Large-scale building extraction in very high-resolution aerial imagery using Mask R-CNN

Stiller, Dorothee; Stark, Thomas; Wurm, Michael; Dech, Stefan; Taubenböck, Hannes

doi:10.1109/jurse.2019.8808977

Cited by 26 publications

(23 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The building detection result is the carrier of the NoS estimation result, not the focus of this paper. Moreover, the problem of building detection using Mask R-CNN has been studied by many scholars [15][16][17][18][19]. Therefore, the performance of NoS estimation task will be taken as the center in the evaluation of experiments.…”

Section: Resultsmentioning

confidence: 99%

Number of Building Stories Estimation from Monocular Satellite Image Using a Modified Mask R-CNN

Tang

2020

Remote Sensing

View full text Add to dashboard Cite

Stereo photogrammetric survey used to be used to extract the height of buildings, then to convert the height to number of stories through certain rules to estimate the number of stories of buildings by means of satellite remote sensing. In contrast, we propose a new method using deep learning to estimate the number of stories of buildings from monocular optical satellite image end to end in this paper. To the best of our knowledge, this is the first attempt to directly estimate the number of stories of buildings from monocular satellite images. Specifically, in the proposed method, we extend a classic object detection network, i.e., Mask R-CNN, by adding a new head to predict the number of stories of detected buildings from satellite images. GF-2 images from nine cities in China are used to validate the effectiveness of the proposed methods. The result of experiment show that the mean absolute error of prediction on buildings whose stories between 1–7, 8–20, and above 20 are 1.329, 3.546, and 8.317, respectively, which indicate that our method has possible application potentials in low-rise buildings, but the accuracy in middle-rise and high-rise buildings needs to be further improved.

show abstract

Section: Resultsmentioning

confidence: 99%

Number of Building Stories Estimation from Monocular Satellite Image Using a Modified Mask R-CNN

Tang

2020

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…A variation of image translation, dropout, and gamma adjustments from [51] is used to increase the original data by a factor of four; each of these augmented image patches is then rotated three times by 90 • . The augmenters are listed in Table II and are chosen based on successful training techniques from [52], [53]. Table I provides insight about the dataset used for training the XFCN.…”

Section: A Data Preprocessing and Data Samplingmentioning

confidence: 99%

Satellite-Based Mapping of Urban Poverty With Transfer-Learned Slum Morphologies

Stark

Wurm

Zhu

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

In the course of global urbanization, poverty in cities has been observed to increase, especially in the Global South. Poverty is one of the major challenges for our society in the upcoming decades, making it one of the most important issues in the Sustainable Development Goals defined by the United Nations. Satellite-based mapping can provide valuable information about slums where insights about the location and size are still missing. Large-scale slum mapping remains a challenge, fuzzy feature spaces between formal and informal settlements, significant imbalance of slum occurrences opposed to formal settlements, and various categories of multiple morphological slum features. We propose a transfer learned fully convolutional Xception network (XFCN), which is able to differentiate between formal built-up structures and the various categories of slums in high resolution satellite data. The XFCN is trained on a large sample of globally distributed slums, located in cities of,

show abstract

“…(2018), who report a mean patch size of informal settlements of 1.6 ha. Due to class imbalance in the data set, we additionally performed data augmentation on tiles with a significant fraction of UV to artificially increase the amount of UV image tiles (Stiller et al 2019). Our experimental set-up of the mapping procedure consists the tiles of the three training regions which are randomly split into a training (70%) and a validation (30%) dataset to perform independent performance evaluation (see Figure 3 for an illustration of the workflow).…”

Section: Preprocessingmentioning

confidence: 99%

Mapping urban villages using fully convolutional neural networks

Mast

Wei

Wurm

2020

Remote Sensing Letters

View full text Add to dashboard Cite

Urban villages are a characteristic settlement type characterized by preserving their morphological characteristics embedded in sharp contrast in modern, highrise developments found especially in fast growing urban agglomerations of China. They serve very important socioeconomic functions in terms of the provision of cheap housing for rural-urban migrants, but they are also considered controversial for local governments. Due to the unprecedented pace of urban growth, especially in the Pearl River Delta region (PRD), up-to-date information on the size and location of urban villages are mostly missing. Large-area but highly detailed data from earth observation platforms can provide crucial information for mapping urban villages based on their characteristic morphologies. This study deploys fully convolutional neural networks for mapping urban villages in the city of Shenzhen. Results of the underlying experiments show that very high mapping accuracies of 84% can be achieved.

show abstract

Large-scale building extraction in very high-resolution aerial imagery using Mask R-CNN

Cited by 26 publications

References 14 publications

Number of Building Stories Estimation from Monocular Satellite Image Using a Modified Mask R-CNN

Number of Building Stories Estimation from Monocular Satellite Image Using a Modified Mask R-CNN

Satellite-Based Mapping of Urban Poverty With Transfer-Learned Slum Morphologies

Mapping urban villages using fully convolutional neural networks

Contact Info

Product

Resources

About