Automatic Classification of Roof Shapes for Multicopter Emergency Landing Site Selection

Castagno, Jeremy D.; Atkins, Ella M.

doi:10.2514/6.2018-3977

Cited by 5 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although this is our first journal publication on roof classification, this paper extends our preliminary work presented at a conference [ 4 ] on both processing methods and application to diverse geographical and architectural environments. Specific contributions include: Over 4500 building roofs spanning three cities have been manually classified and archived with a satellite and LiDAR depth image pair.…”

Section: Introductionmentioning

confidence: 77%

“…Some geographic regions (e.g., Germany) are more likely to have a denser collection of labeled roof shapes through a higher volunteer involvement. Previous work by the authors relied upon pre-labeled roof shapes provided by the OSM database [ 4 ] in Witten, Germany. However, this paper broadens the categories of classifiable roof shapes as well as sampling from diverse regions including small to large city centers.…”

Section: Gis Data Processing Image Generation and Trainingmentioning

confidence: 99%

“… New “complex-flat” and “unknown” roof shape classes enable the machine classifier to distinguish flat roofs with infrastructure (e.g., air conditioning and water towers), unfamiliar roof shapes, and images of poor quality. This paper significantly reduces the set of outliers that previously required manual removal for training and test datasets (from 45% in [ 4 ] down to 5% in this paper). This paper’s test set accuracies represent a reasonable expectation of results when deployed in new areas.…”

Section: Introductionmentioning

confidence: 99%

“…This paper significantly reduces the set of outliers that previously required manual removal for training and test datasets (from 45% in [ 4 ] down to 5% in this paper). This paper’s test set accuracies represent a reasonable expectation of results when deployed in new areas.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Roof Shape Classification from LiDAR and Satellite Image Data Fusion Using Supervised Learning

Castagno

Atkins

2018

Sensors

Self Cite

View full text Add to dashboard Cite

Geographic information systems (GIS) provide accurate maps of terrain, roads, waterways, and building footprints and heights. Aircraft, particularly small unmanned aircraft systems (UAS), can exploit this and additional information such as building roof structure to improve navigation accuracy and safely perform contingency landings particularly in urban regions. However, building roof structure is not fully provided in maps. This paper proposes a method to automatically label building roof shape from publicly available GIS data. Satellite imagery and airborne LiDAR data are processed and manually labeled to create a diverse annotated roof image dataset for small to large urban cities. Multiple convolutional neural network (CNN) architectures are trained and tested, with the best performing networks providing a condensed feature set for support vector machine and decision tree classifiers. Satellite image and LiDAR data fusion is shown to provide greater classification accuracy than using either data type alone. Model confidence thresholds are adjusted leading to significant increases in models precision. Networks trained from roof data in Witten, Germany and Manhattan (New York City) are evaluated on independent data from these cities and Ann Arbor, Michigan.

show abstract

Section: Introductionmentioning

confidence: 77%

Section: Gis Data Processing Image Generation and Trainingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Roof Shape Classification from LiDAR and Satellite Image Data Fusion Using Supervised Learning

Castagno

Atkins

2018

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Assouline et al [20] experimented with raster features and geometric features from a DSM to label rooftops using a random forest classifier. Although classical machine learning-based algorithms show promising results, they invariably face computational complexity challenges caused by the high dimensionality of data sources [21].…”

Section: Pixel-wise Image Classificationmentioning

confidence: 99%

Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-Type Classification

et al. 2019

View full text Add to dashboard Cite

Various deep learning applications benefit from multi-task learning with multiple regression and classification objectives by taking advantage of the similarities between individual tasks. This can result in improved learning efficiency and prediction accuracy for the task-specific models compared to separately trained models. In this paper, we make an observation of such influences for important remote sensing applications like elevation model generation and semantic segmentation tasks from the stereo half-meter resolution satellite digital surface models (DSMs). Mainly, we aim to generate good-quality DSMs with complete, as well as accurate level of detail (LoD)2-like building forms and to assign an object class label to each pixel in the DSMs. For the label assignment task, we select the roof type classification problem to distinguish between flat, non-flat, and background pixels. To realize those tasks, we train a conditional generative adversarial network (cGAN) with an objective function based on least squares residuals and an auxiliary term based on normal vectors for further roof surface refinement. Besides, we investigate recently published deep learning architectures for both tasks and develop the final end-to-end network, which combines different models, as using them first separately, they provide the best results for their individual tasks.which the object belongs. Mainly, building footprint extraction or roof type classification is one of the most challenging, but important problems. It is common to use DSMs as input data for classification tasks regarding buildings [6,7], as depth information provides geometrical silhouettes and allows a better understanding of building forms. Although a vast amount of attempts have already been made on accurate pixel-wise classification [8,9], it still remains a challenging task in practice due to the wide variety of building appearances.In most cases, each task, e.g., depth image generation and pixel-wise image classification, is tackled independently, although they are closely connected. Solving those multiple tasks jointly can enhance the performance of each independent task, as well as speed up computation time. This observation leads to the advantages of multi-task (MT) learning. The approach of simultaneously improving the generalization performance of multiple outputs from a single input was applied to numerous machine learning techniques. As a promising concept for convolutional neural networks (CNNs), MT learning has been proven to leverage a variety of problems successfully, like classification and semantic segmentation [10] or classification and object detection [11]. Due to the fact that different tasks may conflict, MT learning is regarded as the optimization of MT loss, which minimizes a linear combination of contributed single-task loss functions.In this work, we aim to produce good-quality LoD2-like DSMs with realistic building geometries together with dense pixel-wise rooftop classification masks, defining multiple classes, like ground, flat, and ...

show abstract

Deep learning aided web-based procedural modelling of LOD2 city models

2023

View full text Add to dashboard Cite

Use cases such as shadow or solar potential analysis require the use of the LOD2 building models (Level of Detail 2) and the generation of the LOD2 models requires the proper generation of the roof geometries. In general, obtaining roof type information and succeeding generations of the LOD2 models requires expensive aerial surveys and time-consuming construction processes. In this study, a methodology to generate LOD2 building models using only 2D building footprints and aerial imagery is explained to overcome these challenges. Using this methodology, condominiums could be generated as 3D if condominium unit plans are provided as well. The roof type information has been obtained from an aerial image that covers the entire study area using a CNN (Convolutional Neural Network) model with an 89.9 % accuracy rate. Then, the roof geometries have been constructed procedurally by extending and implementing the Straight Skeleton (SS) algorithm for three main types of roofs: at, gable and hipped.These constructed roof geometries have been combined with LOD1 block models generated by extruding the 2D footprints according to the height attribute. The proposed methodology has been developed as a web-based solution utilizing RESTful web services with modern web technologies. In summary, the main novelty of the study is based on two contributions: the extension of the SS algorithm for the construction of roof geometries and the web-based generation of LOD2 building models.

show abstract

Automatic Classification of Roof Shapes for Multicopter Emergency Landing Site Selection

Cited by 5 publications

References 15 publications

Roof Shape Classification from LiDAR and Satellite Image Data Fusion Using Supervised Learning

Roof Shape Classification from LiDAR and Satellite Image Data Fusion Using Supervised Learning

Multi-Task cGAN for Simultaneous Spaceborne DSM Refinement and Roof-Type Classification

Deep learning aided web-based procedural modelling of LOD2 city models

Contact Info

Product

Resources

About