Exploiting ConvNet Diversity for Flooding Identification

Nogueira, Keiller; Fadel, Samuel G.; Dourado, Ícaro Cavalcante; Werneck, Rafael de Oliveira; Muñoz, Javier Alvaro Vargas; Penatti, Otávio Augusto Bizetto; Calumby, Rodrigo Tripodi; Li, Lin Tzy; Santos, Jefersson A. dos; Torres, Ricardo da S.

doi:10.1109/lgrs.2018.2845549

Cited by 67 publications

(45 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Semantic segmentation allows the creation of thematic maps aiming to help in the comprehension of a scene [3]. In fact, semantic labeling has been an essential task for the remote sensing community [4] given that its outcome, the thematic map, generates essential and useful information capable of assisting in the decision making of a wide range of fields, including environmental monitoring, intelligent agriculture [5], disaster relief [6], [7], urban planning [8]. K In the top case, while smaller contexts may not provide enough information for the understanding of the scene, a large context brings more information that may help the model to identify that it is a road with a car on it.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Multicontext Segmentation of Remote Sensing Images Based on Convolutional Networks

Nogueira

Mura

Chanussot

et al. 2019

IEEE Trans. Geosci. Remote Sensing

Self Cite

127

View full text Add to dashboard Cite

Semantic segmentation requires methods capable of learning high-level features while dealing with large volume of data. Towards such goal, Convolutional Networks can learn specific and adaptable features based on the data. However, these networks are not capable of processing a whole remote sensing image, given its huge size. To overcome such limitation, the image is processed using fixed size patches. The definition of the input patch size is usually performed empirically (evaluating several sizes) or imposed (by network constraint). Both strategies suffer from drawbacks and could not lead to the best patch size. To alleviate this problem, several works exploited multi-context information by combining networks or layers. This process increases the number of parameters resulting in a more difficult model to train. In this work, we propose a novel technique to perform semantic segmentation of remote sensing images that exploits a multi-context paradigm without increasing the number of parameters while defining, in training time, the best patch size. The main idea is to train a dilated network with distinct patch sizes, allowing it to capture multi-context characteristics from heterogeneous contexts. While processing these varying patches, the network provides a score for each patch size, helping in the definition of the best size for the current scenario. A systematic evaluation of the proposed algorithm is conducted using four high-resolution remote sensing datasets with very distinct properties. Our results show that the proposed algorithm provides improvements in pixelwise classification accuracy when compared to state-of-the-art methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Dynamic Multicontext Segmentation of Remote Sensing Images Based on Convolutional Networks

Nogueira

Mura

Chanussot

et al. 2019

IEEE Trans. Geosci. Remote Sensing

Self Cite

127

View full text Add to dashboard Cite

show abstract

“…The study demonstrated efficient classification using CNN methods, but it was limited to a single type of image with RGB channels for training the model. Nogueira et al [35] in their study proposed four deep network architecture based on dilated convolutions and deconvolutions layers to distinguish between flooded and non-flooded areas from high resolution remote sensing images. The study outperforms all baseline methods by 1% in terms of Jaccard Index for flooding detection in a new location (unseen by the model network during training).…”

Section: Introductionmentioning

confidence: 99%

Flood Mapping with Convolutional Neural Networks Using Spatio-Contextual Pixel Information

et al. 2019

View full text Add to dashboard Cite

Remote sensing technology in recent years has been regarded the most important source to provide substantial information for delineating the flooding extent to the disaster management authority. There have been numerous studies proposing mathematical or statistical classification models for flood mapping. However, conventional pixel-wise classifications methods rely on the exact match of the spectral signature to label the target pixel. In this study, we propose a fully convolutional neural networks (F-CNNs) classification model to map the flooding extent from Landsat satellite images. We utilised the spatial information from the neighbouring area of target pixel in classification. A total of 64 different models were generated and trained with a variable neighbourhood size of training samples and number of learnable filters. The training results revealed that the model trained with 3 × 3 neighbourhood sized training samples and with 32 convolutional filters achieved the best performance out of the experiments. A new set of different Landsat images covering flooded areas across Australia were used to evaluate the classification performance of the model. A comparison of our proposed classification model to the conventional support vector machines (SVM) classification model shows that the F-CNNs model was able to detect flooded areas more efficiently than the SVM classification model. For example, the F-CNNs model achieved a maximum precision rate (true positives) of 76.7% compared to 45.27% for SVM classification.

show abstract

“…However, the aforementioned models can not be directly utilized for geospatial object detection, because the properties of remote sensing images and natural images are different and the direct application of those models to remote sensing images is not optimal. Researchers have done a lot of work in applying CNN-based models to detect geospatial objects in remote sensing images and achieved remarkable consequences [4,[15][16][17][18][19][20][21][22][23][24][25]45]. For example, the work in [4] utilized a hyperregion proposal network (HRPN) and a cascade of boosted classifiers to detect vehicles in remote sensing images.…”

Section: Geospatial Object Detectionmentioning

confidence: 99%

A Novel Multi-Model Decision Fusion Network for Object Detection in Remote Sensing Images

Guo

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

Object detection in optical remote sensing images is still a challenging task because of the complexity of the images. The diversity and complexity of geospatial object appearance and the insufficient understanding of geospatial object spatial structure information are still the existing problems. In this paper, we propose a novel multi-model decision fusion framework which takes contextual information and multi-region features into account for addressing those problems. First, a contextual information fusion sub-network is designed to fuse both local contextual features and object-object relationship contextual features so as to deal with the problem of the diversity and complexity of geospatial object appearance. Second, a part-based multi-region fusion sub-network is constructed to merge multiple parts of an object for obtaining more spatial structure information about the object, which helps to handle the problem of the insufficient understanding of geospatial object spatial structure information. Finally, a decision fusion is made on all sub-networks to improve the stability and robustness of the model and achieve better detection performance. The experimental results on a publicly available ten class data set show that the proposed method is effective for geospatial object detection.

show abstract

Exploiting ConvNet Diversity for Flooding Identification

Cited by 67 publications

References 9 publications

Dynamic Multicontext Segmentation of Remote Sensing Images Based on Convolutional Networks

Dynamic Multicontext Segmentation of Remote Sensing Images Based on Convolutional Networks

Flood Mapping with Convolutional Neural Networks Using Spatio-Contextual Pixel Information

A Novel Multi-Model Decision Fusion Network for Object Detection in Remote Sensing Images

Contact Info

Product

Resources

About