This post is a collection of questions and answers about the Semi-Automatic Classification Plugin (SCP) and remote sensing which were discussed in the Facebook group and the Google+ Community this month.
These questions vary from supervised classification technique to software issues, and can be useful to the readers of this blog for solving issues about the use of SCP.
These questions vary from supervised classification technique to software issues, and can be useful to the readers of this blog for solving issues about the use of SCP.
Question
I created a point shapefile to use for validation and run the accuracy tool. But when I tried to call it from the Error Matrix Input this shapefile does not appear in the list.
Answer
The reference shapefile must be a polygon vector file. You should convert your points to polygons.
Also, this old tutorial could be useful for you https://fromgistors.blogspot.com/2014/09/accuracy-assessment-using-random-points.html
Also, this old tutorial could be useful for you https://fromgistors.blogspot.com/2014/09/accuracy-assessment-using-random-points.html
Question
I am having troubles signing into the Sentinel-2 download of SCP. I can sign in on the website fine, but in the SCP it keeps asking for me to "Enter Credentials" and never goes further.
Answer
If your Copernicus account was created recently, you should wait a few days before they enable the access through the APIs (used by the plugin).
Question
Is there a way to adjust the spectral signature plot values axis? I am using higher level (surface reflectance) landsat data and the plot looks like this in the picture. I assume it is due to the fact that the values range from about 30 to 6000.
Answer
You should convert the values to float or use the Multiplicative Factor in Band set (see http://semiautomaticclassificationmanual-v5.readthedocs.io/en/latest/main_interface_window.html#band-set-definition).
Question
Is it possible to create a spectral signature file to classify let's say 2000 image and use the same created spectral file to classify another image say 2010? What would be the repercussions of such?
Answer
Theoretically it is possible, but practically I wouldn't recommend to use the same signatures in two images because the classification accuracy could be low (because of the different seasons, weather conditions, etc).
Question
I am doing the interpretation of my spectral signatures obtained from Landsat and Sentinel-2, but I do not understand the difference with the bands 12 (2.19μm).
Answer
The spectral signature depends on spatial resolution which is different for the two images.
When comparing ROI's spectral signatures there could be single pixels of different materials that can modify the average spectral signature.
You should compare the spectral signatures of single pixels. Possibly one or more pixels belong to different materials.
When comparing ROI's spectral signatures there could be single pixels of different materials that can modify the average spectral signature.
You should compare the spectral signatures of single pixels. Possibly one or more pixels belong to different materials.
For any comment or question, join the Facebook group and the Google+ Community about the Semi-Automatic Classification Plugin.