A vast amount of location information exists in unstructured texts, such as social media posts, news stories, scientific articles, web pages, travel blogs, and historical archives. Geoparsing refers to recognizing location references from texts and identifying their geospatial representations. While geoparsing can benefit many domains, a summary of its specific applications is still missing. Further, there is a lack of a comprehensive review and comparison of existing approaches for location reference recognition, which is the first and core step of geoparsing. To fill these research gaps, this review first summarizes seven typical application domains of geoparsing: geographic information retrieval, disaster management, disease surveillance, traffic management, spatial humanities, tourism management, and crime management. We then review existing approaches for location reference recognition by categorizing these approaches into four groups based on their underlying functional principle: rule-based, gazetteer matching–based, statistical learning-–based, and hybrid approaches. Next, we thoroughly evaluate the correctness and computational efficiency of the 27 most widely used approaches for location reference recognition based on 26 public datasets with different types of texts (e.g., social media posts and news stories) containing 39,736 location references worldwide. Results from this thorough evaluation can help inform future methodological developments and can help guide the selection of proper approaches based on application needs.

Extracting precise location information from microblogs is a crucial task in many applications, particularly in disaster response, revealing where damages are, where people need assistance, and where help can be found. A crucial prerequisite to location extraction is place name extraction. In this paper, we present GazPNE: a hybrid approach to place name extraction which fuses rules, gazetteers, and deep learning techniques without requiring any manually annotated data. The core of the approach is to learn the intrinsic characteristics of multi-word place names with deep learning from gazetteers. Specifically, GazPNE consists of a rule-based system to select n-grams from the microblogs that potentially contain place names, and a C-LSTM model that decides if the selected n-gram is a place name or not. The C-LSTM is trained on 388.1 million examples containing 6.8 million positive examples with US and Indian place names extracted from OpenStreetMap and 381.3 million negative examples synthesized by rules. We evaluate GazPNE against the SoTA on a manually annotated 4,500 tweet dataset which contains 9,026 place names from three foods: 2016 in Louisiana (US), 2016 in Houston (US), and 2015 in Chennai (India). GazPNE achieves SotA performance on the test data with an F1 of 0.84.

Façade parsing is an essential process before the 3-D modeling of digital or virtual 3-D city models. The existing grammar-based approaches for façade parsing rely on strong prior knowledge but can obtain façade parts with better structure. Pixelwise-segmentation-based approaches achieve façade parsing with much less knowledge but the resulting structure of façade parts is normally incomplete. Both these approaches are restricted by their high reliance on the data set. Therefore, they cannot be applied for façade parsing with complex scenes. To address this issue, we built a large street-level data set by taking Mapillary images as the training data for more general scenes. At the same time, we propose a new pipeline based on convolutional neural network (CNN) that combines pixelwise segmentation and global object detection to achieve better results for facade parsing. Our pipeline can be applied to façade images after rectification and street-level façade images with complex scenes. The result of the ablation study demonstrates that the design of our pipeline is effective. We test our pipeline on the classic ECP2011 data set and our new large street-level data set. Our pipeline achieves state-of-the-art results for both the data sets: an accuracy of 98.2% and the mean average precision (mAP) of 98.8% on the ECP2011 data set as well as the mAP of 81.1% for façade parts parsing on our street-level data set.

The concept of “human as sensors” defines a new sensing model, in which humans act as sensors by contributing their observations, perceptions, and sensations. This is crucial for the development of Social Internet of Things, which is an integral part of cyber-physical–social systems. Online social media platforms, as the most active places where users act as social sensors, are responsive to real-world events and are useful for gathering situational information in real time. Unfortunately, posts rarely contain structured geographic information, thus hindering their usage for contributing to various challenges, such as emergency response. We address this limitation by introducing a general approach for extracting place names from tweets, named GazPNE2. It combines global gazetteers (i.e., OpenStreetMap and GeoNames), deep learning, and pretrained transformer models (i.e., BERT and BERTweet), which requires no manually annotated data. It can extract place names at both coarse (e.g., city) and fine-grained (e.g., street and POI) levels and place names with abbreviations. To fully evaluate GazPNE2 and compare it with 11 competing approaches, we use 19 public tweet data sets, containing 38 802 tweets and 22 197 places across the world. The results show GazPNE2 achieves a much higher F1 (0.8) than the other approaches. Furthermore, we apply GazPNE2 to three large unannotated tweet data sets related to over 20 crisis events (e.g., coronavirus disease 2019), containing 560 040 tweets. An F1 of 0.84 is achieved on 3000 tweets, which are randomly selected from the three data sets and then manually annotated. Code and data are available on GitHub page: https://github.com/uhuohuy/GazPNE2.

Individual tree segmentation in forest scenes provides a foundation for forest ecosystem modeling and biodiversity assessment applications. Existing approaches work well for the cases where trees do not grow in layers. However, they may fail in the scenario with understory vegetation occlusion and heavily overlapped crowns. In this work, we propose a two-stage solution for individual tree segmentation. This method combines a semantic segmentation module and an instance segmentation module. In the first stage, the semantic segmentation network classifies the point clouds into the tree and non-tree points. In the second stage, the instance segmentation module is utilized by incorporating object detection and post-processing refinement. The combination of semantic network and object detection network roughly extracts trees, filters out the understory vegetation that affects the extraction of small trees, and improves the extraction probability of small trees. Meanwhile, the method of object detection extraction trees can solve tree extraction omissions due to unclear stems. For the overlap crown, first, object detection limits the border of the tree crown of an individual tree, and further segmentation was implemented by refining clustering. Experiments show that our method solves the above problems and achieves state-of-the-art completeness and mean accuracy performances on benchmark datasets.

As the world’s largest crowdsourcing-based street view platform, Mapillary has received considerable attention in both research and practical applications. By February 2019, more than 20,000 users worldwide contributed approximately 6.3 million kilometers of streetscape sequences. In this study, we attempted to get a deep insight into the Mapillary project through an exploratory analysis from the perspective of contributors, including the development of users, the spatiotemporal analysis of active users, the contribution modes (walking, cycling, and driving), and the devices used to contribute. It shows that inequality exists in the distribution of contributed users, similar to that in other volunteered geographic information (VGI) projects. However, the inequality in Mapillary contribution is less than in OpenStreetMap (OSM). Compared to OSM, the other main difference is that the data collection demonstrated obvious seasonal variation because contributions to OSM can be accomplished on a computer, whereas images have to be captured on the streets for Mapillary, and this is considerably affected by seasonal weather.