Automated Building Detection from Airborne LiDAR and Very High-Resolution Aerial Imagery with Deep Neural Network

The detection of buildings in the city is essential in several geospatial domains and for decision-making regarding intelligence for city planning, tax collection, project management, revenue generation, and smart cities, among other areas. In the past, the classical approach used for building detection was by using the imagery and it entailed human–computer interaction, which was a daunting proposition. To tackle this task, a novel network based on an end-to-end deep learning framework is proposed to detect and classify buildings features. The proposed CNN has three parallel stream channels: the first is the high-resolution aerial imagery, while the second stream is the digital surface model (DSM). The third was fixed on extracting deep features using the fusion of channel one and channel two, respectively. Furthermore, the channel has eight group convolution blocks of 2D convolution with three max-pooling layers. The proposed model’s efficiency and dependability were tested on three different categories of complex urban building structures in the study area. Then, morphological operations were applied to the extracted building footprints to increase the uniformity of the building boundaries and produce improved building perimeters. Thus, our approach bridges a significant gap in detecting building objects in diverse environments; the overall accuracy (OA) and kappa coefficient of the proposed method are greater than 80% and 0.605, respectively. The findings support the proposed framework and methodologies’ efficacy and effectiveness at extracting buildings from complex environments.

Download Full-text

Urban Thematic Mapping by Integrating LiDAR Point Cloud with Colour Imagery

GEOMATICA ◽

10.5623/cig2011-060 ◽

2011 ◽

Vol 65 (4) ◽

pp. 375-385 ◽

Cited By ~ 2

Author(s):

Haiyan Guan ◽

Jonathan Li ◽

Michael A. Chapman

Keyword(s):

Urban Areas ◽

Point Cloud ◽

Local Variation ◽

Airborne Lidar ◽

Aerial Imagery ◽

Building Detection ◽

Thematic Mapping ◽

Airborne Lidar Data ◽

Spectral Channels ◽

Height Data

This paper presents an effective approach to integrating airborne lidar data and colour imagery acquired simultaneously for urban mapping. Texture and height information extracted from lidar point cloud is integrated with spectral channels of aerial imagery into an image segmentation process. Then, the segmented polygons are integrated with the extracted geometric features (height information between first- and lastreturn, eigenvalue-based local variation and filtered height data) and spectral features (line segments) into a supervised classifier. The results for two different urban areas in Toronto, Canada, demonstrated that a satisfactory overall accuracy of 84.96% and Kappa of 0.76 were achieved in Scene I, while a building detection rate of 92.11%, comission error of 2.10% and omission error of 9.25% were obtained in Scene II.

Download Full-text

Micro-terrain and canopy feature extraction by breakline and differencing analysis of gridded elevation models : identifying terrain model discontinuities with application to off-road mobility modeling

10.21079/11681/40185 ◽

2021 ◽

Author(s):

S. Blundell

Keyword(s):

High Resolution ◽

Route Planning ◽

Canopy Cover ◽

Airborne Lidar ◽

Added Value ◽

Surface Model ◽

Mobility Modeling ◽

Sparse Vegetation ◽

Elevation Model ◽

Terrain Features

Elevation models derived from high-resolution airborne lidar scanners provide an added dimension for identification and extraction of micro-terrain features characterized by topographic discontinuities or breaklines. Gridded digital surface models created from first-return lidar pulses are often combined with lidar-derived bare-earth models to extract vegetation features by model differencing. However, vegetative canopy can also be extracted from the digital surface model alone through breakline analysis by taking advantage of the fine-scale changes in slope that are detectable in high-resolution elevation models of canopy. The identification and mapping of canopy cover and micro-terrain features in areas of sparse vegetation is demonstrated with an elevation model for a region of western Montana, using algorithms for breaklines, elevation differencing, slope, terrain ruggedness, and breakline gradient direction. These algorithms were created at the U.S. Army Engineer Research Center – Geospatial Research Laboratory (ERDC-GRL) and can be accessed through an in-house tool constructed in the ENVI/IDL environment. After breakline processing, products from these algorithms are brought into a Geographic Information System as analytical layers and applied to a mobility routing model, demonstrating the effect of breaklines as obstacles in the calculation of optimal, off-road routes. Elevation model breakline analysis can serve as significant added value to micro-terrain feature and canopy mapping, obstacle identification, and route planning.

Download Full-text

Urban Building Extraction and Modeling Using GF-7 DLC and MUX Images

Remote Sensing ◽

10.3390/rs13173414 ◽

2021 ◽

Vol 13 (17) ◽

pp. 3414

Author(s):

Heng Luo ◽

Biao He ◽

Renzhong Guo ◽

Weixi Wang ◽

Xi Kuai ◽

...

