Detecting the number of buildings in a single high-resolution SAR image

Cao, Ye; Su, Chen; Yang, Guangbin

doi:10.5721/eujrs20144729

Cited by 10 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some researchers first detect bright-line segments and regularly spaced point-like features and subsequently group them into building regions [9,22,28,29,30]. Alternatively, several studies directly extract building regions of layovers and shadows using segmentation algorithms, such as the mean shift algorithm and conditional random field (CRF) [31], marker controlled watershed algorithm [17], thresholding and morphological operations [32]. Besides data-driven approaches, model-based methods are conducted.…”

Section: Related Workmentioning

confidence: 99%

Large-scale Building Height Retrieval from Single SAR Imagery based on Bounding Box Regression Networks

Sun¹,

Mou²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Building height retrieval from synthetic aperture radar (SAR) imagery is of great importance for urban applications, yet highly challenging owing to the complexity of SAR data. This paper addresses the issue of building height retrieval in large-scale urban areas from a single TerraSAR-X spotlight or stripmap image. Based on the radar viewing geometry, we propose that this problem can be formulated as a bounding box regression problem and therefore allows for integrating height data from multiple data sources in generating ground truth on a larger scale. We introduce building footprints from Geographic information system (GIS) data as complementary information and propose a bounding box regression network that exploits the location relationship between a building's footprint and its bounding box, allowing for fast computation. This is important for large-scale applications. The method is validated on four urban data sets using TerraSAR-X images in both high-resolution spotlight and stripmap modes. Experimental results show that the proposed network can reduce the computation cost significantly while keeping the height accuracy of individual buildings compared to a Faster R-CNN based method. Moreover, we investigate the impact of inaccurate GIS data on our proposed network, and this study shows that the bounding box regression network is robust against positioning errors in GIS data. The proposed method has great potential to be applied to regional or even global scales.

show abstract

Section: Related Workmentioning

confidence: 99%

Large-scale Building Height Retrieval from Single SAR Imagery based on Bounding Box Regression Networks

Sun¹,

Mou²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Zhao et al [8] proposed a building detection method based on marker controlled watershed algorithm. A similar approach that exploited layover and double bounce echoes to detect and determine the number of buildings from a single high resolution image was provided in [9]. Ferro et al [10] also developed a method that was primarily based on extracting a set of low-level bright (lines) and dark (shadows) primitives.…”

Section: Introductionmentioning

confidence: 99%

Buildings Detection in VHR SAR Images Using Fully Convolution Neural Networks

Shahzad

Maurer

Fraundorfer

et al. 2019

IEEE Trans. Geosci. Remote Sensing

103

View full text Add to dashboard Cite

This is the pre-acceptance version, to read the final version please go to IEEE Transactions on Geoscience and Remote Sensing on IEEE Xplore. This paper addresses the highly challenging problem of automatically detecting man-made structures especially buildings in very high resolution (VHR) synthetic aperture radar (SAR) images. In this context, the paper has two major contributions: Firstly, it presents a novel and generic workflow that initially classifies the spaceborne TomoSAR point clouds − generated by processing VHR SAR image stacks using advanced interferometric techniques known as SAR tomography (TomoSAR) − into buildings and non-buildings with the aid of auxiliary information (i.e., either using openly available 2-D building footprints or adopting an optical image classification scheme) and later back project the extracted building points onto the SAR imaging coordinates to produce automatic largescale benchmark labelled (buildings/non-buildings) SAR datasets. Secondly, these labelled datasets (i.e., building masks) have been utilized to construct and train the state-of-the-art deep Fully Convolution Neural Networks with an additional Conditional Random Field represented as a Recurrent Neural Network to detect building regions in a single VHR SAR image. Such a cascaded formation has been successfully employed in computer vision and remote sensing fields for optical image classification but, to our knowledge, has not been applied to SAR images. The results of the building detection are illustrated and validated over a TerraSAR-X VHR spotlight SAR image covering approximately 39 km 2 − almost the whole city of Berlin − with mean pixel accuracies of around 93.84%.

show abstract

“…Built-up areas, such as buildings, infrastructures, or even construction sites, are characterized by high backscattering values, due to multiple reflections or the doublebounce effect [47]. Therefore, land consumption can increase the backscatter if the land cover is characterized by low or intermediate roughness, such as low-vegetation and bare soils before changing.…”

mentioning

confidence: 99%

Land Consumption Monitoring with SAR Data and Multispectral Indices

et al. 2021

View full text Add to dashboard Cite

Land consumption is the increase in artificial land cover, which is a major issue for environmental sustainability. In Italy, the Italian Institute for Environmental Protection and Research (ISPRA) and National System for Environmental Protection (SNPA) have the institutional duty to monitor land consumption yearly, through the photointerpretation of high-resolution images. This study intends to develop a methodology in order to produce maps of land consumption, by the use of the semi-automatic classification of multitemporal images, to reduce the effort of photointerpretation in detecting real changes. The developed methodology uses vegetation indices calculated over time series of images and decision rules. Three variants of the methodology were applied to detect the changes that occurred in Italy between the years 2018 and 2019, and the results were validated using ISPRA official data. The results show that the produced maps include large commission errors, but thanks to the developed methodology, the area to be photointerpreted was reduced to 7,300 km2 (2.4 % of Italian surface). The third variant of the methodology provided the highest detection of changes: 70.4% of the changes larger than 100 m2 (the pixel size) and over 84.0% of changes above 500 m2. Omissions are mainly related to single pixel changes, while larger changes are detected by at least one pixel in most of the cases. In conclusion, the developed methodology can improve the detection of land consumption, focusing photointerpretation work over selected areas detected automatically.

show abstract

Detecting the number of buildings in a single high-resolution SAR image

Cited by 10 publications

References 25 publications

Large-scale Building Height Retrieval from Single SAR Imagery based on Bounding Box Regression Networks

Large-scale Building Height Retrieval from Single SAR Imagery based on Bounding Box Regression Networks

Buildings Detection in VHR SAR Images Using Fully Convolution Neural Networks

Land Consumption Monitoring with SAR Data and Multispectral Indices

Contact Info

Product

Resources

About