def main():
gdal.AllRegister()
path = auxil.select_directory('Choose working directory')
if path:
os.chdir(path)
infile = auxil.select_infile(title='Select an image')
if infile:
inDataset = gdal.Open(infile,GA_ReadOnly)
cols = inDataset.RasterXSize
rows = inDataset.RasterYSize
bands = inDataset.RasterCount
else:
return
pos = auxil.select_pos(bands)
if not pos:
return
bands = len(pos)
dims = auxil.select_dims([0,0,cols,rows])
if dims:
x0,y0,cols,rows = dims
else:
return
class_image = np.zeros((rows,cols),dtype=np.byte)
K = auxil.select_integer(6,'Number of clusters')
max_scale = auxil.select_integer(2,'Maximum scaling factor')
max_scale = min((max_scale,3))
min_scale = auxil.select_integer(0,'Minimum scaling factor')
min_scale = min((max_scale,min_scale))
T0 = auxil.select_float(0.5,'Initial annealing temperature')
beta = auxil.select_float(0.5,'Spatial mixing parameter')
outfile, outfmt = auxil.select_outfilefmt('Select output classification file')
if not outfile:
return
probfile, probfmt = auxil.select_outfilefmt('Select output probability file (optional)')
print '========================='
print ' EM clustering'
print '========================='
print 'infile: %s'%infile
print 'clusters: %i'%K
print 'T0: %f'%T0
print 'beta: %f'%beta
start = time.time()
# read in image and compress
DWTbands = []
for b in pos:
band = inDataset.GetRasterBand(b)
DWTband = auxil.DWTArray(band.ReadAsArray(x0,y0,cols,rows).astype(float),cols,rows)
for i in range(max_scale):
DWTband.filter()
DWTbands.append(DWTband)
rows,cols = DWTbands[0].get_quadrant(0).shape
G = np.transpose(np.array([DWTbands[i].get_quadrant(0,float=True).ravel() for i in range(bands)]))
# initialize membership matrix
n = G.shape[0]
U = np.random.random((K,n))
den = np.sum(U,axis=0)
for j in range(K):
U[j,:] = U[j,:]/den
# cluster at minimum scale
try:
U,Ms,Cs,Ps,pdens = em(G,U,T0,beta,rows,cols)
except:
print 'em failed'
return
# sort clusters wrt partition density
idx = np.argsort(pdens)
idx = idx[::-1]
U = U[idx,:]
# clustering at increasing scales
for i in range(max_scale-min_scale):
# expand U and renormalize
U = np.reshape(U,(K,rows,cols))
rows = rows*2
cols = cols*2
U = ndi.zoom(U,(1,2,2))
U = np.reshape(U,(K,rows*cols))
idx = np.where(U<0.0)
U[idx] = 0.0
den = np.sum(U,axis=0)
for j in range(K):
U[j,:] = U[j,:]/den
# expand the image
for i in range(bands):
DWTbands[i].invert()
G = np.transpose(np.array([DWTbands[i].get_quadrant(0,float=True).ravel() for i in range(bands)]))
# cluster
unfrozen = np.where(np.max(U,axis=0) < 0.90)
try:
U,Ms,Cs,Ps,pdens = em(G,U,0.0,beta,rows,cols,unfrozen=unfrozen)
except:
print 'em failed'
return
print 'Cluster mean vectors'
print Ms
print 'Cluster covariance matrices'
for k in range(K):
print 'cluster: %i'%k
print Cs[k]
# up-sample class memberships if necessary
if min_scale>0:
U = np.reshape(U,(K,rows,cols))
f = 2**min_scale
rows = rows*f
cols = cols*f
U = ndi.zoom(U,(1,f,f))
U = np.reshape(U,(K,rows*cols))
idx = np.where(U<0.0)
U[idx] = 0.0
den = np.sum(U,axis=0)
for j in range(K):
