Transform paper-based cadastral data into digital systems using GIS and end-to-end deep learning techniques

Abstract

Digital systems storing cadastral data in vector format are considered effective due to their ability of offering interactive services to citizens and other land-related systems. The adoption of such systems is ubiquitous, but when adopted, they create two non-compatible systems with paper-based cadastral systems whose information needs to be digitised. This study proposes a new approach that is fast and accurate for transforming paper-based cadastral data into digital systems. The proposed method involves deep-learning techniques of the LCNN and ResNet-50 for detecting cadastral parcels and their numbers, respectively, from the cadastral plans. It also contains four functions defined to speed up transformations and compilations of the cadastral plan’s data in digital systems. The LCNN is trained and validated with 968 samples. The ResNet-50 is trained and validated with 106,000 samples. The Structural-Average-Precision (${\mathit{sAP}}^{10}$) achieved with the LCNN was 0.9057. The Precision, Recall and F1-Score achieved with the ResNet-50 were 0.9650, 0.9648 and 0.9649, respectively. These results confirmed that the new method is accurate enough for implementation, and we tested it with a huge set of data from Tanzania. Its performance from the experimented data shows that the proposed method could effectively transform paper-based cadastral data into digital systems.

Keywords:

• Paper-based cadastral data
• digital cadastral systems
• geographical information systems
• artificial neural networks
• deep learning

Acknowledgement

Sincere thanks go to the Ministry of Lands, Housing and Human Settlements Development of Tanzania for allowing us to access the data used in this study. We also provide our sincere thanks to Hudson Magomba, the principal land surveyor of Bahi in Dodoma and Dr Alex Paul Lubida of the University of Dar es Salaam, Tanzania for a close follow up of the request of data.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data and codes availability statement

The data and codes that support the findings of this study are available in figshare.com and can be accessed using the following link, https://doi.org/10.6084/m9.figshare.20689459. The data (ie cadastral plans) is given as a sample from three regions by hiding some information because it is part of a legal document with individuals and government interests.

Additional information

Funding

This work is funded by the Chongqing Natural Science Foundation (Grant Number: CSTB2022NSCQMSX2069) and the Ministry of Education of China (Grant Number: 19JZD023)

Notes on contributors

Joseph Mango

Joseph Mango is a PhD holder from East China Normal University, Shanghai, and lecturer in the University of Dar es Salaam, Tanzania. His research interests include land surveying, spatiotemporal data modelling, GIS, Remote sensing and Machine learning.

Moyang Wang

Moyang Wang is a PhD student at School of Geographic Sciences of East China Normal University, Shanghai. Her research interests include image interpretation and deep learning.

Senlin Mu

Senlin Mu is a Master’s student at School of Geographic Sciences of East China Normal University, Shanghai. His research interests include GIS and deep reinforcement learning.

Di Zhang

Di Zhang is a PhD student at School of Geographic Sciences of East China Normal University, Shanghai. His research interests include computational geometry, spatio-temporal big data analysis and deep learning.

Jamila Ngondo

Jamila Ngondo is a Lecturer in the Department of Geography and Economics of the Dar es Salaam University College of Education, University of Dar es salaam, Tanzania. Her research interests include physical geography, GIS and remote sensing.

Regina Valerian-Peter

Regina Valerian-Peter is the Assistant Lecturer in Department of Geospatial Sciences and Technology, Ardhi University, Tanzania. Her research interests include geodesy, spatial data modelling, GIS and Remote sensing.

Christophe Claramunt

Christophe Claramunt is a professor of computer science at the French Naval Academy Research Institute. His research is oriented towards theoretical and computational issues of geographical information science and its applications to environmental urban and marine sciences.

Xiang Li

Xiang Li is a professor in the School of Geographic Sciences of East China Normal University. His research interests include spatio-temporal modelling, spatial optimization, and spatial big data application.