Published June 25, 2020
Submission deadline: 30 November 2020
With rapid advances in sensing technologies, a huge amount of geospatial data can now be collected from sensors such as cameras, multi- and hyper-spectral scanners, synthetic aperture radar (SAR), and laser scanners. The sensing platforms include satellites, aircraft, unmanned aerial/ground vehicles, boats, trains, cars, and humans for backpack-carried or handheld sensors. The geometric and semantic information derived from such datasets are critical for making informed decision and solving real world problems. However, how to accurately and reliably extract information from such datasets remains a changing topic in cartography and other geoinformation communities.
Over the past decade, deep learning (DL) has achieved remarkable success first in speech recognition, then image classification followed by many image-based applications, including remote sensing. A recent focus in the areas of remote sensing and computer vision is placed on processing 3D data especially deep learning for point clouds acquired by laser scanners or LiDAR. The International Cartographic Association (ICA) commission on sensor-driven mapping and the Canadian Journal of Remote Sensing aim at publishing a special issue on “Large-scale Machine Learning for Sensor-driven Mapping.” The main objective of this special issue is to promote the thematic research and development applications of deep learning approaches for a variety of mapping problems. The suggested approaches include design and efficient training of new deep learning architectures, end-to-end 3D modelling algorithms, generative adversarial learning, transfer learning, meta learning, semi-supervised, weakly-supervised and self-supervised learning, deep reinforcement learning, neural architecture search (NAS) and large-scale datasets relevant to deep learning. The applications include, but are not limited to:
- Cartographic information extraction from optical and SAR images as well as LiDAR point clouds;
- 3D reconstruction and modelling of the built environments from images and point clouds;
- Multi-sensor data fusion including optical-SAR fusion, point cloud-image fusion, pan-sharpening;
- Registration and segmentation of images and point clouds;
- Cartographic object recognition, classification, and change detection;
- Point clouds from stereo, panoramas, cellphone camera images, oblique photos and satellite images;
- High performance computing for large-scale images and point clouds;
- Large-scale urban modelling from aerial, terrestrial and mobile images and LiDAR point clouds;
- 2D floorplan generation and 3D modelling of indoor images and point clouds;
- Indoor localization and mapping using deep learning;
- Integration of indoor mapping with Building Information Modelling (BIM);
Research Papers, Review Papers, as well as Research Notes, are all welcome and invited.
Prospective authors must follow the regular guidelines of the CJRS-JCT available on the Taylor & Francis website at: https://www.tandfonline.com/action/authorSubmission?show=instructions&journalCode=ujrs20
The targeted publication date for the Special Issue is summer 2021. However, all papers will be published on line individually as soon as they are accepted. The manuscripts must be submitted by 30 November 2020 via the CJRS-JCT ScholarOne Manuscripts website under the Special Issue on “Large-scale Machine Learning for Sensor-driven Mapping” or “Deep Learning for Mapping” in short, at https://mc.manuscriptcentral.com/cjrs-jct/
Regular charges by paper ($400 USD for research or review paper, $300 USD for a research note) will apply for all published papers, as well as the standard peer-review process. Gold Open Access licences are also available.
CJRS-JCT Editor-in-Chief: Monique Bernier, INRS – Québec City, QC, Canada
Guest Editors – CJRS-JCT Special Issue on Large-scale Machine Learning for Sensor-driven Mapping:
|University of Calgary||Calgary, AB, Canada|
|University of Northern British Columbia||Prince George, BC, Canada|
|University of Waterloo||Waterloo, ON, Canada|