Keyword(s):

High Resolution ◽

3D Modeling ◽

3D Visualization ◽

Smart Cities ◽

Three Dimensions ◽

Surface Model ◽

Building Extraction ◽

Double Line ◽

Building Model ◽

Urban Scene

Urban modeling and visualization are highly useful in the development of smart cities. Buildings are the most prominent features in the urban environment, and are necessary for urban decision support; thus, buildings should be modeled effectively and efficiently in three dimensions (3D). In this study, with the help of Gaofen-7 (GF-7) high-resolution stereo mapping satellite double-line camera (DLC) images and multispectral (MUX) images, the boundary of a building is segmented via a multilevel features fusion network (MFFN). A digital surface model (DSM) is generated to obtain the elevation of buildings. The building vector with height information is processed using a 3D modeling tool to create a white building model. The building model, DSM, and multispectral fused image are then imported into the Unreal Engine 4 (UE4) to complete the urban scene level, vividly rendered with environmental effects for urban visualization. The results of this study show that high accuracy of 95.29% is achieved in building extraction using our proposed method. Based on the extracted building vector and elevation information from the DSM, building 3D models can be efficiently created in Level of Details 1 (LOD1). Finally, the urban scene is produced for realistic 3D visualization. This study shows that high-resolution stereo mapping satellite images are useful in 3D modeling for urban buildings and can support the generation and visualization of urban scenes in a large area for different applications.

Download Full-text

On the Reliable Generation of 3D City Models from Open Data

Urban Science ◽

10.3390/urbansci4040047 ◽

2020 ◽

Vol 4 (4) ◽

pp. 47

Author(s):

Renoy Girindran ◽

Doreen S Boyd ◽

Julian Rosser ◽

Dhanya Vijayan ◽

Gavin Long ◽

...

Keyword(s):

High Resolution ◽

Open Source ◽

3D Model ◽

Airborne Lidar ◽

Data Availability ◽

Small Subset ◽

Surface Model ◽

High Resolution Data ◽

3D City Models ◽

City Models

A 3D model communicates more effectively than a 2D model, hence the applications of 3D city models are rapidly gaining significance in urban studies. However, presently, there is a dearth of free of cost, high-resolution 3D city models available for use. This paper offers potential solutions to this problem by providing a globally replicable methodology to generate low-cost 3D city models from open source 2D building data in conjunction with open satellite-based elevation datasets. Two geographically and morphologically different case studies were used to develop and test this methodology: the Chinese city of Shanghai and the city of Nottingham in the UK. The method is based principally on OpenStreetMap (OSM) and Advanced Land Observing Satellite World 3D digital surface model (AW3D DSM) data and use GMTED 2010 DTM data for undulating terrain. Further enhancement of the resultant 3D model, though not compulsory, uses higher resolution elevation models that are not always open source, but if available can be used (i.e., airborne LiDAR generated DTM). Further we test and develop methods to improve the accuracy of the generated 3D models, employing a small subset of high resolution data that are not open source but can be purchased with a minimal budgets. Given these scenarios of data availability are globally applicable and time-efficient for 3D building generation (where 2D building footprints are available), our proposed methodology has the potential to accelerate the production of 3D city models, and thus to facilitate their dependent applications (e.g., disaster management) wherever commercial 3D city models are unavailable.

Download Full-text

Using Unmanned Aerial Vehicle and LiDAR-Derived DEMs to Estimate Channels of Small Tributary Streams

Remote Sensing ◽

10.3390/rs13173380 ◽

2021 ◽

Vol 13 (17) ◽

pp. 3380 ◽

Cited By ~ 1

Author(s):

Joan Grau ◽

Kang Liang ◽

Jae Ogilvie ◽

Paul Arp ◽

Sheng Li ◽

...

Keyword(s):

High Resolution ◽

Unmanned Aerial Vehicle ◽

Slope Position ◽

Surface Model ◽

Stream Channel ◽

Low Resolution ◽

Stream Channels ◽

Aerial Vehicle ◽

Lidar Dem ◽

Small Tributary

Defining stream channels in a watershed is important for assessing freshwater habitat availability, complexity, and quality. However, mapping channels of small tributary streams becomes challenging due to frequent channel change and dense vegetation coverage. In this study, we used an Unmanned Aerial Vehicle (UAV) and photogrammetry method to obtain a 3D Digital Surface Model (DSM) to estimate the total in-stream channel and channel width within grazed riparian pastures. We used two methods to predict the stream channel boundary: the Slope Gradient (SG) and Vertical Slope Position (VSP). As a comparison, the same methods were also applied using low-resolution DEM, obtained with traditional photogrammetry (coarse resolution) and two more LiDAR-derived DEMs with different resolution. When using the SG method, the higher-resolution, UAV-derived DEM provided the best agreement with the field-validated area followed by the high-resolution LiDAR DEM, with Mean Squared Errors (MSE) of 1.81 m and 1.91 m, respectively. The LiDAR DEM collected at low resolution was able to predict the stream channel with a MSE of 3.33 m. Finally, the coarse DEM did not perform accurately and the MSE obtained was 26.76 m. On the other hand, when the VSP method was used we found that low-resolution LiDAR DEM performed the best followed by high-resolution LiDAR, with MSE values of 9.70 and 11.45 m, respectively. The MSE for the UAV-derived DEM was 15.12 m and for the coarse DEM was 20.78 m. We found that the UAV-derived DEM could be used to identify steep bank which could be used for mapping the hydrogeomorphology of lower order streams. Therefore, UAVs could be applied to efficiently map small stream channels in order to monitor the dynamic changes occurring in these ecosystems at a local scale. However, the VSP method should be used to map stream channels in small watersheds when high resolution DEM data is not available.

