Subsetting several S2 Images


I am analysing 10 S2 images for change detection analysis. I have 2 questions

  1. Is there any tool that I can use for subsetting all the images in SNAP (or what is the procedure to do that in a time saving way)?

  2. Is there any way to find out change detection analysis through SNAP? Already have used forest cover change processor, and didn’t work

Thanks in advance


You could use graph builder and then apply batch processing,

I think this is the only one available in SNAP,

1 Like

Thanks falahfakhri,

it is a great help for me. Using it now and trying to figure it out but still don’t know how can I use 10 or more images at the same time.

Do I have to same a .wkt file and load it every time when I do subset? but thanks anyway for the help


Open all your granular in Product Explorer, and then add all opened in the Batch processor, using the highlighted sign, or drag and drop those opened in the Batch processing.

I think you intend to subset the same AOI for all! in this case you could use google earth to delineate your AOI as wkt, and then paste it one time in here,

1 Like

Hello again, Thanks for your great help. I was trying to do subsetting through batch processing but couldn’t get the subset option like your window.

Facing a problem to run forest cover change. Would you please help me in this case

First you should create *xml file from gb, has the subset operator and then you could apply Batch processing.

if you take a look again at my previous post, I copied here again the Batch process has Mygraph.xml file which is loaded from the load tab.

1 Like

yup, Sorry that I have missed it. now it works :slight_smile: what about the forest change cover? did you see it?

Dose your two products have, these inputs,

Yes, I have 4 bands in my images but they are 2, 3, 4, and 8. My current source product and previous source product is ok

The error is already discussed in here,

Source of the post

Now it is showing another problem which I really don’t get cause I did my subsetting all together with 10 images simultaneously. So all have same raster size and the same pixel size as well

Did you subset the similar bands in each product?

Do they are included within one file of each product? (previous one file has four bands and current one file has four bands)

1 Like

Yes, I have done the same as you write. :slight_smile:

Could you please subset only the bands of each product, and don’t subset the AOI, and give try,

Give this try for only two products one current and one previous

Ok, let you know after that.

According to my previous post, Source of the post, your input of each product should has these bands,

In this case you should re- sample each product to the lowest resolution 10 m. of the red band,

I think the bands input causes the error, because your input is not match the requirement

But band 2,3,4 and 8 all are in 10 meter resolutions. Yes according to your previous post I didn’t chose band 11 and 12.

Don’t know but may be that could be a problem

I think yes, because firstly according to the statement it says "must "

And technically according to the operator, it compares too many indices, So inclusively many indices would be calculated using the aforementioned bands,

1 Like

Then it’s my bad. First, I thought they showed it for an example. so if I have 4 bands at least it will work. Second, I am analysing the images with these (2,3,4,and 8) bands only. So it will not work for me, I guess.

Will try QGIS then. Thanks. It was so helpful to chat with you. Come back later with some other problems :smiley: