I have 5 Landsat8 images from 5 different time periods, I'm trying to use a different method other than post classification method. Basically I want to first combine the data and then classify. However, I don't know how to do that exactly in envi.
Could you please clarify your question? It seems a bit confusing. You have 5 images with 5 different time periods and you want to combine them???
Dear Can,
Hi, please check my paper
https://www.researchgate.net/publication/271704516_Possible_field_crops_mapping_and_assessment_in_Lebanon_using_SENTINEL-2_images_and_GIS
Although it is pointing to Sentinel 2 which was not available at that time. But, It will show you how to collect satellite images and then correct them and convert them to NDVI, then stack them in chronological order to do crop mapping. You can use graphs and expert opinion to identify crops or you can do direct classification.
Conference Paper Possible field crops mapping and assessment in Lebanon using...
Hi Can,
Regarding to your task, you firstly need to address the following concerns:
1. what is the main target object you want to analyze, e.g. LULC, NDVI, water quality or else?
2. what are temporal differences between these 5 images? Some of objects, such as vegetations, may vary significantly at different seasons (i.e., seasonality effect).
I'd suggest you perform the classification at first and then combine these results in a Bayesian framework.
Hope it helps,
Best!
Kaixu Bai
You must have co-registered data. You must have images calibrated in the same way. And it is possible to pre-filter images if noise in them is intensive.
- Badges
- Science topic
- Similar topics
- Geoscience
- Remote Sensing