You can use the newer Classification Framework which also uses the SVM algorithm but you can supply one set of training data for multiple images after you generate the classifier:
http://www.harrisgeospati...VISVMClassifier.html

Thank you for the suggestion. I tried to edit the code but I keep on getting error messages. I am very poor with programming. Is there any tools I can use other than using the IDL codes?

I am afraid that the classification framework is available through the API only. What sort of errors are you receiving? Can you run the sample code with the sample data?

I did not try with the given images. I managed to get rod of my error messages and run both SVMClassificationusingObjects and ApplySVMTrainedClassifier. First out put comes as I expect but the second output only have two classes (I am trying to classify into 8 classes). I tried with both sentinel 2 and landsat 8 data and got similar results. I have several questions about applying the classifier.
1. In the example, it asks to update the code as ROIfile = 'TrainingDataROIs.xml' for the ApplySVMTrainedClassifier. But I do not see where to change this. So I added this command from the classification using obects. Is this ok?
2. When updating the gains and offsets, I used the the gain and offset values for the second image that I am trying to classify. Is this correct.?
3. Most of the area in my image are masked. Could this be a problem?

I think I figured out what's wrong. I used the gains and offsets generated from the first set of codes and it worked! Thank you!

I am trying to loop the ApplySVMTrainedClassifier to multiple files. I included some additional codes to do this. I am getting error at line 61 ( raster = e.OpenRaster(file)) saying;

% Invalid input for filename, must be a scalar string
% Execution halted at: APPLYSVMTRAINEDCLASSIFIER 61 C:\Users\ChowFras\AppData\Local\Temp\applysvmtrainedclassifier.pro
% $MAIN$



Below is the code I am using now. I have added a line of stars at the start of the new codes to loop the files and before the line I am getting error. Can you please help me to get this fixed?

PRO ApplySVMTrainedClassifier
COMPILE_OPT IDL2
; Start the application
e = ENVI()

; Open an attribute raster to classify
file = 'FLAASH.hdr'
raster = e.OpenRaster(file)

; Open training data ROIs
ROIfile = 'TrainingDataROIs.xml'
rois = e.OpenROI(ROIfile)

; Normalize the attribute image using the
; previously recorded gains and offsets
gain = [0.00028893383, 0.00027631943,0.00025608195, 0.00016105653, 0.00021829295, 0.00019872814, 0.00018508236, 0.00016900456, 0.00021376657, 0.00029515939]
offset = [-0.24537576, -0.30148087, -0.33031915, -0.48204199, -0.36862689 , -0.50192095, -0.52861240, -0.53657914, -0.53154062, -0.48786404]
normalizedRaster = ENVIGainOffsetRaster(raster, gain, offset)

; Get the trained classifier. This is a persistable object
; in the form of an IDL save file that needs to be restored
; from disk
trainedClassifierURI = 'C:\Users\ChowFras\AppData\Local\Temp\TrainedSVMClassifier.epo'
trainedClassifier = ENVIRestoreObject(trainedClassifierURI)

; Classify a different attribute image
classRaster = ENVIClassifyRaster(normalizedRaster, trainedClassifier)

; Get the classification raster metadata
numClasses = classRaster.Metadata['Classes']
classNames = classRaster.Metadata['Class Names']

; Display the result
view = e.GetView()
layer = view.CreateLayer(raster)
layer2 = view.CreateLayer(classRaster)
view.Zoom, /FULL_EXTENT

******************************************************************************
;Find all the files with .dat extensions in a directory.

search_dir = 'I:\idl'

filelist = file_search(search_dir + '*.hdr')

;Set a base for your output files that will be appended to.

outbase = 'I:\idl\SVM_'



;Loop over every file.

;Open the raster, set parameters, and execute the task.

foreach file, filelist, index do begin
*********************************************
raster = e.OpenRaster(file)

task.input_raster = raster

task.output_raster_uri = outbase + strtrim(index, 2)

task.Execute

endforeach

END