Road extraction techniques based on remote sensing image have significantly advanced. Currently, fully supervised road segmentation neural networks based on remote sensing images require a significant number of densely labeled road samples, limiting their applicability in large-scale scenarios. Consequently, semi-supervised methods that utilize fewer labeled data have gained increasing attention. However, the imbalance between a small quantity of labeled data and a large volume of unlabeled data leads to local detail errors and overall cognitive mistakes in semi-supervised road extraction. To address this challenge, this paper proposes a novel consistency self-training semi-supervised method (CSSnet), which effectively learns from a limited number of labeled data samples and a large amount of unlabeled data. This method integrates self-training semi-supervised segmentation with semi-supervised classification. The semi-supervised segmentation component relies on an enhanced generative adversarial network for semantic segmentation, which significantly reduces local detail errors. The semi-supervised classification component relies on an upgraded mean-teacher network to handle overall cognitive errors. Our method exhibits excellent performance with a modest amount of labeled data. This study was validated on three separate road datasets comprising high-resolution remote sensing satellite images and UAV photographs. Experimental findings showed that our method consistently outperformed state-of-the-art semi-supervised methods and several classic fully supervised methods.

Extracting roads from remote sensing images holds significant practical value across fields like urban planning, traffic management, and disaster monitoring. Current Convolutional Neural Network (CNN) methods, praised for their robust local feature learning enabled by inductive biases, deliver impressive results. However, they face challenges in capturing global context and accurately extracting the linear features of roads due to their localized receptive fields. To address these shortcomings of traditional methods, this paper proposes a novel parallel encoder architecture that integrates a CNN Encoder Module (CEM) with a Transformer Encoder Module (TEM). The integration combines the CEM’s strength in local feature extraction with the TEM’s ability to incorporate global context, achieving complementary advantages and overcoming limitations of both Transformers and CNNs. Furthermore, the architecture also includes a Linear Convolution Module (LCM), which uses linear convolutions tailored to the shape and distribution of roads. By capturing image features in four specific directions, the LCM significantly improves the model’s ability to detect and represent global and linear road features. Experimental results demonstrate that our proposed method achieves substantial improvements on the German-Street Dataset and the Massachusetts Roads Dataset, increasing the Intersection over Union (IoU) of road class by at least 3% and the overall F1 score by at least 2%.

Simulations based on site-specific crop growth models have been widely used to obtain regional yield potential estimates for food security assessments at the regional scale. By dividing a region into nonoverlapping basic spatial units using appropriate zonation schemes, the data required to run a crop growth model can be reduced, thereby improving the simulation efficiency. In this study, we explored the impacts of different zonation schemes on estimating the regional yield potential of the Chinese winter wheat area to obtain the most appropriate spatial zonation scheme of weather sites therein. Our simulated results suggest that the upscaled site-specific yield potential is affected by the zonation scheme and by the spatial distribution of sites. As such, the distribution of a small number of sites significantly affected the simulated regional yield potential under different zonation schemes, and the zonation scheme based on sunshine duration clustering zones could effectively guarantee the simulation accuracy at the regional scale. Using the most influential environmental variable of crop growth models for clustering can get the better zonation scheme to upscale the site-specific simulation results. In contrast, a large number of sites had little effect on the regional yield potential simulation results under the different zonation schemes.

Significant advancements have been achieved in road surface extraction based on high-resolution remote sensing image processing. Most current methods rely on fully supervised learning, which necessitates enormous human effort to label the image. Within this field, other research endeavors utilize weakly supervised methods. These approaches aim to reduce the expenses associated with annotation by leveraging sparsely annotated data, such as scribbles. This paper presents a novel technique called a weakly supervised network using scribble-supervised and edge-mask (WSSE-net). This network is a three-branch network architecture, whereby each branch is equipped with a distinct decoder module dedicated to road extraction tasks. One of the branches is dedicated to generating edge masks using edge detection algorithms and optimizing road edge details. The other two branches supervise the model’s training by employing scribble labels and spreading scribble information throughout the image. To address the historical flaw that created pseudo-labels that are not updated with network training, we use mixup to blend prediction results dynamically and continually update new pseudo-labels to steer network training. Our solution demonstrates efficient operation by simultaneously considering both edge-mask aid and dynamic pseudo-label support. The studies are conducted on three separate road datasets, which consist primarily of high-resolution remote-sensing satellite photos and drone images. The experimental findings suggest that our methodology performs better than advanced scribble-supervised approaches and specific traditional fully supervised methods.