U[j,:] = U[j,:]/den
# classify
labels = np.byte(np.argmax(U,axis=0)+1)
class_image[0:rows,0:cols] = np.reshape(labels,(rows,cols))
rows1,cols1 = class_image.shape
# write to disk
driver = gdal.GetDriverByName(outfmt)
outDataset = driver.Create(outfile,cols1,rows1,1,GDT_Byte)
projection = inDataset.GetProjection()
geotransform = inDataset.GetGeoTransform()
if geotransform is not None:
gt = list(geotransform)
gt[0] = gt[0] + x0*gt[1]
gt[3] = gt[3] + y0*gt[5]
outDataset.SetGeoTransform(tuple(gt))
if projection is not None:
outDataset.SetProjection(projection)
outBand = outDataset.GetRasterBand(1)
outBand.WriteArray(class_image,0,0)
outBand.FlushCache()
outDataset = None
# write class membership probability file if desired
if probfile:
driver = gdal.GetDriverByName(probfmt)
outDataset = driver.Create(probfile,cols,rows,K,GDT_Byte)
if geotransform is not None:
outDataset.SetGeoTransform(tuple(gt))
if projection is not None:
outDataset.SetProjection(projection)
for k in range(K):
probs = np.reshape(U[k,:],(rows,cols))
probs = np.byte(probs*255)
outBand = outDataset.GetRasterBand(k+1)
outBand.WriteArray(probs,0,0)
outBand.FlushCache()
outDataset = None
print 'class probabilities written to: %s'%probfile
inDataset = None
if (outfmt == 'ENVI') and (K<19):
# try to make an ENVI classification header file
hdr = header.Header()
headerfile = outfile+'.hdr'
f = open(headerfile)
line = f.readline()
envihdr = ''
while line:
envihdr += line
line = f.readline()
f.close()
hdr.read(envihdr)
hdr['file type'] ='ENVI Classification'
hdr['classes'] = str(K+1)
classlookup = '{0'
for i in range(1,3*(K+1)):
classlookup += ', '+str(str(ctable[i]))
classlookup +='}'
hdr['class lookup'] = classlookup
hdr['class names'] = ['class %i'%i for i in range(K+1)]
f = open(headerfile,'w')
f.write(str(hdr))
f.close()
print 'classification written to: '+outfile
print 'elapsed time: '+str(time.time()-start)
print '--done------------------------'
def main():
print '================================'
print 'Complex Wishart Change Detection'
print '================================'
print time.asctime()
gdal.AllRegister()
path = auxil.select_directory('Choose working directory')
if path:
os.chdir(path)
# first SAR image
infile1 = auxil.select_infile(title='Choose first SAR image')
if infile1:
inDataset1 = gdal.Open(infile1,GA_ReadOnly)
cols = inDataset1.RasterXSize
rows = inDataset1.RasterYSize
bands = inDataset1.RasterCount
else:
return
m = auxil.select_integer(5,msg='Number of looks')
if not m:
return
print 'first filename: %s'%infile1
print 'number of looks: %i'%m
# second SAR image
infile2 = auxil.select_infile(title='Choose second SAR image')
if not infile2:
return
n = auxil.select_integer(5,msg='Number of looks')
if not n:
return
print 'second filename: %s'%infile2
print 'number of looks: %i'%n
# output file
outfile,fmt = auxil.select_outfilefmt()
if not outfile:
return
# significance level
sig = auxil.select_float(0.01, 'Choose significance level')
print 'Signifcane level: %f'%sig
start = time.time()
print 'co-registering...'
