Building area extraction from the high spatial resolution remote sensing imagery

Shi, Wei; Mao, Zhengyuan; Liu, Jinqing

doi:10.1007/s12145-018-0355-5

Cited by 7 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another important step in the proposed approach is the incorporation of additional information for enhancement of available bands information. This is important for supporting better spectral characteristics, texture, and shape information in medium spatial resolution [11]. This is very much reasonable and important for obtaining better results in building extraction.…”

Section: Introductionmentioning

confidence: 92%

“…In most of the previous works in building extraction, high-resolution satellite images have been used, because there is more spatial information along with a higher texture and geometrical information [11] than medium spatial resolution multispectral satellite images. However, the high-resolution images only have three to four bands, which limits the probability of identifying the proper class of a pixel [12].…”

Section: Introductionmentioning

confidence: 99%

“…So, to better generalize the research for any area, the multi-spectral satellite images like Sentinel-2 have been used in this work. However, in the case of multi-spectral satellite images, the major challenge still seen in previous approaches comes from the side of spatial resolutions, the miss-classification of pixel values, proper segregation of building boundaries in closely situated buildings of an area, and a smaller number of training datasets [11,14]. So, to improve on the above-mentioned challenges under building extraction, the proposed approach uses the multi-spectral bands, which helps in discriminating a variety of features more clearly and reduces class confusion.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

6+: A Novel Approach for Building Extraction from a Medium Resolution Multi-Spectral Satellite

et al. 2022

View full text Add to dashboard Cite

For smart, sustainable cities and urban planning, building extraction through satellite images becomes a crucial activity. It is challenging in the medium spatial resolution. This work proposes a novel methodology named ‘6+’ for improving building extraction in 10 m medium spatial resolution multispectral satellite images. Data resources used are Sentinel-2A satellite images and OpenStreetMap (OSM). The proposed methodology merges the available high-resolution bands, super-resolved Short-Wave InfraRed (SWIR) bands, and an Enhanced Normalized Difference Impervious Surface Index (ENDISI) built-up index-based image to produce enhanced multispectral satellite images that contain additional information on impervious surfaces for improving building extraction results. The proposed methodology produces a novel building extraction dataset named ‘6+’. Another dataset named ‘6 band’ is also prepared for comparison by merging super-resolved bands 11 and 12 along with all the highest spatial resolution bands. The building ground truths are prepared using OSM shapefiles. The models specific for extracting buildings, i.e., BRRNet, JointNet, SegUnet, Dilated-ResUnet, and other Unet based encoder-decoder models with a backbone of various state-of-art image segmentation algorithms, are applied on both datasets. The comparative analyses of all models applied to the ‘6+’ dataset achieve a better performance in terms of F1-Score and Intersection over Union (IoU) than the ‘6 band’ dataset.

show abstract

Section: Introductionmentioning

confidence: 92%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

6+: A Novel Approach for Building Extraction from a Medium Resolution Multi-Spectral Satellite

et al. 2022

View full text Add to dashboard Cite

show abstract

“…While Y. Xu et al [18] were able to detect buildings with clear boundaries using U-net on very-high-resolution satellite imagery, these kinds of data using optical sensors unfortunately cannot capture Earth's surface when it is covered by clouds. The problem applies to some accurate building detection methods, which have a possibility to be extended for using in building change detection, that use the characteristic of the shape of building roofs from a very-high-resolution imagery [19][20][21]. As well as being hindered by cloud contamination, these images are also difficult to obtain especially when studying an event in the distant past.…”

Section: Introductionmentioning

confidence: 99%

CORN: An Alternative Way to Utilize Time-Series Data of SAR Images in Newly Built Construction Detection

