I want to develop several LULC maps of my study area for the last 30 years. But the fact is, there is no land-use data available for the previous years officially (except a few high-resolution Google Map). If I develop a site training dataset for land use classification using the old satellite images by interpretation from the local experts, will it be methodologically sound?
It will provide a first order approximation depending on how similar the same land use category area looks for the older image compared to the most recent one.
Probably you could interpret older imagery yourself with the same criteria you interpret the current ones. Please refer to our various papers on various retrogressive LULC change detections, including the below:
Article A comparison of fine resolution census and image-based natio...
Article Forecasting the pattern and pace of Fagus forest expansion i...
Hi,
The involvement of local experts will be justified, since each locality has its own direct and indirect photointerpretation features. This saves time on the job.
date is the limitation of used quality. Inventory of the date is security blanket. to provide more consistent better quality of basic date. in other to provide such a quality service. Personal build up inventory mind, who don't know about experience that are going to happen. the more you know the less inventory.
Hein Van Gils Thanks Sir, I am going through the works you have provided
Hi Gils, your assumption is general acceptable, while needs further evaluation by concrete method you adopted.
Hi everyone! Maybe I found something what I was searching for. Please have a look on the attachment if you are interested. Thanks.
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography
- Mapping
- Maps