registerSAR.registerSAR(infile1,infile2,'warp.tif','GTiff')
infile2 = 'warp.tif'
inDataset2 = gdal.Open(infile2,GA_ReadOnly)
cols2 = inDataset2.RasterXSize
rows2 = inDataset2.RasterYSize
bands2 = inDataset2.RasterCount
if (bands != bands2) or (cols != cols2) or (rows != rows2):
print 'Size mismatch'
return
if bands == 9:
print 'Quad polarimetry'
# C11 (k1)
b = inDataset1.GetRasterBand(1)
k1 = m*b.ReadAsArray(0,0,cols,rows)
# C12 (a1)
b = inDataset1.GetRasterBand(2)
a1 = b.ReadAsArray(0,0,cols,rows)
b = inDataset1.GetRasterBand(3)
im = b.ReadAsArray(0,0,cols,rows)
a1 = m*(a1 + 1j*im)
# C13 (rho1)
b = inDataset1.GetRasterBand(4)
rho1 = b.ReadAsArray(0,0,cols,rows)
b = inDataset1.GetRasterBand(5)
im = b.ReadAsArray(0,0,cols,rows)
rho1 = m*(rho1 + 1j*im)
# C22 (xsi1)
b = inDataset1.GetRasterBand(6)
xsi1 = m*b.ReadAsArray(0,0,cols,rows)
# C23 (b1)
b = inDataset1.GetRasterBand(7)
b1 = b.ReadAsArray(0,0,cols,rows)
b = inDataset1.GetRasterBand(8)
im = b.ReadAsArray(0,0,cols,rows)
b1 = m*(b1 + 1j*im)
# C33 (zeta1)
b = inDataset1.GetRasterBand(9)
zeta1 = m*b.ReadAsArray(0,0,cols,rows)
# C11 (k2)
b = inDataset2.GetRasterBand(1)
k2 = n*b.ReadAsArray(0,0,cols,rows)
# C12 (a2)
b = inDataset2.GetRasterBand(2)
a2 = b.ReadAsArray(0,0,cols,rows)
b = inDataset2.GetRasterBand(3)
im = b.ReadAsArray(0,0,cols,rows)
a2 = n*(a2 + 1j*im)
# C13 (rho2)
b = inDataset2.GetRasterBand(4)
rho2 = b.ReadAsArray(0,0,cols,rows)
b = inDataset2.GetRasterBand(5)
im = b.ReadAsArray(0,0,cols,rows)
rho2 = n*(rho2 + 1j*im)
# C22 (xsi2)
b = inDataset2.GetRasterBand(6)
xsi2 = n*b.ReadAsArray(0,0,cols,rows)
# C23 (b2)
b = inDataset2.GetRasterBand(7)
b2 = b.ReadAsArray(0,0,cols,rows)
b = inDataset2.GetRasterBand(8)
im = b.ReadAsArray(0,0,cols,rows)
b2 = n*(b2 + 1j*im)
# C33 (zeta2)
b = inDataset2.GetRasterBand(9)
zeta2 = n*b.ReadAsArray(0,0,cols,rows)
k3 = k1 + k2
a3 = a1 + a2
rho3 = rho1 + rho2
xsi3 = xsi1 + xsi2
b3 = b1 + b2
zeta3 = zeta1 + zeta2
det1 = k1*xsi1*zeta1 + 2*np.real(a1*b1*np.conj(rho1)) - xsi1*(abs(rho1)**2) - k1*(abs(b1)**2) - zeta1*(abs(a1)**2)
det2 = k2*xsi2*zeta2 + 2*np.real(a2*b2*np.conj(rho2)) - xsi2*(abs(rho2)**2) - k2*(abs(b2)**2) - zeta2*(abs(a2)**2)
det3 = k3*xsi3*zeta3 + 2*np.real(a3*b3*np.conj(rho3)) - xsi3*(abs(rho3)**2) - k3*(abs(b3)**2) - zeta3*(abs(a3)**2)
p = 3
f = p**2
cst = p*((n+m)*np.log(n+m)-n*np.log(n)-m*np.log(m))
rho = 1. - (2.*p**2-1.)*(1./n + 1./m - 1./(n+m))/(6.*p)
omega2 = -(p*p/4.)*(1. - 1./rho)**2 + p**2*(p**2-1.)*(1./n**2 + 1./m**2 - 1./(n+m)**2)/(24.*rho**2)
elif bands == 4:
print 'Dual polarimetry'
# C11 (k1)
b = inDataset1.GetRasterBand(1)
k1 = m*b.ReadAsArray(0,0,cols,rows)
