Abstract Semantic segmentation of remote sensing imagery (RSI) is critical in many domains due to the diverse landscapes and different sizes of geo-objects that RSI contains, making semantic segmentation challenging. In this paper, a convolutional network, named Adaptive Feature Fusion UNet (AFF-UNet), is proposed to optimize the semantic segmentation performance. The model has three key aspects: (1) dense skip connections architecture and an adaptive feature fusion module that adaptively weighs different levels of feature maps to achieve adaptive feature fusion, (2) a channel attention convolution block that obtains the relationship between different channels using a tailored configuration, and (3) a spatial attention module that obtains the relationship between different positions. AFF-UNet was evaluated on two public RSI datasets and was quantitatively and qualitatively compared with other models. Results from the Potsdam dataset showed that the proposed model achieved an increase of 1.09% over DeepLabv3 + in terms of the average F1 score and a 0.99% improvement in overall accuracy. The visual qualitative results also demonstrated a reduction in confusion of object classes, better performance in segmenting different sizes of object classes, and better object integrity. Therefore, the proposed AFF-UNet model optimizes the accuracy of RSI semantic segmentation.
Ocean primary productivity generated by phytoplankton is critical for ocean ecosystems and the global carbon cycle. Accurate ocean primary productivity forecasting months in advance is beneficial for marine management. Previous persistence-based prediction studies ignore the temporal memories of multiple relevant factors and the seasonal forecasting skill drops quickly with increasing lead time. On the other hand, the emerging ensemble climate forecasts are not well considered as new predictability sources of ocean conditions. Here we proposed a joint forecasting model by combining the seasonal climate predictions from ten heterogeneous models and the temporal memories of relevant factors to examine the monthly predictability of ocean productivity from 0.5- to 11.5-month lead times. The results indicate that a total of ~90% and ~20% productive oceans are expected to be skillfully predicted by the combination of seasonal SST predictions and local memory at 0.5- and 4.5-month leads, respectively. The joint forecasting model improves by 10% of the skillfully predicted areas at 6.5-month lead relative to the prediction by productivity persistence. The hybrid data-driven and model-driven forecasting approach improves the predictability of ocean productivity relative to individual predictions, of which the seasonal climate predictions contribute largely to the skill improvement over the equatorial Pacific and Indian Ocean. These findings highlight the advantages of the integration of climate predictions and temporal memory for ocean productivity forecasting and may provide useful seasonal forecasting information for ocean ecosystem management.
Yellow River Basin urban agglomeration (YRBU) is the main carrier of regional socio-economic development in the Yellow River Basin, and its eco-environmental quality, urbanization, and coupling coordination degree are facing higher demands. It is of great significance for the development of YRBU to understand the interactive coupling relationship between the eco-environment and urbanization development from the multi-scale perspective. This research intended to understand the spatio-temporal characteristics of eco-environmental quality, urbanization, and coupling coordination degree in the study area from 2013 to 2021. We proposed an Adjusted Remote Sensing Ecological Index (A-RSEI), integrated Sentinel-2A, Landsat 8, and other remote sensing data to evaluate the eco-environmental quality of the study area, from 2013 to 2021. Coupled coordination degree (CCD) model was used to obtain the CCD between eco-environmental quality and urbanization. In addition, spatio-temporal and multi-scale analysis was carried out from the perspectives of urban agglomeration, municipal, county, and pixel scales. Combined with spatial autocorrelation analysis and Tapio decoupling model, the CCD was further explored. The results show that the proposed A-RSEI model is more suitable for monitoring the eco-environmental quality of the Yellow River Basin. The coupling coordination degree of eco-environment and urbanization in most regions of the study area are rising in a relatively green development trend. The multi-scale analysis among eco-environmental quality, urbanization, and CCD can not only indicate the impact of the central city on its surrounding areas but also help to describe the details of CCD combined with the terrain. The comprehensive discrimination of urban agglomeration and county scale is helpful to express the relationship between urbanization and eco-environmental quality centered on a certain city. The results can provide scientific support for eco-environment protection and high-quality development of the Yellow River Basin.
A robust and successful learning methodology based on sequential Relevance Vector Machine Regression (RVR) for identifying correct matches and mismatches from initial SIFT matching points is proposed. We introduce a nonlinear matching function between the corresponding points set from the given image pairs. The sequential RVR algorithm is used to learn the matching function relationship; correct matches and mismatches can be detected by checking the residuals whether they are consistent with the matching function models. Experiments show that the proposed method can efficiently pick out the mismatches and preserve the correct matches, especially on the larger view angle matching condition, and outperforms to state-of-the art approaches.
In this paper, an active shape model (ASM) based facial feature localization strategy is proposed, which employs a local binary pattern (LBP) probability model. Due to the computation simplicity and illumination insensitivity of LBP texture descriptor and the learning ability of the probability model, the algorithm is robust and fast. In addition, component-based ASM is used to impose reasonable constraints on the shape. Multi-state shape and texture models with state classifier are trained to handle highly flexible components, i.e. eyes and mouth. Our database consisting of tens of persons with various expressions and illuminations is used to train and verify the proposed algorithm. The experiments demonstrate its accuracy, efficiency and robustness.
Abrupt changes in environmental temperature, wind and humidity can lead to great threats to human life safety. The Gansu marathon disaster of China highlights the importance of early warning of hypothermia from extremely low appar-ent temperature (AT). Here a deep convolutional neural network model together with a statistical downscaling framework is developed to forecast environmental factors for 1 to 12 h in advance to evaluate the effectiveness of deep learning for AT prediction at 1 km resolution. The experiments use data for temperature, wind speed and relative humidity in ERA-5 and the results show that the developed deep learning model can predict the upcoming extreme low temperature AT event in the Gansu marathon region several hours in advance with better accuracy than climatological and persistence forecast-ing methods. The hypothermia time estimated by the deep learning method with a heat loss model agrees well with the observed estimation at 3-hour lead. Therefore, the developed deep learning forecasting method is effective for short-term AT prediction and hypothermia warnings at local areas.
With the land use data of Dangyang City, Yichang City, Hubei Province in 2000, 2010, and 2020, 13 factors of nature, economy, and society were selected as the driving factors affecting the change of Cropland by using the land-use transition matrix and the kernel density analysis. In this study, the dynamic degree model of cropland use was introduced, combined with the geographic detector method, to explore the spatiotemporal variation characteristics and main driving factors of cropland in various towns in Dangyang City from 2000 to 2020. The results showed that the distribution of cropland generally showed a decreasing trend from southeast to northwest. From 2010 to 2020, the cropland in Dangyang City changed significantly, and the cropland area decreased. Economic and social factors turn out to be the main factors affecting the distribution of cropland. When interacting with other factors, the impact on cropland is weakened, and natural factors have interactive effects on cropland.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)