Deep Learning Approach for Rock Outcrops Identification

Kwok, Coco Yin Tung; Wong, Man Sing; Ho, Hung Chak; Lo, Frankie L.C.; Ko, Florence W.Y.

doi:10.1109/eorsa.2018.8598577

Cited by 5 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The advent of the Fourth Industrial Revolution stimulates the development of artificial intelligence, where geoscience studies have seamlessly adapted machine learning applications into their methodology (Caté et al 2017;Chen and Schuster 2020;Liu et al 2020;Feng et al 2021;Ramdani et al 2021). The study of carbonate outcrop analogs is perhaps one of the branches of Geoscience that does not see an extensive application of machine learning despite its potential (Francis et al 2014;Kirsch et al 2018;Kwok et al 2018;Marques Junior et al 2020;Ramdani et al 2021). Converting seismic images into outcrops can be viewed as an image-to-image translation problem.…”

Section: Methodology Overviewmentioning

confidence: 99%

Acoustic impedance to outcrop: Presenting near-surface seismic data as a virtual outcrop in carbonate analog studies

Ramdani

Perbawa

Puspita³

et al. 2022

The Leading Edge

View full text Add to dashboard Cite

Outcrop analogs play a central role in understanding subseismic interwell depositional facies heterogeneity of carbonate reservoirs. Outcrop geologists rarely utilize near-surface seismic data due to the limited vertical resolution and difficulty visualizing seismic signals as “band-limited rocks.” This study proposes a methodology using a combination of forward modeling and conditional generative adversarial network (cGAN) to translate seismic-derived acoustic impedance (AI) into a pseudo-high-resolution virtual outcrop. We tested the methodology on the Hanifa reservoir analog outcropping in Wadi Birk, Saudi Arabia. We interpret a 4 km long outcrop photomosaic from a digital outcrop model (DOM) for its depositional facies, populate the DOM with AI properties, and forward calculate the band-limited AI of the DOM facies using colored inversion. We pair the synthetic band-limited AI with DOM facies and train them using a cGAN. Similarly, we pair the DOM facies with outcrop photos and train them using a cGAN. We chain the two trained networks and apply them to the approximately 600 m long seismic-derived AI data acquired just behind the outcrop. The result translates AI images into a virtual outcrop “behind-the-outcrop” model. This virtual outcrop model is a visual medium that operates at a resolution and format more familiar to outcrop geologists. This model resolves subseismic stratigraphic features such as the intricate downlap-onlap stratal termination at scales of tens of centimeters and the outline of buildup facies, which are otherwise unresolvable in the band-limited AI.

show abstract

Section: Methodology Overviewmentioning

confidence: 99%

Acoustic impedance to outcrop: Presenting near-surface seismic data as a virtual outcrop in carbonate analog studies

Ramdani

Perbawa

Puspita³

et al. 2022

The Leading Edge

View full text Add to dashboard Cite

show abstract

“…Existing methods for the automatic detection of rocky outcrops are mostly based on satellite images (Kwok et al 2018), which can cause rocky outcrops under forest canopy to be overlooked, or on the delineation from DTM based only on slope values (Smith and Mullins 2021), which can miss small outcrops on flat areas or misinterpret rocky outcrops as steep (soil-)covered slopes. Therefore, we developed a new two-step approach for the quantitative detection of rocky outcrops (Fig.…”

Section: Rocky Outcrops Detectionmentioning

confidence: 99%

Revealing functional responses in habitat selection of rocky features and rugged terrain by Eurasian lynx (Lynx lynx) using LiDAR data

Čonč,

Oliveira,

Belotti

et al. 2024

Landsc Ecol

View full text Add to dashboard Cite

Context Many carnivores are attracted to rugged terrain, rocky areas, and conspicuous relief features. However, most of the previous research is limited to general topographical habitat characteristics and rarely consider the effects of microhabitat characteristics. Objectives We used the Eurasian lynx (Lynx lynx) as a model species to investigate the effects of microhabitat characteristics and human infrastructure on habitat selection. We also tested whether there is evidence for a functional response in habitat selection across a large gradient of habitat availability. Methods We developed a new approach for detecting rocky outcrops from airborne LiDAR data. In combination with other remote sensing techniques and GPS-telemetry data, we assessed lynx habitat selection and functional responses across two geologically contrasting areas in Europe. Results We detected > 1 million rocky outcrops and confirmed their strong selection by lynx. Lynx also selected steep, rugged, and rocky areas, especially for day-resting sites. Furthermore, lynx avoided paths during the day but selected them and other linear anthropogenic infrastructure during the night, indicating the behaviour-specific impact of human infrastructure. We also observed a functional response in the selection of rocky and rugged areas, as lynx’ selection of such habitats increased with their lower availability. This highlights the importance of preserving such terrains, especially when they are rare in a landscape. Conclusions Our results highlight the importance of incorporating remote sensing techniques and data on microhabitat features in animal habitat selection research. We also recommend caution when developing new infrastructure for human recreation or promoting its use near geomorphological features and in rugged terrain.

show abstract

“…They are primarily improving visual differences between lithologies by transforming spectral bands into separable combinations according to their diagnostic characteristics, or making a comparison of similarity in the spectrum between the collected remote sensing data and spectral reflectance measured in the laboratory. The second group of methods, namely machine learning, involving random forest [12], [18], support vector machine [19], [20], and neural network [21], [22], they provide a way to quantitatively describe the distribution of geological features compared with the traditional methods.…”

Section: Introductionmentioning

confidence: 99%

Mapping of Himalaya Leucogranites Based on ASTER and Sentinel-2A Datasets Using a Hybrid Method of Metric Learning and Random Forest

Wang¹,

Dong²

2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

The widely distributed leucogranite belt in the Himalayan orogeny is expected to have excellent potential for developing rare metal mineralization. Finding a way to effectively map the spatial distribution of leucogranites would be a significant contribution to rare metal exploration. Research shows remote sensing technology has long been recognized the significance in geological works, which greatly promoted mineral exploration in a cost-effective manner, especially in the Himalayan orogenic belt with poor natural environment. However, several challenges still exist in relation to the limited spectral band and spatial resolution of remote sensing images, as well as the onerous data processing. In this context, this study sought to resolve these two issues by applying a hybrid approach that comprises image fusion, metric learning, and random forest methods. For the first challenge, multisource and multisensor remote sensing data were integrated to provide more comprehensive spatial texture characteristics and spectral information. To address the second challenge, this study used a hybrid method of metric learning and random forest to promote computing efficiency and classification accuracy. This process is illustrated through a case study of lithological mapping in Cuonadong dome, the northern part of the Himalayan orogeny belt. Seven target lithological units were effectively discriminated with an 85.75% overall accuracy. This provides an important scientific basis for further exploration for rare metal deposits in the Himalayan orogeny belt, and a way of thinking for detecting geological features under harsh natural conditions.

show abstract

Deep Learning Approach for Rock Outcrops Identification

Cited by 5 publications

References 23 publications

Acoustic impedance to outcrop: Presenting near-surface seismic data as a virtual outcrop in carbonate analog studies

Acoustic impedance to outcrop: Presenting near-surface seismic data as a virtual outcrop in carbonate analog studies

Revealing functional responses in habitat selection of rocky features and rugged terrain by Eurasian lynx (Lynx lynx) using LiDAR data

Mapping of Himalaya Leucogranites Based on ASTER and Sentinel-2A Datasets Using a Hybrid Method of Metric Learning and Random Forest

Contact Info

Product

Resources

About