Brazildam: A Benchmark Dataset for Tailings Dam Detection

Ferreira, Edemir; Brito, Matheus; Balaniuk, Remis; Alvim, Mário S.; Santos, Jefersson A. dos

doi:10.5194/isprs-archives-xlii-3-w12-2020-343-2020

Cited by 5 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors of [15] suggest that an ensemble of Inception and ResNet modules is an effective architecture for land cover classification. Current remote sensing research does not fully exploit hyperparameter tuning to further improve these models; researchers have mainly considered optimizing a subset of hyperparameters using a parameter sweep approach [6,18]. The authors of [13] have considered AutoML for a specific application of high-throughput image-based plant phenotyping.…”

Section: Deep Learning In Remote Sensingmentioning

confidence: 99%

Automated Machine Learning for Satellite Data: Integrating Remote Sensing Pre-trained Models into AutoML Systems

Salinas

Baratchi

Rijn

et al. 2021

Machine Learning and Knowledge Discovery in Databases. Applied Data Science Track

View full text Add to dashboard Cite

Current AutoML systems have been benchmarked with traditional natural image datasets. Differences between satellite images and natural images (e.g., bit-wise resolution, the number, and type of spectral bands) and lack of labeled satellite images for training models, pose open questions about the applicability of current AutoML systems on satellite data. In this paper, we demonstrate how AutoML can be leveraged for classification tasks on satellite data. Specifically, we deploy the Auto-Keras system for image classification tasks and create two new variants, IMG-AK and RS-AK, for satellite image classification that respectively incorporate transfer learning using models pre-trained with (i) natural images (using ImageNet) and (ii) remote sensing datasets. For evaluation, we compared the performance of these variants against manually designed architectures on a benchmark set of 7 satellite datasets. Our results show that in 71% of the cases the AutoML systems outperformed the best previously proposed model, highlighting the usefulness of a customized satellite data search space in AutoML systems. Our RS-AK variant performed better than IMG-AK for small datasets with a limited amount of training data. Furthermore, it found the best automated model for the datasets composed of near-infrared, green, and red bands.

show abstract

Section: Deep Learning In Remote Sensingmentioning

confidence: 99%

Automated Machine Learning for Satellite Data: Integrating Remote Sensing Pre-trained Models into AutoML Systems

Salinas

Baratchi

Rijn

et al. 2021

Machine Learning and Knowledge Discovery in Databases. Applied Data Science Track

View full text Add to dashboard Cite

show abstract

“…Relating to the Sentinel-2 images acquisition, we followed exactly the same protocol that was proposed by [23]. In this protocol, Google Earth Engine [1] was used to download the data using the place's geographical coordinates.…”

Section: Methodology a Datasetsmentioning

confidence: 99%

Scene classification using a combination of aerial and ground images

Machado¹

2021

Anais Estendidos Da XXXIV Conference on Graphics, Patterns and Images (SIBRAPI Estendido 2021)

View full text Add to dashboard Cite

lt is undeniable that aerial images can provide useful information for a large variety of tasks, such as disaster relief, and urban planning. But, since these images only see the Earth from one point of view, some applications may benefit from complementary information provided by other perspective views of the scene, such as ground-level images. Despite a large number of public image repositories for both georeferenced photos and aerial images (such as Google Maps, and Street View), there is a lack of public datasets that allow studies that exploit the complementarity of aerial+ground imagery. Given this, we present two new publicly available datasets named AiRound and CV-BrCT. Using both, we tackled the scene classification task in 2 different scenarios. The first one has a fully-paired image set, while the second has missing samples. In both situations, we used deep learning and feature fusion algorithms. To handle missing samples, we proposed a content-based image retrieval framework.

show abstract

“…A visual inspection of each reference using Google Earth images was required to address this problem and correct the target coordinates. [35] used the same databases to create the BrazilDAM, a public dataset based on Sentinel-2 and Landsat-8 satellite images covering the tailings dams cataloged by the ANM.…”

Section: Data Acquisitionmentioning

confidence: 99%

Mining and Tailings Dam Detection In Satellite Imagery Using Deep Learning

Balaniuk¹,

Isupova²,

Reece³

2020

Preprint

View full text Add to dashboard Cite

This work explores the combination of free cloud computing, free open-source software, and deep learning methods to analyse a real, large-scale problem: the automatic country-wide identification and classification of surface mines and mining tailings dams in Brazil. Locations of officially registered mines and dams were obtained from the Brazilian government open data resource. Multispectral Sentinel-2 satellite imagery, obtained and processed at the Google Earth Engine platform, was used to train and test deep neural networks using the TensorFlow 2 API and Google Colab platform. Fully Convolutional Neural Networks were used in an innovative way, to search for unregistered ore mines and tailing dams in large areas of the Brazilian territory. The efficacy of the approach is demonstrated by the discovery of 263 mines that do not have an official mining concession. This exploratory work highlights the potential of a set of new technologies, freely available, for the construction of low cost data science tools that have high social impact. At the same time, it discusses and seeks to suggest practical solutions for the complex and serious problem of illegal mining and the proliferation of tailings dams, which pose high risks to the population and the environment, especially in developing countries. Code is made publicly available at: https://github.com/remis/mining-discovery-with-deep-learning.

show abstract

Brazildam: A Benchmark Dataset for Tailings Dam Detection

Cited by 5 publications

References 12 publications

Automated Machine Learning for Satellite Data: Integrating Remote Sensing Pre-trained Models into AutoML Systems

Automated Machine Learning for Satellite Data: Integrating Remote Sensing Pre-trained Models into AutoML Systems

Scene classification using a combination of aerial and ground images

Mining and Tailings Dam Detection In Satellite Imagery Using Deep Learning

Contact Info

Product

Resources

About