Road extraction from remote sensing images is significant for urban planning, intelligent transportation, and vehicle navigation. However, it is challenging to automatically extract roads from remote sensing images because the scale difference of roads in remote sensing images varies greatly, and slender roads are difficult to identify. Moreover, the road in the image is often blocked by the shadows of trees and buildings, which results in discontinuous and incomplete extraction results. To solve the above problems, this paper proposes a multiscale feature encoding and long-range context-aware network (MECA-Net) for road extraction. MECA-Net adopts an encoder–decoder structure and contains two core modules. One is the multiscale feature encoding module, which aggregates multiscale road features to improve the recognition ability of slender roads. The other is the long-range context-aware module, which consists of the channel attention module and the strip pooling module, and is used to obtain sufficient long-range context information from the channel dimension and spatial dimension to alleviate road occlusion. Experimental results on the open DeepGlobe road dataset and Massachusetts road dataset indicate that the proposed MECA-Net outperforms the other eight mainstream networks, which verifies the effectiveness of the proposed method.

Estimating clumping indices is important for determining the leaf area index (LAI) of forest canopies. The spatial distribution of the clumping index is vital for LAI estimation. However, the neglect of woody tissue can result in biased clumping index estimates when indirectly deriving them from the gap probability and LAI observations. It is difficult to effectively and automatically extract woody tissue from digital hemispherical photos. In this study, a method for the automatic detection of trunks from Terrestrial Laser Scanning (TLS) data was used. Between-crown and within-crown gaps from TLS data were separated to calculate the clumping index. Subsequently, we analyzed the gap probability, clumping index, and LAI estimates based on TLS and HemiView data in consideration of woody tissue (trunks). Although the clumping index estimated from TLS had better agreement (R² = 0.761) than that from HemiView, the change of angular distribution of the clumping index affected by the trunks from TLS data was more obvious than with the HemiView data. Finally, the exclusion of the trunks led to a reduction in the average LAI by ~19.6% and 8.9%, respectively, for the two methods. These results also showed that the detection of woody tissue was more helpful for the estimation of clumping index distribution. Moreover, the angular distribution of the clumping index is more important for the LAI estimate than the average clumping index value. We concluded that woody tissue should be detected for the clumping index estimate from TLS data, and 3D information could be used for estimating the angular distribution of the clumping index, which is essential for highly accurate LAI field measurements.

Imagery geometry models (IGMs) of the high-resolution satellite images (HRSIs) are always of great interest in the photogrammetry and remote sensing community for the raising new kinds of sensors and imaging systems. Especially the generalized sensor models (GSMs) have been widely used for positioning of satellite images, and the accuracy are already validated. Since Back propagation (BP) neural network is a better choice for the two key reasons of the replacement of physical sensor models by generalized sensor models, numerous mathematical estimations for every specialized sensor, and secret equations of the IGMs. Experiments are carried out to test the approximation accuracy of the new generalized sensor model. And the experimental results show that, the BP neural network is of extremely high accuracy for satellite imagery photogrammetric restitution.