Using historical maps in scientific studies : applications, challenges, and best practices

Using historical maps in scientific studies : applications, challenges, and best practices / Yao-Yi Chiang, Weiwei Duan, Stefan Leyk, Johannes H. Uhl, Craig A. Knoblock.

This book illustrates the first connection between the map user community and the developers of digital map processing technologies by providing several applications, challenges, and best practices in working with historical maps. After the introduction chapter, in this book, Chapter 2 presents a va...

Full description

Saved in:
Bibliographic Details
Main Authors: Chiang, Yao-Yi (Author), Duan, Weiwei (Author), Leyk, Stefan (Author), Uhl, Johannes H. (Author), Knoblock, Craig A., 1962- (Author)
Format: eBook
Language:English
Published: Cham, Switzerland : Springer, [2020]
Series:SpringerBriefs in geography,
Subjects:
Historical geography > Maps.
historical maps.
Historical geography
atlases.
Atlases
Maps
Atlases.
Maps.
Atlas.
Cartes géographiques.
Online Access:Click for online access
Table of Contents:
  • Intro; Acknowledgments; Contents; 1 Introduction; 1.1 Book Objectives; 1.2 Book Structure; References; 2 Historical Map Applications and Processing Technologies; 2.1 Introduction; 2.2 Applications of Historical Maps; 2.3 Case Studies of Map Processing Technologies; 2.3.1 Case Study I: Semi-Automatic Symbol Recognition from Map Scans; 2.3.1.1 SURF (Speeded Up Robust Features) Matching; 2.3.1.2 Histogram Matching; 2.3.1.3 Result Consolidation; 2.3.1.4 Results and Discussion; 2.3.2 Case Study II: Multi-Model, Context-Based Automatic Symbol Recognition from Map Scans
  • 2.3.2.1 Graphics Sampling Using Contextual Information2.3.2.2 Results and Discussion; 2.3.3 Case Study Discussion and Outlook; 2.4 Chapter Summary; References; 3 Creating Structured, Linked Geographic Data from Historical Maps: Challenges and Trends; 3.1 Introduction; 3.2 Finding Relevant Historical Maps; 3.3 Converting Map Content to Machine-Readable Formats and Record Uncertainty; 3.3.1 Crowdsourcing Approaches; 3.3.2 Semi-automatic Approaches; 3.3.3 Multi-Model, Context-Based, Automatic Approaches; 3.4 Modeling and Publishing Map Content; 3.5 Chapter Summary; References
  • 4 Training Deep Learning Models for Geographic Feature Recognition from Historical Maps4.1 Introduction; 4.2 Challenges in Using CNNs on Historical Maps; 4.2.1 Accurate Boundary Delineation of Geographic Features; 4.2.2 Scarce Training Data for Cartographic Documents; 4.3 Overview of Semantic Segmentation for Geographic Feature Recognition from Map Scans; 4.3.1 VGG16: The 16-layer Very Deep Convolutional Networks for Large-Scale Image Recognition; 4.3.2 GoogLeNet; 4.3.3 ResNet; 4.3.4 The Encoder and Decoder Architecture for Semantic Segmentation
  • 4.3.5 Multi-Scale Pyramids of Feature Images for Semantic Segmentation4.4 Overview of Transfer Learning for Geographic Feature Recognition from Map Scans; 4.5 Experiment; 4.5.1 Experimental Data, Settings and Evaluation Metrics; 4.5.2 Experiment I: The Impact of Backbone CNNs: FCN-VGG16, FCN-GoogLeNet, and FCN-ResNet; 4.5.3 Experiment II: The Impact of Transfer Learning Strategies: PSPNet; 4.5.3.1 Experiment III: Modified PSPNet; 4.6 Chapter Summary; References; 5 Summary and Discussion; 5.1 Book Summary; A Railroad Recognition Results

Similar Items