class map

geosentinel/L2ASCLtoJSON.py

@@ -25,29 +25,25 @@ SENTINEL_PATH = config['PATHS']['PATH_GEOSENTINEL']
 sys.path.append(SENTINEL_PATH)
 
 from geosentinel import rasterWkt
-
+CLASS_MAP = {0:0, 1:1, 2:2, 3:3, 4:4, 5:5, 6:6, 7:7, 8:8, 9:8, 10:8, 11:9}
 #-------------------------------------------------------------------------------
 def L2ASCLtoDict(filename, wkt):
 	''' 
 	L2ASCLtoDict
 	'''
+	classMap = lambda x: [ CLASS_MAP[xx] for xx in x ]
+
 	inputImage = gdal.Open(filename)
 	rows, cols, geotransform = inputImage.RasterYSize, inputImage.RasterXSize, inputImage.GetGeoTransform()
 	data = inputImage.GetRasterBand(1).ReadAsArray(0,0,cols,rows)
-#	count = Counter([])
-#	for r in data:
-#		count = count + Counter(r)
-#	sclDict = {str(k):v for k,v in dict(count).items()}
-#	print ("data")
-#	print (sclDict)
 
 	data = np.bitwise_and(data, rasterWkt.getPolygonArray(inputImage, wkt))
 	count = Counter([])
 	for r in data:
-		count = count + Counter(r)
+		count = count + Counter(classMap(r))
+
 	sclDict = {str(k):v for k,v in dict(count).items()}
-#	print ("data & POLY")
-#	print (sclDict)
+
 	return sclDict
 #-------------------------------------------------------------------------------
 def main(argv):

geosentinel/splitSCL.py

@@ -20,7 +20,8 @@ SCL_COLOR = [
 		(102,204,255,255), #THIN_CIRRUS
 		(255,153,255,255), #SNOW
 		 ]
-
+CLASS_MAP = {0:0, 1:1, 2:2, 3:3, 4:4, 5:5, 6:6, 7:7, 8:8, 9:8, 10:8, 11:9}
+CLASS_GROUP = [[0], [1], [2], [3], [4], [5], [6], [7], [8,9,10], [11]]
 #-------------------------------------------------------------------------------
 def splitSCL(filename, outdir):
 	inputImage = gdal.Open(filename)
@@ -33,10 +34,12 @@ def splitSCL(filename, outdir):
 	sr.ImportFromWkt(projection)
 	driverTiff = gdal.GetDriverByName('GTiff')
 
-	for c in range(len(SCL_COLOR)):
+#	for c in range(len(SCL_COLOR)):
+	for c in range(len(CLASS_GROUP)):
 		outputfile=outdir+filename[:-4]+"_"+str(c)+".tif"
 		output = driverTiff.Create(outputfile, cols, rows, 4, gdal.GDT_Byte)
-		classMask  = inputImage.GetRasterBand(1).ReadAsArray(0,0,cols,rows)==c
+#		classMask  = inputImage.GetRasterBand(1).ReadAsArray(0,0,cols,rows)==c
+                classMask  = np.isin(inputImage.GetRasterBand(1).ReadAsArray(0,0,cols,rows), CLASS_GROUP[c])
 
 		for b in range(1, output.RasterCount+1):
 			output.GetRasterBand(b).SetNoDataValue(0)