IMG2nDSM: Height Estimation from Single Airborne RGB Images with Deep Learning

Karatsiolis, Savvas; Kamilaris, Andreas; Cole, Ian R.

doi:10.3390/rs13122417

Cited by 21 publications

(10 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Interestingly, neural networks (NN) predicting depth from single images also use object interactions, like shadows, to identify objects in the scene [37]. To infer the nDSMs, we use a model developed in previous work [38] that converts single RGB images to nDSMs and we pre-train it with the data used for the SRR task. The model inferring the nDSMs from RGB images uses a U-Net architecture [39] to compress an image into smaller representations that the model then decodes to form the elevation map.…”

Section: A the Ndsm-based Lossmentioning

confidence: 99%

“…The model inferring the nDSMs from RGB images uses a U-Net architecture [39] to compress an image into smaller representations that the model then decodes to form the elevation map. For a detailed model architecture and details regarding its training, we kindly refer the readers to [38]. Figure 1 shows examples of inferring the elevation map of an aerial image via the nDSM model.…”

Section: A the Ndsm-based Lossmentioning

confidence: 99%

See 1 more Smart Citation

Exploiting Digital Surface Models for Inferring Super-Resolution for Remotely Sensed Images

Karatsiolis

Padubidri

Kamilaris

2022

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

Despite the plethora of successful Super-Resolution Reconstruction (SRR) models applied to natural images, their application to remote sensing imagery tends to produce poor results. Remote sensing imagery is often more complicated than natural images and has its peculiarities such as being of lower resolution, it contains noise, and often depicting large textured surfaces. As a result, applying non-specialized SRR models like the Enhanced Super Resolution Generative Adversarial Network (ESRGAN) on remote sensing imagery results in artifacts and poor reconstructions. To address these problems, we propose a novel strategy for enabling an SRR model to output realistic remote sensing images: instead of relying on feature-space similarities as a perceptual loss, the model considers pixel-level information inferred from the normalized Digital Surface Model (nDSM) of the image. This allows the application of betterinformed updates during the training of the model which sources from a task (elevation map inference) that is closely related to remote sensing. Nonetheless, the nDSM auxiliary information is not required during production i.e., the model infers a superresolution image without additional data. We assess our model on two remotely sensed datasets of different spatial resolutions that also contain the DSMs of the images: the DFC2018 dataset and the dataset containing the national LiDAR fly-by of Luxembourg. We compare our model with ESRGAN and we show that it achieves better performance and does not introduce any artifacts in the results. In particular, the results for the high-resolution DFC2018 dataset are realistic and almost indistinguishable from the ground truth images.

show abstract

Section: A the Ndsm-based Lossmentioning

confidence: 99%

Section: A the Ndsm-based Lossmentioning

confidence: 99%

Exploiting Digital Surface Models for Inferring Super-Resolution for Remotely Sensed Images

Karatsiolis

Padubidri

Kamilaris

2022

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several estimation problems has been solved using deep learning approaches for buildings as shown in Table 1. The buildings heights estimation has been deeply investigated in several previous works Karatsiolis et al [36], Li et al [37], Liu et al [38], and Cao and Huang [39].…”

Section: Related Workmentioning

confidence: 99%

“…In Karatsiolis et al [36], the authors introduced a deep learning model that combines architectural characteristics extracted through U-NET supported with residual connections and learned the height estimation by mapping the aerial RGB images. They achieved an RMSE value of 1.6.…”

Section: Related Workmentioning

confidence: 99%

DCRN: An Optimized Deep Convolutional Regression Network for Building Orientation Angle Estimation in High-Resolution Satellite Images

Shahin

Almotairi

2021

Electronics

View full text Add to dashboard Cite

Recently, remote sensing satellite image analysis has received significant attention from geo-information scientists. However, the current geo-information systems lack automatic detection of several building characteristics inside the high-resolution satellite images. The accurate extraction of buildings characteristics helps the decision-makers to optimize urban planning and achieve better decisions. Furthermore, Building orientation angle is a very critical parameter in the accuracy of automated building detection algorithms. However, the traditional computer vision techniques lack accuracy, scalability, and robustness for building orientation angle detection. This paper proposes two different approaches to deep building orientation angle estimation in the high-resolution satellite image. Firstly, we propose a transfer deep learning approach for our estimation task. Secondly, we propose a novel optimized DCRN network consisting of pre-processing, scaled gradient layer, deep convolutional units, dropout layers, and regression end layer. The early proposed gradient layer helps the DCRN network to extract more helpful information and increase its performance. We have collected a building benchmark dataset that consists of building images in Riyadh city. The images used in the experiments are 15,190 buildings images. In our experiments, we have compared our proposed approaches and the other approaches in the literature. The proposed system has achieved the lowest root mean square error (RMSE) value of 1.24, the lowest mean absolute error (MAE) of 0.16, and the highest adjusted R-squared value of 0.99 using the RMS optimizer. The cost of processing time of our proposed DCRN architecture is 0.0113 ± 0.0141 s. Our proposed approach has proven its stability with the input building image contrast variation for all orientation angles. Our experimental results are promising, and it is suggested to be utilized in other building characteristics estimation tasks in high-resolution satellite images.

show abstract

“…Comparison of the results of current main methods in DFC2018 dataset.Because the img2ndsm experiment from[75] on the DFC2018 dataset uses different partition strategies of the training set and test set, we have conducted relevant experiments again according to the dataset partition method in relevant papers.…”

mentioning

confidence: 99%

Multi-Task Learning of Relative Height Estimation and Semantic Segmentation from Single Airborne RGB Images

Liu

Wang

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

The generation of topographic classification maps or relative heights from aerial or remote sensing images represents a crucial research tool in remote sensing. On the one hand, from auto-driving, three-dimensional city modeling, road design, and resource statistics to smart cities, each task requires relative height data and classification data of objects. On the other hand, most relative height data acquisition methods currently use multiple images. We find that relative height and geographic classification data can be mutually assisted through data distribution. In recent years, with the rapid development of artificial intelligence technology, it has become possible to estimate the relative height from a single image. It learns implicit mapping relationships in a data-driven manner that may not be explicitly available through mathematical modeling. On this basis, we propose a unified, in-depth learning structure that can generate both estimated relative height maps and semantically segmented maps and perform end-to-end training. Compared with the existing methods, our task is to perform both relative height estimation and semantic segmentation tasks simultaneously. We only need one picture to obtain the corresponding semantically segmented images and relative heights simultaneously. The model’s performance is much better than that of equivalent computational models. We also designed dynamic weights to enable the model to learn relative height estimation and semantic segmentation simultaneously. At the same time, we have conducted good experiments on existing datasets. The experimental results show that the proposed Transformer-based network architecture is suitable for relative height estimation tasks and vastly outperforms other state-of-the-art DL (Deep Learning) methods.

show abstract

IMG2nDSM: Height Estimation from Single Airborne RGB Images with Deep Learning

Cited by 21 publications

References 49 publications

Exploiting Digital Surface Models for Inferring Super-Resolution for Remotely Sensed Images

Exploiting Digital Surface Models for Inferring Super-Resolution for Remotely Sensed Images

DCRN: An Optimized Deep Convolutional Regression Network for Building Orientation Angle Estimation in High-Resolution Satellite Images

Multi-Task Learning of Relative Height Estimation and Semantic Segmentation from Single Airborne RGB Images

Contact Info

Product

Resources

About