I am working on LULC change detection by applying change detection techniques like image differencing, image rationing, PCA, Tasseled Cap transformation which produce change or no-change binary output. After that I have done supervised classification of original image and then applied post classification. Now I want to do cross tabulation between the change detection results derived for different techniques against the post classification data. But I could not understand how to do?
Hi. Refer to this.
http://www.sciencedirect.com/science/article/pii/S1110982315000332
For analyse the change detection, you can use the ENVI software. there are two tools available for change detection analysis. otherwise you can apply statistical methods to find out the changes.
You can opt for spectral matching techniques to determine LULC changes as part of your post- classification processing. Nevertheless, following is some useful rersource what you are looking for:
http://www.sciencedirect.com/science/article/pii/S0273117712004218
http://www.ipublishing.co.in/jggsvol1no12010/EIJGGS1006.pdf
http://sedac.ciesin.columbia.edu/data/collection/lulc/methods
http://www.asprs.org/a/publications/pers/2007journal/september/2007_sep_1029-1040.pdf
Hope this helps.
In addition to Eic's work two examples of land cover change. Both proceed to a forecast. Have fun
Article What Drives Conversion of Tropical Forest in Carrasco Provin...
Article Forecasting the pattern and pace of Fagus forest expansion i...
Hi
you are looking simplest way to solve this problem, in my view , the " combine" operation in Arcmap is the simplest one.
Spatial analyse \ local\ combine
Good Luck
Dear Sridevi Jadav,
I suggest you to download the free software Map Comparison Kit (http://mck.riks.nl/) from which you can retrieve also the contingency table. A link to the publication in which we used the software is attached
I hope it could be useful for you.
http://www.tandfonline.com/doi/abs/10.1080/15230406.2015.1072736?journalCode=tcag20#.Vuk1GvnhDcs
Dear Jadav,
I think it is better to use spectral matching techniques or statistical methods
Regards,
You can use the Arcgis to cross information or multivariate correlation.
Hey! Post classification change detection is simple and most widely used method. However, it is very sensitive to the classification accuracies. You can use ARCGIS's 'Spatial Analyst Tools --> Zonal --> Tabulate Area' to accomplish the task. You can refer the https://www.researchgate.net/publication/283837494_Spatiotemporal_urban_expansion_in_Pune_metropolis_India_using_remote_sensing
to understand how the results can be interpreted.
All the best
Article Spatiotemporal urban expansion in Pune metropolis, India usi...
Use Erdas imagine Accuracy assessment tools to evaluate your LU/LC Data analysis
in any land cover change please.
You can have a look at this:
Remote Sensing of Environment 80 (2002) 185–201
Status of land cover classification accuracy assessment
Also, if you are workimg with binary data, ROC curves are good to assess accuracy.
An example:
Land-cover change model validation by an ROC method for the Ipswich watershed, Massachusetts, USA
Agriculture, Ecosystems & Environment
Volume 85, Issues 1–3, June 2001, Pages 239–248
I will advise you to same way: Erdas imagine Accuracy assessment tools to evaluate your LU/LC also you can use kappa index but it is not usefull at latest studies. You will use maximum likelihood algorithm and accuracy measures like 0.82, 0.90 ....
there is a lots of techniques to evaluate land use land cover change. over laying technique is one of them you can follow this link for a details methodology:
https://www.researchgate.net/publication/301797360_Analyzing_land_surface_temperature_distribution_in_response_to_land_useland_cover_change_using_split_window_algorithm_and_spectral_radiance_model_in_Sundarban_Biosphere_Reserve_India
Article Analyzing land surface temperature distribution in response ...
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India