The present paper presents techniques for polyline simplification based on an artificial neural network within the constraints of generalization knowledge. The proposed method measures polyline shape characteristics that influence polyline simplification using abstracted descriptors and then introduces these descriptors into the artificial neural network as input properties. In total, 18 descriptors categorized into three types are presented in detail. In a second approach, map simplification principles are abstracted as controllers, imposed after the output layer of the trained artificial neural network to make the polyline simplification comply with these principles. This study worked with three controllers – a basic controller and two knowledge-based controllers. These descriptors and controllers abstracted from generalization knowledge were tested in experiments to determine their efficacy in polyline simplification based on the artificial neural network. The experimental results show that the utilization of abstracted descriptors and controllers can constrain the artificial neural network-based polyline simplification according to polyline shape characteristics and simplification principles.

There remains several challenges that are encountered in the task of extracting buildings from aerial imagery using convolutional neural networks (CNNs). First, the tremendous complexity of existing building extraction networks impedes their practical application. In addition, it is arduous for networks to sufficiently utilize the various building features in different images. To address these challenges, we propose an efficient network called MSL-Net that focuses on both multiscale building features and multilevel image features. First, we use depthwise separable convolution (DSC) to significantly reduce the network complexity, and then we embed a group normalization (GN) layer in the inverted residual structure to alleviate network performance degradation. Furthermore, we extract multiscale building features through an atrous spatial pyramid pooling (ASPP) module and apply long skip connections to establish long-distance dependence to fuse features at different levels of the given image. Finally, we add a deformable convolution network layer before the pixel classification step to enhance the feature extraction capability of MSL-Net for buildings with irregular shapes. The experimental results obtained on three publicly available datasets demonstrate that our proposed method achieves state-of-the-art accuracy with a faster inference speed than that of competing approaches. Specifically, the proposed MSL-Net achieves 90.4%, 81.1% and 70.9% intersection over union (IoU) values on the WHU Building Aerial Imagery dataset, Inria Aerial Image Labeling dataset and Massachusetts Buildings dataset, respectively, with an inference speed of 101.4 frames per second (FPS) for an input image of size 3 × 512 × 512 on an NVIDIA RTX 3090 GPU. With an excellent tradeoff between accuracy and speed, our proposed MSL-Net may hold great promise for use in building extraction tasks.

Detecting buildings, segmenting building footprints, and extracting building edges from high-resolution remote sensing images are vital in applications such as urban planning, change detection, smart cities, and map-making and updating. The tasks of building detection, footprint segmentation, and edge extraction affect each other to a certain extent. However, most previous works have focused on one of these three tasks and have lacked a multitask learning framework that can simultaneously solve the tasks of building detection, footprint segmentation and edge extraction, making it difficult to obtain smooth and complete buildings. This study proposes a novel multiscale and multitask deep learning framework to consider the dependencies among building detection, footprint segmentation, and edge extraction while completing all three tasks. In addition, a multitask feature fusion module is introduced into the deep learning framework to increase the robustness of feature extraction. A multitask loss function is also introduced to balance the training losses among the various tasks to obtain the best training results. Finally, the proposed method is applied to open-source building datasets and large-scale high-resolution remote sensing images and compared with other advanced building extraction methods. To verify the effectiveness of multitask learning, the performance of multitask learning and single-task training is compared in ablation experiments. The experimental results show that the proposed method has certain advantages over other methods and that multitask learning can effectively improve single-task performance.

Convolutional neural networks (CNNs) perform well in tasks of segmenting buildings from remote sensing images. However, the intraclass heterogeneity of buildings is high in images, while the interclass homogeneity between buildings and other nonbuilding objects is low. This leads to an inaccurate distinction between buildings and complex backgrounds. To overcome this challenge, we propose an Attentional Feature Learning Network (AFL-Net) that can accurately extract buildings from remote sensing images. We designed an attentional multiscale feature fusion (AMFF) module and a shape feature refinement (SFR) module to improve building recognition accuracy in complex environments. The AMFF module adaptively adjusts the weights of multi-scale features through the attention mechanism, which enhances the global perception and ensures the integrity of building segmentation results. The SFR module captures the shape features of the buildings, which enhances the network capability for identifying the area between building edges and surrounding nonbuilding objects and reduces the over-segmentation of buildings. An ablation study was conducted with both qualitative and quantitative analyses, verifying the effectiveness of the AMFF and SFR modules. The proposed AFL-Net achieved 91.37, 82.10, 73.27, and 79.81% intersection over union (IoU) values on the WHU Building Aerial Imagery, Inria Aerial Image Labeling, Massachusetts Buildings, and Building Instances of Typical Cities in China datasets, respectively. Thus, the AFL-Net offers the prospect of application for successful extraction of buildings from remote sensing images.

In the era of information explosion, Chinese social media has become a repository for massive geographic information; however, its unique unstructured nature and diverse expressions are challenging to toponym entity recognition. To address this problem, we propose a Chinese social media named entity recognition (CSMNER) model to improve the accuracy and robustness of toponym recognition in Chinese social media texts. By combining the BERT (Bidirectional Encoder Representations from Transformers) pre-trained model with an improved IDCNN-BiLSTM-CRF (Iterated Dilated Convolutional Neural Network- Bidirectional Long Short-Term Memory- Conditional Random Field) architecture, this study innovatively incorporates a boundary extension module to effectively extract the local boundary features and contextual semantic features of the toponym, successfully addressing the recognition challenges posed by noise interference and language expression variability. To verify the effectiveness of the model, experiments were carried out on three datasets: WeiboNER, MSRA, and the Chinese social named entity recognition (CSNER) dataset, a self-built named entity recognition dataset. Compared with the existing models, CSMNER achieves significant performance improvement in toponym recognition tasks.

An important high-precision building vector mapping method automatically delineates building polygons from high-resolution remote sensing images. Deep learning methods have greatly improved the accuracy of automatic building segmentation in remote sensing images. However, building polygons in vector forms have a compact and regular structured expression effect, which corresponds more with the application requirements of cartography and geographic information systems (GIS). We propose a vector mapping method for buildings in remote sensing images with joint semantic-geometric learning to generate building polygon vectors in remote sensing images automatically. The method, aiming to provide cartography and GIS data sources, consists of three modules: multi-task segmentation, contour regularization, and polygon optimization. To reduce missing extractions and mis-extractions and obtain a complete building segmentation mask, the multi-task segmentation module performs joint semantic-geometric learning on three related tasks: building instance detection, pixel-wise contour segmentation, and edge extraction. The regularization module normalizes the segmentation mask expression using geometric constraints and image information, whereas the polygon optimization module combines geometric constraints with deep learning methods to ensure vectorization quality. The experimental results show that the proposed method adapts well to building vector extraction tasks under different scenarios and can generate building vector polygons that match the ground truth labels. This method offers significant advantages in solving problems, such as building polygon irregularity and vertex offset.