# C12 (a1)
b = inDataset1.GetRasterBand(2)
a1 = b.ReadAsArray(0,0,cols,rows)
b = inDataset1.GetRasterBand(3)
im = b.ReadAsArray(0,0,cols,rows)
a1 = m*(a1 + 1j*im)
# C22 (xsi1)
b = inDataset1.GetRasterBand(4)
xsi1 = m*b.ReadAsArray(0,0,cols,rows)
# C11 (k2)
b = inDataset2.GetRasterBand(1)
k2 = n*b.ReadAsArray(0,0,cols,rows)
# C12 (a2)
b = inDataset2.GetRasterBand(2)
a2 = b.ReadAsArray(0,0,cols,rows)
b = inDataset2.GetRasterBand(3)
im = b.ReadAsArray(0,0,cols,rows)
a2 = n*(a2 + 1j*im)
# C22 (xsi2)
b = inDataset2.GetRasterBand(4)
xsi2 = n*b.ReadAsArray(0,0,cols,rows)
k3 = k1 + k2
a3 = a1 + a2
xsi3 = xsi1 + xsi2
det1 = k1*xsi1 - abs(a1)**2
det2 = k2*xsi2 - abs(a2)**2
det3 = k3*xsi3 - abs(a3)**2
p = 2
cst = p*((n+m)*np.log(n+m)-n*np.log(n)-m*np.log(m))
f = p**2
rho = 1-(2*f-1)*(1./n+1./m-1./(n+m))/(6.*p)
omega2 = -f/4.*(1-1./rho)**2 + f*(f-1)*(1./n**2+1./m**2-1./(n+m)**2)/(24.*rho**2)
elif bands == 1:
print 'Single polarimetry'
# C11 (k1)
b = inDataset1.GetRasterBand(1)
k1 = m*b.ReadAsArray(0,0,cols,rows)
# C11 (k2)
b = inDataset2.GetRasterBand(1)
k2 = n*b.ReadAsArray(0,0,cols,rows)
k3 = k1 + k2
det1 = k1
det2 = k2
det3 = k3
p = 1
cst = p*((n+m)*np.log(n+m)-n*np.log(n)-m*np.log(m))
f = p**2
rho = 1-(2.*f-1)*(1./n+1./m-1./(n+m))/(6.*p)
omega2 = -f/4.*(1-1./rho)**2+f*(f-1)*(1./n**2+1./m**2-1./(n+m)**2)/(24.*rho**2)
else:
print 'Incorrect number of bands'
return
idx = np.where(det1 <= 0.0)
det1[idx] = 0.0001
idx = np.where(det2 <= 0.0)
det2[idx] = 0.0001
idx = np.where(det3 <= 0.0)
det3[idx] = 0.0001
lnQ = cst+m*np.log(det1)+n*np.log(det2)-(n+m)*np.log(det3)
# test statistic
Z = -2*rho*lnQ
# change probabilty
P = (1.-omega2)*stats.chi2.cdf(Z,[f])+omega2*stats.chi2.cdf(Z,[f+4])
P = ndimage.filters.median_filter(P, size = (3,3))
# change map
a255 = np.ones((rows,cols),dtype=np.byte)*255
a0 = a255*0
c11 = np.log(k1+0.0001)
min1 =np.min(c11)
max1 = np.max(c11)
c11 = (c11-min1)*255.0/(max1-min1)
c11 = np.where(c11<0,a0,c11)
c11 = np.where(c11>255,a255,c11)
c11 = np.where(P>(1.0-sig),a0,c11)
cmap = np.where(P>(1.0-sig),a255,c11)
# write to file system
driver = gdal.GetDriverByName(fmt)
outDataset = driver.Create(outfile,cols,rows,2,GDT_Float32)
geotransform = inDataset1.GetGeoTransform()
if geotransform is not None:
outDataset.SetGeoTransform(geotransform)
projection = inDataset1.GetProjection()
if projection is not None:
outDataset.SetProjection(projection)
outBand = outDataset.GetRasterBand(1)
outBand.WriteArray(Z,0,0)
outBand.FlushCache()
outBand = outDataset.GetRasterBand(2)
outBand.WriteArray(P,0,0)
outBand.FlushCache()
outDataset = None
print 'test statistic and probabilities written to: %s'%outfile
