I have a Corine 2000 land use data set and i want to update it using a Worldview Satellite Image with 8 channels.. WHat would be the best way you think?
From which year are the Worldview images? What do you mean by updating land use map - to reconstruct land use in 2000 or to detect land use change in period from 2000 to the year for which you have the imagery?
While your imagery must be newer than 2007, you must consider time delay and possible changes in land use that have occured.
Second, geometric resolution is very different. CORINE map has the resolution of 100 meters and minimum mapping unit is 100 hectares. Worldview MS imagery has the resolution of about 2 meters.
You should conduct land use classification of Worldview scenes (some softwares are Multispec, Semi-Automatic Classification Plugin for QGIS, ArcGIS). The CORINE map could be useful in selection of training data for your classification, but you must be sure that land use didn't change in the meantime. However, the high resolution of your imagery is alone sufficient for land use classification.
Hi!
There is a significant difference between the two images: CORINE has a medium spatial resolution (100m) and WORLDVIEW, a high spatial resolution (about 2m pixel width). This implies in the definition of what geographical objects you want to identify (level of detail) or, in other words, you need to choose a suitable scale for the phenomenons you want to classify/map (i.e.: you want the trees or the orchard?).
Then, you should segment the WORLDVIEW image to "create" an image that "fits" the scale and/or the geographical objects you chose to identify (maximum level of detail).
In the classification process, it is very likely that you find problems with the segregation of thematic classes with similar spectral responses. To solve that, you can choose a specific target (i.e.: bare soil), identify and use the areas mapped as a mask to identify the other thematic classes that may have similar spatial response.
I hope this can help you.
Best regards.
Will try the supervised classification then and will see what i get.. one more thing is that my area of interest is covered by 3 different images of different date.. Will there be a problem; I plan to do an atmospheric correction.. But will it be enough;
Yes, I think that you will have a problem, because if these dates are also in a different season, you' ll have different illumination (different pixel values). Also I don't think that the atmospheric correction will solve this problem. Atm. Cor. is used when want to take off the atmospheric effects (like scattering between radiation and atmospheric particles such us H20, CO2 etc) of your image and see the real information that your study area has.
Endorsing what Catherine said, probably you will need to do the same classification process for each image.
Then, you can use the larger or more comprehensive thematic image to avoid lose much time with the concatenation.
Success.
My images are recent. 2014-2015 ..Also i have some fieldwork data and i know what type of vegetation exists in several locations.. Thanks everyone for your advice. I will start from scratch with the classification only from the satellite images , and will only have th corine data as a supplementary tool for comparisons..
Cartosat-3 is an advanced Indian Earth Observation satellite built and developed by ISRO, which will replace the IRS series. It has a panchromatic resolution of 0.25 metres making it the imaging satellite with highest resolution in the world and Mx of 1 metre with a high quality resolution which is a major improvement from the previous payloads in the Cartosat series. Potential uses include weather mapping, cartography or army defence, and strategic applications.
- Badges
- Science topic
- Similar topics
- Geoscience
- Remote Sensing