A Residual-Inception U-Net (RIU-Net) Approach and Comparisons with U-Shaped CNN and Transformer Models for Building Segmentation from High-Resolution Satellite Images

Sarıtürk, Batuhan; Şeker, Dursun Zafer

doi:10.3390/s22197624

Cited by 18 publications

(8 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The quality of model training was determining using the intersection over union (IoU) in relation to the intersection area and union area [18].…”

Section: Accuracy Assessmentmentioning

confidence: 99%

A modern method for building damage evaluation using deep learning approach - Case study: Flash flooding in Derna, Libya

Sellami,

Rhinane

2024

E3S Web Conf.

View full text Add to dashboard Cite

Year after year, floods become more and more a frequent and destructive force of nature, causing significant infrastructure losses and loss of life. An accurate and rapid assessment is required to determine the degree of contamination. The present study proposes a modern method for building damage assessment using deep learning during the flash flood of Derna, Libya. For this reason, we first exploited SAR satellite data, captured before and after the flood, to accurately determine the flood extent. Next, the footprint of affected buildings within this extent was extracted using a deep learning approach (U-Net model) based on high-resolution satellite imagery (30 cm) from MAXAR. Finally, an additional analysis was carried out using VIIRS VNP46A2 data (500 m spatial resolution) to analyse the night light assessment. The results demonstrate the effectiveness of this method, given that 5877 buildings were submerged by water and 2002 buildings were totally or partially destroyed. Also taking into account the estimated night light, Derna's power supply was reduced by over 90% after the floods. The suggested approach is an effective tool for comprehending the global effects of floods and aiding in relief efforts.

show abstract

“…The quality of model training was determining using the intersection over union (IoU) in relation to the intersection area and union area [18].…”

Section: Accuracy Assessmentmentioning

confidence: 99%

A modern method for building damage evaluation using deep learning approach - Case study: Flash flooding in Derna, Libya

Sellami,

Rhinane

2024

E3S Web Conf.

View full text Add to dashboard Cite

show abstract

“…Other works, such as [ 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 ], implemented what is known as Inception-like blocks, which use deep conventional networks, and the residual connections introduced in [ 26 ] to extract the outputs from each layer and concatenated them for the output, as shown in the example in Figure 4 , mimicking the operation using the Inception layer depth-wise by allowing the extracted feature at multiple receptive fields to be processed at the output layer. However, while this approach can be suitable for certain applications such as classification, a drop in the spatial features accumulates as we move deeper, diminishing the spatial accuracy of the larger LRF values, as illustrated in Figure 5 , where it can be seen that a bias towards features at the centre starts to increase, impairing the capability of the layer to accurately position where the feature is located and decreasing its efficiency in applications such as object detection.…”

Section: Width-based Layer Design (Inception and Inception-like Appro...mentioning

confidence: 99%

An Adaptive Kernels Layer for Deep Neural Networks Based on Spectral Analysis for Image Applications

Shoura

Leung

Balaji

2023

Sensors

View full text Add to dashboard Cite

As the pixel resolution of imaging equipment has grown larger, the images’ sizes and the number of pixels used to represent objects in images have increased accordingly, exposing an issue when dealing with larger images using the traditional deep learning models and methods, as they typically employ mechanisms such as increasing the models’ depth, which, while suitable for applications that have to be spatially invariant, such as image classification, causes issues for applications that relies on the location of the different features within the images such as object localization and change detection. This paper proposes an adaptive convolutional kernels layer (AKL) as an architecture that adjusts dynamically to images’ sizes in order to extract comparable spectral information from images of different sizes, improving the features’ spatial resolution without sacrificing the local receptive field (LRF) for various image applications, specifically those that are sensitive to objects and features locations, using the definition of Fourier transform and the relation between spectral analysis and convolution kernels. The proposed method is then tested using a Monte Carlo simulation to evaluate its performance in spectral information coverage across images of various sizes, validating its ability to maintain coverage of a ratio of the spectral domain with a variation of around 20% of the desired coverage ratio. Finally, the AKL is validated for various image applications compared to other architectures such as Inception and VGG, demonstrating its capability to match Inception v4 in image classification applications, and outperforms it as images grow larger, up to a 30% increase in accuracy in object localization for the same number of parameters.

show abstract

“…Fully Convolutional Networks (FCNs) are a class of deep learning models without classi cation level used to identify local regions and pixel level categories in images [55]. The emergence of FCNs marks the beginning of the transition from convolutional neural networks (CNN) to fully connected layers.…”

Section: Fcn Modelmentioning

confidence: 99%

Research on Deep Learning Based Genetic Intelligent Identification Method for Historical Buildings: A Case Study of Chinese Baroque Architecture in Harbin, China

Shao,

Sun

2023

Preprint

View full text Add to dashboard Cite

The protection of historical buildings is limited by low-quality style imitation and large-scale demolition and reconstruction, and the work process requires a high investment of human and material resources, which restricts the inheritance and development of this material cultural heritage. How to achieve precise monitoring and protection of historical building style is a key issue that needs to be urgently solved. The gene of historical architecture is the basic unit that controls the style of historical architecture. Identifying the gene of historical architecture is to identify the decisive factor that controls the style of historical architecture, which is of great significance for precise monitoring and protection of the style of historical architecture. At present, there are subjective qualitative, difficult to quantify, poor recognition accuracy, and low efficiency in reasoning and recognition in research on genetic identification of historical buildings. Therefore, this article takes the Chinese Baroque architecture in Harbin, China as an example, drawing on the principles of biological gene recognition, referring to the methods of architectural gene recognition in cultural geography and architecture, and using improved U-net models, traditional U-net models, FCN models, and Efficient Net models that incorporate channel attention mechanisms to intelligently identify historical building genes, obtaining the optimal intelligent recognition method for historical building genes based on deep learning. Research has shown that the accuracy of the improved U-net model incorporating channel attention mechanism is 0.69, which is 0.04, 0.07, and 0.01 higher than traditional U-net (0.65), FCN (0.62), and Efficient net (0.68), respectively. Therefore, the improved U-net model is the optimal method for intelligent identification of historical building genes. Research can provide new tools and methods for the intelligent identification of historical building genes.

show abstract

A Residual-Inception U-Net (RIU-Net) Approach and Comparisons with U-Shaped CNN and Transformer Models for Building Segmentation from High-Resolution Satellite Images

Cited by 18 publications

References 60 publications

A modern method for building damage evaluation using deep learning approach - Case study: Flash flooding in Derna, Libya

A modern method for building damage evaluation using deep learning approach - Case study: Flash flooding in Derna, Libya

An Adaptive Kernels Layer for Deep Neural Networks Based on Spectral Analysis for Image Applications

Research on Deep Learning Based Genetic Intelligent Identification Method for Historical Buildings: A Case Study of Chinese Baroque Architecture in Harbin, China

Contact Info

Product

Resources

About