Download Full-text

Air Quality Estimation in Ukraine Using SDG 11.6.2 Indicator Assessment

10.20944/preprints202110.0299.v1 ◽

2021 ◽

Author(s):

Andrii Shelestov ◽

Hanna Yailymova ◽

Bohdan Yailymov ◽

Nataliia Kussul

Keyword(s):

European Union ◽

Air Quality ◽

High Resolution ◽

City Planning ◽

Smart Cities ◽

The European Union ◽

City Development ◽

European Cities ◽

Monitoring Service ◽

Functional Areas

Ukraine is an associate member of the European Union and in the coming years it is expected that all the data and services already used by European Union countries will become available for Ukraine. An important program, which is the basis for building European monitoring services for Smart Cities, is the Copernicus program. The two most important services of this program are Copernicus Land Monitoring Service (CLMS) and Copernicus Atmosphere Monitoring Service (CAMS). CLMS provides important information on Land Use in Europe. In the context of Smart Cities, the most valuable one is the Urban Atlas service, which is related to local CLMS services and provides a detailed digital city plan in vector form, which is segmented into small functional areas classified by the CORIN nomenclature. The Urban Atlas is a geospatial layer with high-resolution, which is built for all European cities with a population of more than 100,000 that combines high-resolution sat-ellite data, city segmentation by blocks and functional areas, important city infrastructure, etc. This product is used as a basis for city planning and obtaining analytics on the most important indicators of city development including air quality monitoring. For Ukraine, such geospatial products are not provided under the Copernicus program. It is important to start work on its development and implementation as early as possible, so that when the first city atlas appears, Ukraine will be ready to work with it together with the European community. This requires preparing the basis for na-tional research and training national stakeholders and users to use this product. To make this happen it’s necessary to have national geospatial product, which can be used as an analogue of the city atlas. In this article authors analyzed the existing methods of air quality assessment and assessment of the SDG indicator 11.6.2 achieving for European cities, based on which the indicator 11.6.2 for Ukraine for 5 years was evaluated for the first time. The obtained results are analyzed and the values of indicator 11.6.2 for Ukraine are compared with European countries.

Download Full-text

VERY HIGH RESOLUTION LAND USE AND LAND COVER MAPPING USING PLEIADES-1 STEREO IMAGERY AND MACHINE LEARNING

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xliii-b2-2020-675-2020 ◽

2020 ◽

Vol XLIII-B2-2020 ◽

pp. 675-682

Author(s):

D. James ◽

A. Collin ◽

A. Mury ◽

S. Costa

Keyword(s):

Land Use ◽

High Resolution ◽

Land Cover ◽

Large Scale ◽

Near Infrared ◽

Airborne Lidar ◽

Support Vector ◽

Surface Model ◽

Global Changes ◽

Very High

Abstract. Anthropocene is featured with increasing human population and global changes that strongly affect landscapes at an unprecedented pace. As a flagship, the coastal fringe is subject to an accelerated conversion of natural areas into agricultural ones, in turn, into urban ones, generating hazardous soil artificialization. Very high resolution (VHR) technologies such as airborne LiDAR or UAV imageries are good assets to model the topography and classify the land use/land cover (LULC), helping local management. Even if their spatial resolution suits with the management scale, their extent covers a few km2, making large-scale monitoring complex and time-consuming. VHR spaceborne imagery has a great potential to address this spatial challenge given its regional acquisition. This research proposes to evaluate the capabilities of a Pleiades-1 stereo-satellite multispectral imagery (blue, green, red, BGR, and near-infrared, NIR) to both model the surface topography and classify LULC. Horizontal and vertical accuracies of the photogrammetry-driven digital surface model (DSM) attain 0.53 m and 0.65 m, respectively. Nine LULC generic classes are studied using the maximum likelihood (ML) and support vector machine (SVM) algorithms. The classification accuracy of the basic BGR (reaching 84.64 % and 76.13 % with ML and SVM, respectively) is improved by the DSM contribution (5.49 % and 2.91 % for ML and SVM, respectively), and the NIR contribution (6.78 % and 3.89 % for ML and SVM, respectively). The gain of the DSM-NIR combination totals 8.91 % and 8.40 % for ML and SVM, respectively, making the ML-based full combination the best performance (93.55 %).

