Rapid building detection using machine learning

Cohen, Joseph; Ding, Wei; Kuhlman, Caitlin; Chen, Aijun; Di, Liping

doi:10.1007/s10489-016-0762-6

Cited by 37 publications

(17 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A classifier of MSCNs is designed for building detection in this study due to its capability of non-linear estimation and the robustness of object classification under complex backgrounds. Another classifier, named Random Forest, has been proven to perform efficiently in the classification of building and non-building regions in the literature [59], in which an experiment comparing Random Forest with MSCNs was conducted to test the effectiveness of the MSCN classifier. Multiple features were extracted to classify using Random Forest and compared to deep features.…”

Section: Comparisons Of Mscns and Random Forest Classifiermentioning

confidence: 99%

Building Extraction from UAV Images Jointly Using 6D-SLIC and Multiscale Siamese Convolutional Networks

Zhou

Chen

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

Automatic building extraction using a single data type, either 2D remotely-sensed images or light detection and ranging 3D point clouds, remains insufficient to accurately delineate building outlines for automatic mapping, despite active research in this area and the significant progress which has been achieved in the past decade. This paper presents an effective approach to extracting buildings from Unmanned Aerial Vehicle (UAV) images through the incorporation of superpixel segmentation and semantic recognition. A framework for building extraction is constructed by jointly using an improved Simple Linear Iterative Clustering (SLIC) algorithm and Multiscale Siamese Convolutional Networks (MSCNs). The SLIC algorithm, improved by additionally imposing a digital surface model for superpixel segmentation, namely 6D-SLIC, is suited for building boundary detection under building and image backgrounds with similar radiometric signatures. The proposed MSCNs, including a feature learning network and a binary decision network, are used to automatically learn a multiscale hierarchical feature representation and detect building objects under various complex backgrounds. In addition, a gamma-transform green leaf index is proposed to truncate vegetation superpixels for further processing to improve the robustness and efficiency of building detection, the Douglas–Peucker algorithm and iterative optimization are used to eliminate jagged details generated from small structures as a result of superpixel segmentation. In the experiments, the UAV datasets, including many buildings in urban and rural areas with irregular shapes and different heights and that are obscured by trees, are collected to evaluate the proposed method. The experimental results based on the qualitative and quantitative measures confirm the effectiveness and high accuracy of the proposed framework relative to the digitized results. The proposed framework performs better than state-of-the-art building extraction methods, given its higher values of recall, precision, and intersection over Union (IoU).

show abstract

Section: Comparisons Of Mscns and Random Forest Classifiermentioning

confidence: 99%

Building Extraction from UAV Images Jointly Using 6D-SLIC and Multiscale Siamese Convolutional Networks

Zhou

Chen

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Various building detection algorithms have been developed, with cited precisions ranging from 80-90% [16]- [18]. Accurate algorithms rely on a combination of feature extraction techniques and machine learning.…”

Section: Google Earth Engine Simulator For Obstacle Location Extmentioning

confidence: 99%

Environment-Aware Deployment of Wireless Drones Base Stations with Google Earth Simulator

French

Mozaffari

Eldosouky

et al. 2019

2019 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)

View full text Add to dashboard Cite

In this paper, a software-based simulator for the deployment of base station-equipped unmanned aerial vehicles (UAVs) in a cellular network is proposed. To this end, the Google Earth Engine platform and its included image processing functions are used to collect geospatial data and to identify obstacles that can disrupt the line-of-sight (LoS) communications between UAVs and ground users. Given such geographical information, three environment-aware optimal UAV deployment scenarios are investigated using the developed simulator. In the first scenario, the positions of UAVs are optimized such that the number of ground users covered by UAVs is maximized. In the second scenario, the minimum number of UAVs needed to provide full coverage for all ground users is determined. Finally, given the load requirements of the ground users, the total flight time (i.e., energy) that the UAVs need to completely serve the ground users is minimized. Simulation results using a real area of the Virginia Tech campus show that the proposed environment-aware drone deployment framework with Google Earth input significantly enhances the network performance in terms of coverage and energy consumption, compared to classical deployment approaches that do not exploit geographical information. In particular, the results show that the proposed approach yields a coverage enhancement by a factor of 2, and a 65% improvement in energy-efficiency. The results have also shown the existence of an optimal number of drones that leads to a maximum wireless coverage performance.

show abstract

“…Beside preventing rasterization artefacts along the edges, another effect of the canonical orientation is that, no rotation invariants need to be learned during training.. Furthermore, the relation between X coordinate and Y coordinate can be enhanced, which can simplify the learning process and increase the training accuracy in some degree (Cohen et al, 2016).…”

Section: Preparation Of Input Rastersmentioning

confidence: 99%

Application of Deep Learning for 3D building generalization

Filippovska

Schmidt

et al. 2019

Proc. Int. Cartogr. Assoc.

View full text Add to dashboard Cite

<p><strong>Abstract.</strong> The generalization of 3D buildings is a challenging task, which needs to consider geometry information, semantic content and topology relations of 3D buildings. Although many algorithms with detailed and reasonable designs have been developed for the 3D building generalization, there are still cases that could be further studied. As a fast-growing technique, Deep Learning has shown its ability to build complex concepts out of simpler concepts in many fields. Therefore, in this paper, Deep Learning is used to solve the regression (generalization of individual 3D building) and classification problems (classification of roof type) simultaneously. Firstly, the test dataset is generated and labelled with the generalization results as well as the classification of roof types. Buildings with saddleback, half-hip, and hip roof are selected as the experimental objects since their generalization results can be uniformly represented by a common vector which aims to meet the compatible representation of Deep Learning. Then, the pre-trained ResNet50 is used as the baseline network. The optimal model capacity is searched within an extensive ablation study in the framework of the building generalization problem. After that, a multi-task network is built by adding a branch of classification to the above network, which is in parallel with the generalization branch. In the process of training, the imbalance problems of tasks and classes are solved by adjusting their donations to the total loss function. It is found that less error rate is obtained after adding a classification branch. For the final results, two improved metrics are used to evaluate the generalization performance. The accuracy of generalization reached over 95% for horizontal information and 85% for height, while the accuracy of classification reached 100% on the test dataset.</p>

show abstract

Rapid building detection using machine learning

Cited by 37 publications

References 26 publications

Building Extraction from UAV Images Jointly Using 6D-SLIC and Multiscale Siamese Convolutional Networks

Building Extraction from UAV Images Jointly Using 6D-SLIC and Multiscale Siamese Convolutional Networks

Environment-Aware Deployment of Wireless Drones Base Stations with Google Earth Simulator

Application of Deep Learning for 3D building generalization

Contact Info

Product

Resources

About