When performing remote sensing image segmentation, practitioners often encounter various challenges, such as a strong imbalance in the foreground–background, the presence of tiny objects, high object density, intra-class heterogeneity, and inter-class homogeneity. To overcome these challenges, this paper introduces AerialFormer, a hybrid model that strategically combines the strengths of Transformers and Convolutional Neural Networks (CNNs). AerialFormer features a CNN Stem module integrated to preserve low-level and high-resolution features, enhancing the model’s capability to process details of aerial imagery. The proposed AerialFormer is designed with a hierarchical structure, in which a Transformer encoder generates multi-scale features and a multi-dilated CNN (MDC) decoder aggregates the information from the multi-scale inputs. As a result, information is taken into account in both local and global contexts, so that powerful representations and high-resolution segmentation can be achieved. The proposed AerialFormer was benchmarked on three benchmark datasets, including iSAID, LoveDA, and Potsdam. Comprehensive experiments and extensive ablation studies show that the proposed AerialFormer remarkably outperforms state-of-the-art methods.

CORONA satellite imagery, preserving an account of the earth's surface from 40 years ago, is a most important archaeological survey tool and we have often sung its praises. Here the authors use new procedures to extend the competence and revelations of CORONA even further. Stereo pairs derived from images taken from fore and aft of the satellite give three dimensional images of landscapes and even individual sites. Techniques of modelling and rectification restore the sites to their original shape without recourse to survey on the ground – in many cases no longer possible since the sites have been buried, inundated or erased. The ingenuity shown here indicates that results from CORONA are only going to get better.

Potential visual impact on coastal lands has emerged as a major concern in the development of offshore wind facilities in the United States and Europe. Optimal siting of offshore facilities requires accurate knowledge of the relationship between distance and the visibility of wind turbines. Past assessments of offshore wind turbine visibility were based on smaller turbines and facilities in use at the time and underestimate visibility for current projects, which use more and larger larger turbines. This study is a preliminary assessment of the visibility of offshore wind facilities in the United Kingdom. Study objectives included identifying the maximum distances the facilities could be seen in both daytime and nighttime views and assessing the effect of distance on visual contrasts associated with the facilities. Results showed that small to moderately sized facilities were visible to the unaided eye at distances greater than 42 km [26 miles (mi)], with turbine blade movement visible up to 39 km (24 mi). At night, aerial hazard navigation lighting was visible at distances greater than 39 km (24 mi). The observed wind facilities were judged to be a major focus of visual attention at distances up to 16 km (10 mi), were noticeable to casual observers at distances of almost 29 km (18 mi), and were visible with extended or concentrated viewing at distances beyond 40 km (25 mi).