basename = os.path.basename(outfile)
name, ext = os.path.splitext(basename)
outfile=outfile.replace(name,name+'_cmap')
outDataset = driver.Create(outfile,cols,rows,3,GDT_Byte)
geotransform = inDataset1.GetGeoTransform()
if geotransform is not None:
outDataset.SetGeoTransform(geotransform)
projection = inDataset1.GetProjection()
if projection is not None:
outDataset.SetProjection(projection)
outBand = outDataset.GetRasterBand(1)
outBand.WriteArray(cmap,0,0)
outBand.FlushCache()
outBand = outDataset.GetRasterBand(2)
outBand.WriteArray(c11,0,0)
outBand.FlushCache()
outBand = outDataset.GetRasterBand(3)
outBand.WriteArray(c11,0,0)
outBand.FlushCache()
outDataset = None
print 'change map image written to: %s'%outfile
print 'elapsed time: '+str(time.time()-start)
def main():
print '================================'
print 'Complex Wishart Change Detection'
print '================================'
print time.asctime()
gdal.AllRegister()
path = auxil.select_directory('Choose working directory')
if path:
os.chdir(path)
# first SAR image
infile1 = auxil.select_infile(title='Choose first SAR image')
if infile1:
inDataset1 = gdal.Open(infile1, GA_ReadOnly)
cols = inDataset1.RasterXSize
rows = inDataset1.RasterYSize
bands = inDataset1.RasterCount
else:
return
m = auxil.select_integer(5, msg='Number of looks')
if not m:
return
print 'first filename: %s' % infile1
print 'number of looks: %i' % m
# second SAR image
infile2 = auxil.select_infile(title='Choose second SAR image')
if not infile2:
return
n = auxil.select_integer(5, msg='Number of looks')
if not n:
return
print 'second filename: %s' % infile2
print 'number of looks: %i' % n
# output file
outfile, fmt = auxil.select_outfilefmt()
if not outfile:
return
# significance level
sig = auxil.select_float(0.01, 'Choose significance level')
print 'Signifcane level: %f' % sig
start = time.time()
print 'co-registering...'
registerSAR.registerSAR(infile1, infile2, 'warp.tif', 'GTiff')
infile2 = 'warp.tif'
inDataset2 = gdal.Open(infile2, GA_ReadOnly)
cols2 = inDataset2.RasterXSize
rows2 = inDataset2.RasterYSize
bands2 = inDataset2.RasterCount
if (bands != bands2) or (cols != cols2) or (rows != rows2):
print 'Size mismatch'
return
if bands == 9:
print 'Quad polarimetry'
# C11 (k1)
b = inDataset1.GetRasterBand(1)
k1 = m * b.ReadAsArray(0, 0, cols, rows)
# C12 (a1)
b = inDataset1.GetRasterBand(2)
a1 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset1.GetRasterBand(3)
im = b.ReadAsArray(0, 0, cols, rows)
a1 = m * (a1 + 1j * im)
# C13 (rho1)
b = inDataset1.GetRasterBand(4)
rho1 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset1.GetRasterBand(5)