Download Full-text

Near-Field High-Resolution Maps of the Ridgecrest Earthquakes from Aerial Imagery

Seismological Research Letters ◽

10.1785/0220210234 ◽

2021 ◽

Author(s):

Alba M. Rodriguez Padilla ◽

Mercedes A. Quintana ◽

Ruth M. Prado ◽

Brian J. Aguilar ◽

Thomas A. Shea ◽

...

Keyword(s):

High Resolution ◽

Hazard Analysis ◽

Near Field ◽

Lateral Spreading ◽

Airborne Lidar ◽

Aerial Imagery ◽

Training Dataset ◽

Surface Rupture ◽

High Level ◽

Airborne Lidar Data

Abstract High-resolution maps of surface rupturing earthquakes are essential tools for quantifying rupture hazard, understanding the mechanics of rupture propagation, and interpreting evidence of past earthquakes in the landscape. We present highly detailed maps of five portions of the surface rupture of the 2019 Ridgecrest earthquakes, derived from 5 cm per pixel aerial imagery and 2–20 cm per pixel unmanned aerial vehicle imagery. Our high-resolution maps cover areas of complexity and distributed deformation, sections in which strain is very localized, and areas where the rupture breaks through sediment and bedrock, ensuring sampling of the diverse rupture styles of this earthquake sequence. These maps reveal the near-field deformation of the surface rupture with a high level of detail, resolving the extent of secondary fracturing, lateral spreading, and liquefaction features that are below the resolution of airborne lidar data, field mapping, and geodesy. These data may serve as a machine learning training dataset, and offer opportunities for detailed kinematic analysis and high-resolution probabilistic displacement hazard analysis.

Download Full-text

Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning

Remote Sensing ◽

10.3390/rs12223797 ◽

2020 ◽

Vol 12 (22) ◽

pp. 3797

Author(s):

David Radke ◽

Daniel Radke ◽

John Radke

Keyword(s):

Deep Learning ◽

High Resolution ◽

San Francisco ◽

Aerial Imagery ◽

Surface Model ◽

Vegetation Growth ◽

Vegetation Height ◽

Bay Area ◽

Model Complex ◽

Elevation Data

Measuring and monitoring the height of vegetation provides important insights into forest age and habitat quality. These are essential for the accuracy of applications that are highly reliant on up-to-date and accurate vegetation data. Current vegetation sensing practices involve ground survey, photogrammetry, synthetic aperture radar (SAR), and airborne light detection and ranging sensors (LiDAR). While these methods provide high resolution and accuracy, their hardware and collection effort prohibits highly recurrent and widespread collection. In response to the limitations of current methods, we designed Y-NET, a novel deep learning model to generate high resolution models of vegetation from highly recurrent multispectral aerial imagery and elevation data. Y-NET’s architecture uses convolutional layers to learn correlations between different input features and vegetation height, generating an accurate vegetation surface model (VSM) at 1×1 m resolution. We evaluated Y-NET on 235 km2 of the East San Francisco Bay Area and find that Y-NET achieves low error from LiDAR when tested on new locations. Y-NET also achieves an R2 of 0.83 and can effectively model complex vegetation through side-by-side visual comparisons. Furthermore, we show that Y-NET is able to identify instances of vegetation growth and mitigation by comparing aerial imagery and LiDAR collected at different times.

Download Full-text

AUTOMATIC EXTRACTION OF BUILDING OUTLINE FROM HIGH RESOLUTION AERIAL IMAGERY

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprsarchives-xli-b3-419-2016 ◽

2016 ◽

Vol XLI-B3 ◽

pp. 419-423 ◽

Cited By ~ 2

Author(s):

Yandong Wang

Keyword(s):

High Resolution ◽

Point Cloud ◽

Automatic Generation ◽

Aerial Imagery ◽

Building Detection ◽

New Approach ◽

Building Roof ◽

High Resolution Imagery ◽

Point Cloud Classification

In this paper, a new approach for automated extraction of building boundary from high resolution imagery is proposed. The proposed approach uses both geometric and spectral properties of a building to detect and locate buildings accurately. It consists of automatic generation of high quality point cloud from the imagery, building detection from point cloud, classification of building roof and generation of building outline. Point cloud is generated from the imagery automatically using semi-global image matching technology. Buildings are detected from the differential surface generated from the point cloud. Further classification of building roof is performed in order to generate accurate building outline. Finally classified building roof is converted into vector format. Numerous tests have been done on images in different locations and results are presented in the paper.

Download Full-text