Comparison of Different Convolutional Neural Network Architectures for Satellite Image Segmentation

Khryashchev, Vladimir; Ivanovsky, Leonid; Pavlov, Vladimir; Ostrovskaya, Anna; Rubtsov, Anton

doi:10.23919/fruct.2018.8588071

Cited by 27 publications

(9 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite the lack of training data, Deep networks have proven to outperform at extracting mid-and high-level abstract and discriminative semantic features from images. Recent studies indicate that the feature representations learned by CNNs are greatly effective in semantic segmentation (Long et al 2015;Khryashchev et al 2018).…”

Section: Deep Learning Approaches For Semantic Image Segmentationmentioning

confidence: 99%

Intelligent mapping of irrigated areas from Landsat 8 images using transfer learning

Benbahria

Sebari²,

Hajji³

et al. 2021

International Journal of Engineering and Geosciences

View full text Add to dashboard Cite

The lack of reliable and up-to-date data in developing countries is a major obstacle to sustainable development. In Morocco, where groundwater withdrawals by farmers are very intensive and informal, maps describing and monitoring the extension of irrigated areas are scarce and labor-intensive to obtain. In this paper a novel transfer learning algorithm is proposed to map irrigated areas at different stages of an agricultural cycle from Landsat 8 images. The results obtained displays satisfactory performance over traditional machine learning algorithms. On a small dataset, we initially tested three well known deep learning architectures (SegNet, DenseNet and Unet). The results obtained were not satisfactory. So, to get high performance, we rely on a transfer learning architecture combining UNet with ResNet50 backbone (trained on 2012 ILSVRC ImageNet dataset) as a baseline after a phase where different configurations were tested. In the first part of this study, we compared the use of three optimization methods: Adam and two variants of Stochastic Gradient Descent (SGD) associated with two techniques (Cyclical Learning Rate and Warm Restart) to find the optimal learning rate and then test the impact of data augmentation on the overall accuracies. Data augmentation had improved the overall accuracy for the three methods. Adam based method from 94% to 97% with mean IoU of 0,79 (for all land cover classes) and 0,86 for irrigated areas class. For SGD based methods, the overall accuracy had increased from 91% to 94% with mean IoU of 0,75 (for all land cover classes) and 0,82 for irrigated areas class. As we are interested in having irrigated areas maps at different key periods of the agricultural cycle, we also explored, in the second part of this study, the temporal generalization of the best model.

show abstract

Section: Deep Learning Approaches For Semantic Image Segmentationmentioning

confidence: 99%

Intelligent mapping of irrigated areas from Landsat 8 images using transfer learning

Benbahria

Sebari²,

Hajji³

et al. 2021

International Journal of Engineering and Geosciences

View full text Add to dashboard Cite

show abstract

“…Training network In the EO (Earth Observation) task that extracts the information from the satellite map, such as building segmentation and road segmentation, it was observed that U-Net [23] showed the best performance [24], [25], thus U-Net is used for learning. The numerical map mentioned in Section II is processed and used as labeled data.…”

Section: Hardware Configurationmentioning

confidence: 99%

BRM Localization: UAV Localization in GNSS-Denied Environments Based on Matching of Numerical Map and UAV Images

Choi

Myung

2020

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Localization is one of the most important technologies needed to use Unmanned Aerial Vehicles (UAVs) in actual fields. Currently, most UAVs use GNSS to estimate their position. Recently, there have been attacks that target the weaknesses of UAVs that use GNSS, such as interrupting GNSS signal to crash the UAVs or sending fake GNSS signals to hijack the UAVs. To avoid this kind of situation, this paper proposes an algorithm that deals with the localization problem of the UAV in GNSS-denied environments. We propose a localization method, named as BRM (Building Ratio Map based) localization, for a UAV by matching an existing numerical map with UAV images. The building area is extracted from the UAV images. The ratio of buildings that occupy in the corresponding image frame is calculated and matched with the building information on the numerical map. The position estimation is started in the range of several km 2 area, so that the position estimation can be performed without knowing the exact initial coordinate. Only freely available maps are used for training data set and matching the ground truth. Finally, we get real UAV images, IMU data, and GNSS data from UAV flight to show that the proposed method can achieve better performance than the conventional methods.

show abstract

“…This type of neural networks has a special architecture, aimed at fast and high-quality detection of various objects [1]. The research of work of developed models continues the research, which was provided in papers [7,9]. Keras library with Tensorflow framework as a backend was used for development of CNN.…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

“…As it is said in [7], U-Net consists of 2 parts: an encoder (left) and a decoder (right). The encoder is a neural network which has a typical CNN architecture.…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Application of Satellite Image Segmentation for Urban Planning Optimization

2019

Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

This article presents research results of a convolution neural network for building detection on high-resolution aerial images of Planet database. Jaccard index was used for analysis of the quality of machine learning algorithm. This index of similarity compares results of algorithms with real masks. The masks were sliced on smaller parts together with images before training of developed model. The convolution neural network was launched on NVIDIA DGX-1 supercomputer, which was provided by AIcenter of P.G Demidov Yaroslavl State University. The problem of building detection on satellite images can be put into practice for urban planning, building control, search of the best locations for outlets etc.

show abstract

Comparison of Different Convolutional Neural Network Architectures for Satellite Image Segmentation

Cited by 27 publications

References 6 publications

Intelligent mapping of irrigated areas from Landsat 8 images using transfer learning

Intelligent mapping of irrigated areas from Landsat 8 images using transfer learning

BRM Localization: UAV Localization in GNSS-Denied Environments Based on Matching of Numerical Map and UAV Images

Application of Satellite Image Segmentation for Urban Planning Optimization

Contact Info

Product

Resources

About