et al. 2020

View full text Add to dashboard Cite

The limitations in obtaining sufficient datasets for training deep learning networks is preventing many applications from achieving accurate results, especially when detecting new constructions using time-series satellite imagery, since this requires at least two images of the same scene and it must contain new constructions in it. To tackle this problem, we introduce Chronological Order Reverse Network (CORN)-an architecture for detecting newly built constructions in time-series SAR images that does not require a large quantity of training data. The network uses two U-net adaptations to learn the changes between images from both Time 1-Time 2 and Time 2-Time 1 formats, which allows it to learn double the amount of changes in different perspectives. We trained the network with 2028 pairs of 256 × 256 pixel SAR images from ALOS-PALSAR, totaling 4056 pairs for the network to learn from, since it learns from both Time 1-Time 2 and Time 2-Time 1. As a result, the network can detect new constructions more accurately, especially at the building boundary, compared to the original U-net trained by the same amount of training data. The experiment also shows that the model trained with CORN can be used with images from Sentinel-1. The source code is available at https://github.com/Raveerat-titech/CORN. Remote Sens. 2020, 12, 990 2 of 25 Remote Sens. 2020, 12, 990 3 of 25 the same ground truth data. While normally, the detection of new buildings is supposed to use the data in Time 1-Time 2 format, our proposed architecture takes both Time 1-Time 2 and Time 2-Time 1 formats of data to allow learning based on both of the changing features to make it more viable. With this architecture, the amount of training data of the network appears to be doubled. This allows the network to be trained with a greater variation of data, and can result an increased detection accuracy without having to use more SAR data or create any additional ground truths. Moreover, CORN has the potential to use SAR images from other satellites and other environments because the training back and forth causes the model to be more robust.In summary, the objective of this paper is to cope with the lack of training data when training deep learning networks, which leads to inaccurate results in newly built construction detection. We do so by proposing a network architecture called "CORN", which doubles the training set by reversing the chronological order of the dataset. CORN contains two U-net adaptations; one trains on Time 1-Time 2 images, and the other one trains on Time 2-Time 1 images. The proposed network not only increases the detection accuracy, but can also be used in a greater variety of settings of the data; specifically, images from other satellites and other acquisition conditions, including the terrain of the testing area.

show abstract

“…With the aim of building extraction, this method is used in this paper to generate rules automatically from remote sensing data without expert intervention. In the last decades, considerable attempts have been made to develop various methods for the detection of different types of objects in aerial and satellite images, especially buildings (Lhomme, He et al 2009, Ahmadi, Zoej et al 2010, Akçay and Aksoy 2010, Benedek, Descombes et al 2011, Stankov and He 2014, Sun, Pahlavani et al 2017, Xu, Wu et al 2018, Shi, Mao et al 2019. With the increasing availability and wide utilization of high resolution imagery, object-based image analysis (OBIA) has become a new approach or paradigm to classify or map satellite images into meaningful objects (Teo and Chen 2004, Walter 2004, Blaschke 2010, Pang, Hu et al 2014, Toure, Stow et al 2016, Li, Zhang et al 2018, Gavankar and Ghosh 2019.…”

Section: Introductionmentioning

confidence: 99%

Automatic Building Extraction Using a Decision Tree Object-Based Classification on Joint Use of Aerial and Lidar Data

Nejad

Pahlavani

Bigdeli

2019

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

View full text Add to dashboard Cite

Abstract. Updating digital maps is a challenging task that has been considered for many years and the requirement of up-to-date urban maps is universal. One of the main procedures used in updating digital maps and spatial databases is building extraction which is an active research topic in remote sensing and object-based image analysis (OBIA). Since in building extraction field a full automatic system is not yet operational and cannot be implemented in a single step, experts are used to define classification rules based on a complex and subjective “trial-and-error” process. In this paper, a decision tree classification method called, C4.5, was adopted to construct an automatic model for building extraction based on the remote sensing data. In this method, a set of rules was derived automatically then a rule-based classification is applied to the remote sensing data include aerial and lidar images. The results of experiments showed that the obtained rules have exceptional predictive performance.

show abstract

Building area extraction from the high spatial resolution remote sensing imagery

Cited by 7 publications

References 18 publications

6+: A Novel Approach for Building Extraction from a Medium Resolution Multi-Spectral Satellite

6+: A Novel Approach for Building Extraction from a Medium Resolution Multi-Spectral Satellite

CORN: An Alternative Way to Utilize Time-Series Data of SAR Images in Newly Built Construction Detection

Automatic Building Extraction Using a Decision Tree Object-Based Classification on Joint Use of Aerial and Lidar Data

Contact Info

Product

Resources

About