Extracting Wetlands from Swiss Historical Maps with ConvolutionalNeural Networks

Jiao, Chenjing; Heitzler, Magnus; Hurni, Lorenz

doi:10.21862/avhm2020.03

Cited by 6 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, scientists rely on historical maps in order to unlock the cartographic information concerning wetland areas. The extraction techniques differ depending on the characteristics of the maps (e.g., quality, spatial scale, colorimetric scale), ranging from the most basic (manual or semi-automatic vectorisation of the features) to the more complex automatic raster-processing techniques (Baily, 2007;Chiang et al, 2014) that can integrate neural networks (Jiao et al, 2020).…”

Section: Methodsmentioning

confidence: 99%

Mapping Over 80 Years of Wetland Sensitivity to Human Intervention. The Spatial Dynamics of the Lakes and Wetlands of the Jijia-Iași Wetlands Ramsar Site in 1935–2018

Albulescu,

Minea,

Larion

et al. 2023

Journal of Environmental Engineering and Landscape Management

View full text Add to dashboard Cite

Wetlands are at the top of the list in terms of valuable ecosystem services, at the same time being one of the most sensitive ecosystems that are subject to human-induced changes. The delicate balance between their ever-changing waterlogged and dry areas, together with the associated rich flora and fauna, are easily disturbed by human drivers, which are also responsible for long-term land use conversions. This paper aims to 1) document the spatial dynamics of the wetland areas and lakes in the Jijia-Iași Wetlands Ramsar Site, using historical cartographic materials and modern land use data, and 2) pinpoint the anthropic drivers that shaped these dynamics, via a literature review. This Ramsar Site constitutes an eloquent example of a wetland landscape that was mainly shaped by human intervention. A timeline is used to put into perspective the resulting maps and synthesised findings, and several inferences regarding the past, present, and future sensitivity of the wetlands in the study area are formulated. In 1935–2018, the two analysed elements followed opposite trends: while the wetland areas decreased from 10.61% of the study area to 4.79%, the lake features increased in size, from 0.68% of the Jijia-Iași Wetlands to 10.84%. In order to explain these changes, anthropogenic interventions were divided into three types of management (detrimental high human pressure, beneficial high human pressure, and beneficial low human pressure). In the long term scientific works, the unbalanced dynamics proved to be beneficial to the environment, as the construction of water collections determined a biodiversity boost and the designation of the study area as a Ramsar Site. This paper complements the scientific work focusing on “the Delta of Moldavia”, highlighting the practical implications of the management strategies applied during the reference period.

show abstract

Section: Methodsmentioning

confidence: 99%

Mapping Over 80 Years of Wetland Sensitivity to Human Intervention. The Spatial Dynamics of the Lakes and Wetlands of the Jijia-Iași Wetlands Ramsar Site in 1935–2018

Albulescu,

Minea,

Larion

et al. 2023

Journal of Environmental Engineering and Landscape Management

View full text Add to dashboard Cite

show abstract

“…The Swiss Siegfried map is a comprehensive Swiss topographical map series published between 1872 and 1949 (Heitzler and Hurni, 2020;Jiao et al, 2020). The map series depicts geographical features including buildings, roads, railways, hydrological features, contour lines, toponyms, etc.…”

Section: Siegfried Map and The Symbols In The Black Layermentioning

confidence: 99%

A fast and effective deep learning approach for road extraction from historical maps by automatically generating training data with symbol reconstruction

Jiao

Heitzler

Hurni

2022

International Journal of Applied Earth Observation and Geoinfor

View full text Add to dashboard Cite

“…Supervised machine learning algorithms, such as Deep Convolutional Neural Networks (CNN) have proved to be more effective for such object detection tasks than conventional segmentation methods. CNN-s have been used in various map vectorization applications, such as automatic label extraction (Laumer et al 2020), wetland extraction (Jiao, Heitzler, and Hurni 2020) and improved image segmentation by predictions of areal symbol locations (Groom et al 2020). Saeedimoghaddam and Stepinski used CNN to detect road intersection points and achieved an average of 90% accuracy, with 82% of intersection points extracted (Saeedimoghaddam and Stepinski 2020).…”

Section: Introductionmentioning

confidence: 99%

Automatic vectorization of point symbols on archive maps using deep convolutional neural network

Vassányi

Gede

2021

Proc. Int. Cartogr. Assoc.

View full text Add to dashboard Cite

Abstract. Archive topographical maps are a key source of geographical information from past ages, which can be valuable for several science fields. Since manual digitization is usually slow and takes much human resource, automatic methods are preferred, such as deep learning algorithms. Although automatic vectorization is a common problem, there have been few approaches regarding point symbols. In this paper, a point symbol vectorization method is proposed, which was tested on Third Military Survey map sheets using a Mask Regional Convolutional Neural Network (MRCNN). The MRCNN implementation uses the ResNet101 network improved with the Feature Pyramid Network architecture and is developed in a Google Colab environment. The pretrained network was trained on four point symbol categories simultaneously. Results show 90% accuracy, while 94% of symbols detected for some categories on the complete test sheet.

show abstract

Extracting Wetlands from Swiss Historical Maps with ConvolutionalNeural Networks

Cited by 6 publications

References 0 publications

Mapping Over 80 Years of Wetland Sensitivity to Human Intervention. The Spatial Dynamics of the Lakes and Wetlands of the Jijia-Iași Wetlands Ramsar Site in 1935–2018

Mapping Over 80 Years of Wetland Sensitivity to Human Intervention. The Spatial Dynamics of the Lakes and Wetlands of the Jijia-Iași Wetlands Ramsar Site in 1935–2018

A fast and effective deep learning approach for road extraction from historical maps by automatically generating training data with symbol reconstruction

Automatic vectorization of point symbols on archive maps using deep convolutional neural network

Contact Info

Product

Resources

About