Getting oil spill bit mask in snap from SLC product

[sjpsentinel1]

i have tried with varying window size, lowered the threshold shift and maintained the cluster size as 2sq.km.
i obtained the results as shown

is there any other ways to cover the zone with spill bit mask?
how can i extract the scatter/intensity plot for an area? i need to show the variation in the intensity of dark spots and nor- dark zones?
could you suggest me on the other feature interpretation , like wave, currents from this data ?

[ABraun]

You can apply stronger filters which enhance the dark areas befor the masking process.
You can draw a polygon and investigate the histogram of oil and non-oil areas.

[sjpsentinel1]

Getting oil zone from Non-oil Zone:
i took the layer "oil spill detected with mask "
i have created profile plot with sigma0 vs pixels

it is giving the bar chart as displayed. it gives/shows the values, when the cursor is moved.
In this profile plot,
1.How can i eliminate the shaded line? ie. the standard deviation
2. i need to draw, plot line(line graph), instead of bar chart, suggest me how i can do it ?
3.why it is considering only the masked zone, if i navigate the cursor out of the mask, it is showing NaN?
4.how can i merge, the chart with sigma band, so that , it should display the value of sigma0 on the polyline itself, i.e on the drawn polyline itself.
5. Which filters, i can apply before masking in the oil spill detection tool?

Kindly suggest

[ABraun]

To investigate the darkness of pixels I would rather recommend a histogram instead of a profile plot.

If you want to draw a plot along the line you should place the line on the image intensities instead of the _msk_detection output file. This would give you something more like this:

[sjpsentinel1]

i made the histo plot , but i could not able to interpret it

Then i have constructed the profile plot with sigma0 intensity as suggested.

i could able to get some signatures. i.e the values of intensity (<=0.004). i am taking this as a threshold value.
May i assume the pixels, which have this value, as oil ? can i refine this further ?

[ABraun]

? can i refine this further ?

you can either test if some filters improve the spectral plot in terms of a better separability or use the oil spill module directy on the image

[sjpsentinel1]

could you please give me the index of sigma0 values for Heavy fuel oils, so that i can do better seperation from the look alikes.
An index will definitely support, this seperation.

[ABraun]

maybe you find something helpful here, especially chapter 11
http://www.sarusersmanual.com/

[sjpsentinel1]

Thank you.

1.i have imported an ESRI shapefile into SNAP. By default, it is appearing as white. How i can change the colour of it ?

2. How i can change the colour of the drawn geometry?
3. How can i export the sigma0_vv in netcdf format? as i need to merge into an existing netcdf files .

[ABraun]

2. How i can change the colour of the drawn geometry?

using the mask manager

3. How can i export the sigma0_vv in netcdf format? as i need to merge into an existing netcdf files .

File > Export > lists several netcdf formats

[sjpsentinel1]

in the mask manager, i could able to change the colour of the particles, but they look tiny. How can i increase the size of them?

[sjpsentinel1]

In the layer manager, the particles look bigger, but i could not able to change the colour of them.

[sjpsentinel1]

I need to add band math for getting sigma0 from amplitude, could you please suggest an equation ?
where can i get the digital number ?
how can i compute the amplitude ? however it is displayed in pixel info

[sjpsentinel1]

can i square the amplitudes for geting intensity?
How our SNAP converts the amplitudes into Sigma0_vv intensity?
By the way, The intensities of bit mask show higher range (500 to 1000) but the sigma0_vv intensities are in lesser range (0.004 to 1.5) why so ?
generally sigma0 is a dimensionless number, then why it is measured in db?
Please suggest me on the above .

[ABraun]

on DN, amplitude and calibration, please see here:

Sentinel-1A GRD product

Since the DN is roughly just the square root of the backscatter coefficient (aka normalized radar cross section or sigma nought), which is a dimensionless quantity, the DN itself is dimensionless as well. There calibration equations for Sentinel-1 are described in detail in this document and in the SNAP help as well. There is, of course, no need to do the calibration yourself. You can use the Calibration operator in SNAP (Radar→Radiometric→Calibrate) to convert DN to sigma/beta/gamma nought.…

on the sense of data in db scale, please see here:

dB or DN for image processing?

Hi all, Since the original information (digital numbers DN) can be converted to backscatter (sigma0) values, or even to sigma0 in dB: what type of image information should I use for digital image processing? What value(s) is(are) better for image processing? or are the three (DN, sigma0, sigma0 dB) equally accepted and just depend on my application? I must add that I'm not interested in using classification linked to a specific value of backscatter from a structure, I want to use segmentation a…

[sjpsentinel1]

Thank you ABraun.
whether the calibration carried out in oil spill detection belongs to Distributed or point target ?

[sjpsentinel1]

while doing the radiometric calibration of GRD products , will the radiometric calibration be done for distributed or point targets ?

[sjpsentinel1]

while doing the oil spill detection, the sigma0 is computed from the intensity values for Sigma0_vv in intensity. But amplitude values are taken for computing Sigma0_vv in bil mask. Why so?
Hope intensity = amplitude^2

[junlu]

Yes, intensity = amplitude^2

[MikeLareta]

0.1

what was the value of the window size you ve used? 61 the default, 512 like someone suggests or a different one? what is the logic regarding how to set a proper window size?