1. Mountain Image Analysis Suite (MIAS): A new plugin for converting oblique images to landcover maps in QGIS.
- Author
-
Wright, Claire, Bone, Chris, Mathews, Darcy, Tricker, James, Wright, Ben, and Higgs, Eric
- Subjects
- *
IMAGE analysis , *PLUG-ins (Computer programs) , *RELIEF models , *REMOTE-sensing images , *GEOGRAPHIC information systems - Abstract
The objective of this article is to present a novel GIS plugin for classifying and georeferencing high‐resolution oblique imagery with the intention of creating landcover datasets for spatial analysis. The Mountain Image Analysis Suite (MIAS) is a newly released plugin for the open‐source software, QGIS. MIAS was developed with images from Mountain Legacy Project, the world's largest systematic collection of high‐resolution mountain images. It works with both grayscale and color imagery, including historical images that predate aerial and satellite imagery. MIAS encompasses four tools for classifying and georeferencing oblique images. The plugin accesses pretrained deep learning models from a PyTorch‐based segmentation network to automate the classification of landcover in oblique images. Monoplotting is accomplished through the construction of a virtual photograph simulating the view from the camera using a shaded relief model. Once the virtual photograph is produced, the user aligns the classified image to the virtual photograph using a set of control points. This allows the creation of a classified and georeferenced raster representing the landcover for the area visible in the original oblique image. Similar workflows to the one contained in MIAS have been used for landcover mapping with oblique images to a high level of accuracy. However, MIAS is the first piece of software to bring all stages of image analysis into a single platform. MIAS has many applications across diverse fields such as mountain research, ecological restoration, community‐based mapping, environmental planning, and more. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF