Machine Learning Classification of Buildings for Map Generalization

Lee, Jaeeun; Jang, Hanme; Yang, Jonghyeon; Yu, Kiyun

doi:10.3390/ijgi6100309

Cited by 27 publications

(19 citation statements)

References 15 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhou and Li (2017) compare different Machine Learning approaches concerning their capability to select important road links for road network generalization. Lee et al (2017) use different Machine Learning methods to classify buildings as a prior step for their generalization. An approach for generalization of lines from traffic trajectories using Deep Learning has been presented by Thiemann et al (2018).…”

Section: State Of the Artmentioning

confidence: 99%

Building Generalization Using Deep Learning

Sester

Feng

Thiemann

2018

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Cartographic generalization is a problem, which poses interesting challenges to automation. Whereas plenty of algorithms have been developed for the different sub-problems of generalization (e.g. simplification, displacement, aggregation), there are still cases, which are not generalized adequately or in a satisfactory way. The main problem is the interplay between different operators. In those cases the benchmark is the human operator, who is able to design an aesthetic and correct representation of the physical reality.Deep Learning methods have shown tremendous success for interpretation problems for which algorithmic methods have deficits. A prominent example is the classification and interpretation of images, where deep learning approaches outperform the traditional computer vision methods. In both domains &ndash; computer vision and cartography &ndash; humans are able to produce a solution; a prerequisite for this is, that there is the possibility to generate many training examples for the different cases. Thus, the idea in this paper is to employ Deep Learning for cartographic generalizations tasks, especially for the task of building generalization. An advantage of this task is the fact that many training data sets are available from given map series. The approach is a first attempt using an existing network.In the paper, the details of the implementation will be reported, together with an in depth analysis of the results. An outlook on future work will be given.

show abstract

Section: State Of the Artmentioning

confidence: 99%

Building Generalization Using Deep Learning

Sester

Feng

Thiemann

2018

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…This Special Issue assembles six novel contributions in different areas of GeoData-driven machine learning. Topics span different disciplines of GIScience: generation of street address from satellite imagery [1], land-cover classification of polarimetric Synthetic Aperture Radar (PolSAR) images [2], extraction of buildings from maps to perform generalization [3], land-cover classification from satellite image time series [4], automatic selection of buildings based on cartographic constraints [5] and satellite image retrieval and recommendation [6].…”

mentioning

confidence: 99%

“…Deep learning is used in two works [1,4]: in the former, a convolutional neural network extracts roads from satellite images as an initial step to generate street addresses, while in the latter, a sequential convolutional recurrent neural network provides robust end-to-end land-cover and land-use mapping from satellite image time series. Decision trees in the form of single decision trees [3] or random forests [2] are used for building detection in the first and land-cover classification in the second contribution. Lee, et al [3] also test other standard machine learning approaches such as support vector machines, k-nearest-neighbor and naïve Bayes classification to explore which methodologies are most appropriate for map generalization purposes.…”

mentioning

confidence: 99%

“…Decision trees in the form of single decision trees [3] or random forests [2] are used for building detection in the first and land-cover classification in the second contribution. Lee, et al [3] also test other standard machine learning approaches such as support vector machines, k-nearest-neighbor and naïve Bayes classification to explore which methodologies are most appropriate for map generalization purposes. Map generalization is also at the core of [5], who approach building selection with genetic algorithms that are constrained by cartographic and contextual knowledge.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Foreword to the Special Issue on Machine Learning for Geospatial Data Analysis

Wegner

Roscher

Volpi

et al. 2018

IJGI

View full text Add to dashboard Cite

Advances in machine learning research are pushing the limits of geographical information sciences (GIScience) by offering accurate procedures to analyze small-to-big GeoData. This Special Issue groups together six original contributions in the field of GeoData-driven GIScience that focus mainly on three different areas: extraction of semantic information from satellite imagery, image recommendation, and map generalization. Different technical approaches are chosen for each sub-topic, from deep learning to latent topic models.

show abstract

“…Although these methods have advantages of rapidly processing large amounts of data and ensuring high accuracy in a stand-alone environment, they also have disadvantages. The algorithms performance gradually varying depending on the data quality [22]. Transfer learning, such as that proposed in [23], transfers the learned and trained model parameters to a new model to help train the new model.…”

mentioning

confidence: 99%

Method of Constructing Point Generalization Constraints Based on the Cloud Platform

Zhou

Shen

Yang

et al. 2018

IJGI

View full text Add to dashboard Cite

As an important part of spatial data, the point feature has always been an essential element in web maps and navigation maps. With the development of location-based services and the rapid increase in volunteered geographic information and social media data, the amount of point data is increasing day by day, resulting in inevitable problems of overlay and congestion during visualization. Map generalization provides multiple algorithms that can be used to select, aggregate and make typification of points or point clusters. For the generalization of point data, however, the traditional stand-alone computing environment has difficulty with real-time realization. Currently, the rapid development of cloud computing technology provides a powerful support for improving the efficiency of map generalization. However, compared with the stand-alone environment, the data decomposition and the real-time display of point generalization in the cloud platform imposes higher requirements on the point generalization constraints, which play an important role in point-generalized process control. Based on the computational characteristics of the cloud platform, this paper analyzes the changes in point generalization constraints. In addition, our work proposes the constraints of point generalization based on the cloud platform and its construction method, builds a prototype system based on the Hadoop cloud platform. Our prototype system is tested using typical experimental data. Its efficiency and the quality of its results is examined. The results show that the efficiency and quality of point selection can be significantly improved by controlling the point generalization process with the generalization constraints in the cloud computing environment proposed in this paper. This paper provides a possible way for the realization of map generalization in the cloud computing environment. Its usability with real data and with many users accessing it will be the focus of further research. popularization of social media, the Internet of Things, sensor networks and so forth, the types and numbers of point data have grown dramatically. At the same time, along with the expansion of mobile phone terminal map applications and the increase in wireless network bandwidth, the number of point data displayed by mobile phones is also increasing rapidly. When displaying a large number of point features, many visual problems are inevitably encountered [2]. For example, when performing zooming operations, crowding and overlay between symbols can easily occur. This outcome is even more important in the case of maps for visually impaired users, where the usable map load is much more limited [3]. Using a smaller version of map symbols is a solution only in the case of a small difference between the original and target scale. Online filtering alleviates this problem to some extent, but it only acts as a solution for computer problems and cannot reflect the spatial relationship between the features.Cartographic generalization is an indispensable technique for ...

show abstract

Machine Learning Classification of Buildings for Map Generalization

Cited by 27 publications

References 15 publications

Building Generalization Using Deep Learning

Building Generalization Using Deep Learning

Foreword to the Special Issue on Machine Learning for Geospatial Data Analysis

Method of Constructing Point Generalization Constraints Based on the Cloud Platform

Contact Info

Product

Resources

About