3D Building models were generated by using Google Sketchup Pro and Google Earth.

Then shadow casted 2D orthographic raster models were exported as TIFF or JPEG format.

For further analysis with respect to shadow casted roof top areas, exported 2D raster models have to be classified or segmented. This process wanted to be executed for extraction of shadow polygons which casted over building roof tops and roof-tops open to sun light as well.

  • To achieving this objective unsupervised classification was done. But unfortunately shadows which casted over rooftops and ground area were being merged into one single class. Rooftop area shadow could not be extracted separately.
  • Supervised classification could not be done, because of unable to get training samples. (RGB values of training samples were same for shadow areas and open areas)
  • Therefore I figured out Matlab Application for image feature extraction.
  • I used some sort of extraction techniques such as Color threshold, RGB value extraction
  • But unfortunately they all were with lack of satisfaction.
  • Even though using Matlab with different approach may be result reliable output for this purpose. But up to now, I could not find out appropriate codes.

I have to extract features separately as output of image classificationenter image description here.

closed as too broad by PolyGeo Jan 24 '17 at 20:15

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 1
    What's your actual question here? – Dan C Jan 11 '17 at 18:47
  • may you suggest a method for extracting bunch of pixels of a raster image as preserved location of their? – mr.rifas ahamed Jan 11 '17 at 19:45
  • My objective is shadow casted roof-top area and roof-top area where, those were open to sunlight have to be extracted for further analysis using GIS. – mr.rifas ahamed Jan 11 '17 at 19:46