im = b.ReadAsArray(0, 0, cols, rows)
rho1 = m * (rho1 + 1j * im)
# C22 (xsi1)
b = inDataset1.GetRasterBand(6)
xsi1 = m * b.ReadAsArray(0, 0, cols, rows)
# C23 (b1)
b = inDataset1.GetRasterBand(7)
b1 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset1.GetRasterBand(8)
im = b.ReadAsArray(0, 0, cols, rows)
b1 = m * (b1 + 1j * im)
# C33 (zeta1)
b = inDataset1.GetRasterBand(9)
zeta1 = m * b.ReadAsArray(0, 0, cols, rows)
# C11 (k2)
b = inDataset2.GetRasterBand(1)
k2 = n * b.ReadAsArray(0, 0, cols, rows)
# C12 (a2)
b = inDataset2.GetRasterBand(2)
a2 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset2.GetRasterBand(3)
im = b.ReadAsArray(0, 0, cols, rows)
a2 = n * (a2 + 1j * im)
# C13 (rho2)
b = inDataset2.GetRasterBand(4)
rho2 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset2.GetRasterBand(5)
im = b.ReadAsArray(0, 0, cols, rows)
rho2 = n * (rho2 + 1j * im)
# C22 (xsi2)
b = inDataset2.GetRasterBand(6)
xsi2 = n * b.ReadAsArray(0, 0, cols, rows)
# C23 (b2)
b = inDataset2.GetRasterBand(7)
b2 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset2.GetRasterBand(8)
im = b.ReadAsArray(0, 0, cols, rows)
b2 = n * (b2 + 1j * im)
# C33 (zeta2)
b = inDataset2.GetRasterBand(9)
zeta2 = n * b.ReadAsArray(0, 0, cols, rows)
k3 = k1 + k2
a3 = a1 + a2
rho3 = rho1 + rho2
xsi3 = xsi1 + xsi2
b3 = b1 + b2
zeta3 = zeta1 + zeta2
det1 = k1 * xsi1 * zeta1 + 2 * np.real(
a1 * b1 * np.conj(rho1)) - xsi1 * (abs(rho1)**2) - k1 * (
abs(b1)**2) - zeta1 * (abs(a1)**2)
det2 = k2 * xsi2 * zeta2 + 2 * np.real(
a2 * b2 * np.conj(rho2)) - xsi2 * (abs(rho2)**2) - k2 * (
abs(b2)**2) - zeta2 * (abs(a2)**2)
det3 = k3 * xsi3 * zeta3 + 2 * np.real(
a3 * b3 * np.conj(rho3)) - xsi3 * (abs(rho3)**2) - k3 * (
abs(b3)**2) - zeta3 * (abs(a3)**2)
p = 3
f = p**2
cst = p * ((n + m) * np.log(n + m) - n * np.log(n) - m * np.log(m))
rho = 1. - (2. * p**2 - 1.) * (1. / n + 1. / m - 1. /
(n + m)) / (6. * p)
omega2 = -(p * p / 4.) * (1. - 1. / rho)**2 + p**2 * (p**2 - 1.) * (
1. / n**2 + 1. / m**2 - 1. / (n + m)**2) / (24. * rho**2)
elif bands == 4:
print 'Dual polarimetry'
# C11 (k1)
b = inDataset1.GetRasterBand(1)
k1 = m * b.ReadAsArray(0, 0, cols, rows)
# C12 (a1)
b = inDataset1.GetRasterBand(2)
a1 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset1.GetRasterBand(3)
im = b.ReadAsArray(0, 0, cols, rows)
a1 = m * (a1 + 1j * im)
# C22 (xsi1)
b = inDataset1.GetRasterBand(4)
xsi1 = m * b.ReadAsArray(0, 0, cols, rows)
# C11 (k2)
b = inDataset2.GetRasterBand(1)
k2 = n * b.ReadAsArray(0, 0, cols, rows)
# C12 (a2)
b = inDataset2.GetRasterBand(2)
a2 = b.ReadAsArray(0, 0, cols, rows)
b = inDataset2.GetRasterBand(3)
