Deep learning to detect built cultural heritage from satellite imagery. - Spatial distribution and size of vernacular houses in Sumba, Indonesia -

Monna, Fabrice; Rolland, Tanguy; Denaire, Anthony; Navarro, Nicolas; Granjon, Ludovic; Barbé, Rémi; Chateau-Smith, Carmela

doi:10.1016/j.culher.2021.10.004

Cited by 26 publications

(12 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hanus and Evans (2015) proposed a method for semi-automated pond extraction from LiDAR data in tropical forests. Monna et al (2021) proposed a deep learning method to detect built cultural heritage from satellite imagery in tropical forests, but the canopy is not as dense as in our study area. Our study only uses the spectral or intensity information of the images, and additional work is needed to test whether textural variables will be helpful for such a mapping.…”

Section: Discussionmentioning

confidence: 93%

DETECTION OF MAYA SACBEOB (SACBES) USING OPTICAL AND SAR IMAGERY IN NORTHERN PETÉN, MEXICO

LaRocque

Leblon²,

Ek³

et al. 2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. The pre-Contact Maya peoples build roads (Sacbeob) to facilitate movement between and within cities and surroundings. Given the dense forest cover in the Lowland Maya region now known as Yucatan, ground surveys of sacbeob are time-consuming and challenging to perform. Remote sensing can be a good alternative as it offers the advantages of extensive regional coverage, zero disturbances to cultural resources, and an opportunity to acquire data in less accessible areas on a cost-effective basis. LiDAR technology is highly valuable to detect man-made structures, but this technology is costly and is time-consuming to acquire data over a large area. Satellite imagery presents an alternative for mapping large areas. Previous studies documented linear features that could represent sections of sacbeob between Calakmul and El Mirador using Landsat-5 TM green, red, and near-infrared images (Folan et al. 1995). This study used Landsat-7 and Landsat-8 optical images having more bands and radar imagery (Sentinel-1 C-VH&VV, and Alos-1 PalSAR L-HH&HV) to connect sacbeob mapped by Folan et al. (1995) between three ancient Maya cities (Calakmul, El Mirador, and Uxul). Our results suggest that radar images with the capacity to penetrate dense forest cover can contribute to the identification and documentation of ancient road systems. Such a study provides an ideal starting point for targeted ground verification.

show abstract

Section: Discussionmentioning

confidence: 93%

DETECTION OF MAYA SACBEOB (SACBES) USING OPTICAL AND SAR IMAGERY IN NORTHERN PETÉN, MEXICO

LaRocque

Leblon²,

Ek³

et al. 2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…Precision evaluates how well the model identifies the positive samples. High precision indicates that the model has a low number of false positives 13 . Recall evaluates how well the model detects the positive samples.…”

Section: Proposed Approach For Dispersed House Mappingmentioning

confidence: 99%

“…However, the research was limited by the need to involve over 6000 participants and the six-month duration required to annotate houses in rural areas of certain African countries. Monna et al 13 detected vernacular houses automatically in Sumba Island (Indonesia), using satellite imagery and several DL powerful architectures, such as Inception v2 14 and ResNet 50 15 . A weakness of this research is that they rely on satellite images obtained from Microsoft Bing, which may have limitations in terms of geometric and radiometric consistency across the study area.…”

Section: Introductionmentioning

confidence: 99%

Mapping dispersed houses in rural areas of Colombia by exploiting planet satellite images with convolutional neural networks

Arrechea-Castillo,

Solano-Correa,

Muñoz-Ordóñez

et al. 2023

Geospatial Informatics XIII

View full text Add to dashboard Cite

The Sustainable Development Goal (SDG) number 11 aims at making cities and human settlements more inclusive, safe, resilient, and sustainable. Complying with SDG 11 is a difficult task, especially when considering rural settlements where: (i) population settles in a dispersed manner; and (ii) geography complexity and social dynamics of the area make it difficult to monitor and capture data. One example of such areas can be found in the South-West of Colombia, in the Las Piedras River sub-basin. The National Administrative Department of Statistics in Colombia (DANE in Spanish) aims at mapping the population and houses in dispersed and difficult-to-access rural settlements in an accurate and continuous way. Nevertheless, there are several difficulties (derived from the in-situ way of collecting the data) that prevent such data from being generated. This research presents a methodology to carry out an updated mapping of rural areas with high spatial resolution data coming from PlanetScope (3 m). Such a mapping considers the dynamics of housing growth, focusing on dispersed and difficult-to-access rural settlements. To this aim, Convolutional Neural Networks (CNNs) are used together with PlanetScope data, allowing to account for average houses size (≥12 𝑚2) in the study area. Preliminary results show a detection accuracy above 95%, in average, according to geography complexity.

show abstract

“…Its potential for monitoring pedestrian, vehicular, and water traffic with the use of surveillance cameras has been widely demonstrated (Chen et al, 2021;Chen et al, 2020;Zhou et al, 2018). Similarly, a range of studies have demonstrated that deep learning can also be successfully applied to very high resolution satellite imagery for detection of a range of objects, including motorised traffic (Froidevaux et al, 2020), infrastructure (Monna et al, 2021), and wildlife (Duporge et al, 2021;Guirado et al, 2019). Nevertheless, ship detection has so far focused heavily on the marine (Zhang et al, 2021;Nie et al, 2020) and lacustrine environments (Duan et al, 2019), with limited efforts, largely based on thresholding, within riverine and estuarine settings (Zhang et al, 2019;Gruel et al, 2022;Gruel and Latrubesse, 2021;Hackney et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Monitoring riverine traffic from space: The untapped potential of remote sensing for measuring human footprint on inland waterways

Śmigaj

Hackney

Diem

et al. 2023

Science of The Total Environment

View full text Add to dashboard Cite

Deep learning to detect built cultural heritage from satellite imagery. - Spatial distribution and size of vernacular houses in Sumba, Indonesia -

Cited by 26 publications

References 31 publications

DETECTION OF MAYA SACBEOB (SACBES) USING OPTICAL AND SAR IMAGERY IN NORTHERN PETÉN, MEXICO

DETECTION OF MAYA SACBEOB (SACBES) USING OPTICAL AND SAR IMAGERY IN NORTHERN PETÉN, MEXICO

Mapping dispersed houses in rural areas of Colombia by exploiting planet satellite images with convolutional neural networks

Monitoring riverine traffic from space: The untapped potential of remote sensing for measuring human footprint on inland waterways

Contact Info

Product

Resources

About