it also complicates the radar system designing, signal processing, data analyzing, and system testing. The SRAL raw signal is the result of the interaction between the transmitted signal and the complex terrain surfaces. Generally, the SRAL is used in ocean applications since the PDF can be derived from different ocean surfaces. However, the actual acquired SRAL waveforms can be from complex and variable surface, which may include mountains, cities, forests, lakes, and other types of terrain surfaces. These waveforms contain information about the surfaces of the Earth although difficult to analyze. This paper focuses on the study of the raw signal simulation of SRAL for complex terrain surfaces including land surfaces. The principle of radar altimeter, the electromagnetic characteristics of land targets, and the SRAL signal processing are analyzed. The geometric models are extracted from digital elevation model, and the scattering characteristics of the terrain surface are established with three scattering models. Finally, three cases of terrain surfaces, namely, ice surface, mountain area, and urban area, are simulated and validated against actual SRAL signals measured by the Sentinel‐3 SRAL sensor.

Underdetermined blind source separation is a signal processing technique that is more suitable for practical applications and aims to separate the source signals from the mixed signals. The mixing matrix estimation is a major step in the underdetermined blind source separation. Since the current methods for estimating the mixing matrix have the disadvantages of insufficient accuracy or weak noise immunity, a two-stage single-source point screening that combines the cosine angle algorithm and the L1-norm optimization algorithm is proposed. During the first stage, the first-stage single-source points are extracted from the original mixed signals using the cosine angle algorithm. During the second stage, based on the L1-norm optimization algorithm, the reference single-source points are extracted from the original mixed signals. The reference single-source points are then clustered to obtain the clustering center, which is defined as the reference center. In combination with the reference center, the deviation and interference points in the first-stage single-source points are eliminated by the cosine distance. The remaining signal points are considered as the second-stage single-source points, which are clustered to obtain the mixing matrix estimation. Experiments on simulated and speech signals show that the proposed method can obtain more accurate and robust mixing matrix estimation, leading to better separation of the source signals.