im = b.ReadAsArray(0, 0, cols, rows)
a2 = n * (a2 + 1j * im)
# C22 (xsi2)
b = inDataset2.GetRasterBand(4)
xsi2 = n * b.ReadAsArray(0, 0, cols, rows)
k3 = k1 + k2
a3 = a1 + a2
xsi3 = xsi1 + xsi2
det1 = k1 * xsi1 - abs(a1)**2
det2 = k2 * xsi2 - abs(a2)**2
det3 = k3 * xsi3 - abs(a3)**2
p = 2
cst = p * ((n + m) * np.log(n + m) - n * np.log(n) - m * np.log(m))
f = p**2
rho = 1 - (2 * f - 1) * (1. / n + 1. / m - 1. / (n + m)) / (6. * p)
omega2 = -f / 4. * (1 - 1. / rho)**2 + f * (f - 1) * (
1. / n**2 + 1. / m**2 - 1. / (n + m)**2) / (24. * rho**2)
elif bands == 1:
print 'Single polarimetry'
# C11 (k1)
b = inDataset1.GetRasterBand(1)
k1 = m * b.ReadAsArray(0, 0, cols, rows)
# C11 (k2)
b = inDataset2.GetRasterBand(1)
k2 = n * b.ReadAsArray(0, 0, cols, rows)
k3 = k1 + k2
det1 = k1
det2 = k2
det3 = k3
p = 1
cst = p * ((n + m) * np.log(n + m) - n * np.log(n) - m * np.log(m))
f = p**2
rho = 1 - (2. * f - 1) * (1. / n + 1. / m - 1. / (n + m)) / (6. * p)
omega2 = -f / 4. * (1 - 1. / rho)**2 + f * (f - 1) * (
1. / n**2 + 1. / m**2 - 1. / (n + m)**2) / (24. * rho**2)
else:
print 'Incorrect number of bands'
return
idx = np.where(det1 <= 0.0)
det1[idx] = 0.0001
idx = np.where(det2 <= 0.0)
det2[idx] = 0.0001
idx = np.where(det3 <= 0.0)
det3[idx] = 0.0001
lnQ = cst + m * np.log(det1) + n * np.log(det2) - (n + m) * np.log(det3)
# test statistic
Z = -2 * rho * lnQ
# change probabilty
P = (1. - omega2) * stats.chi2.cdf(Z, [f]) + omega2 * stats.chi2.cdf(
Z, [f + 4])
P = ndimage.filters.median_filter(P, size=(3, 3))
# change map
a255 = np.ones((rows, cols), dtype=np.byte) * 255
a0 = a255 * 0
c11 = np.log(k1 + 0.0001)
min1 = np.min(c11)
max1 = np.max(c11)
c11 = (c11 - min1) * 255.0 / (max1 - min1)
c11 = np.where(c11 < 0, a0, c11)
c11 = np.where(c11 > 255, a255, c11)
c11 = np.where(P > (1.0 - sig), a0, c11)
cmap = np.where(P > (1.0 - sig), a255, c11)
# write to file system
driver = gdal.GetDriverByName(fmt)
outDataset = driver.Create(outfile, cols, rows, 2, GDT_Float32)
geotransform = inDataset1.GetGeoTransform()
if geotransform is not None:
outDataset.SetGeoTransform(geotransform)
projection = inDataset1.GetProjection()
if projection is not None:
outDataset.SetProjection(projection)
outBand = outDataset.GetRasterBand(1)
outBand.WriteArray(Z, 0, 0)
outBand.FlushCache()
outBand = outDataset.GetRasterBand(2)
outBand.WriteArray(P, 0, 0)
outBand.FlushCache()
outDataset = None
print 'test statistic and probabilities written to: %s' % outfile
basename = os.path.basename(outfile)
name, ext = os.path.splitext(basename)
outfile = outfile.replace(name, name + '_cmap')
outDataset = driver.Create(outfile, cols, rows, 3, GDT_Byte)
geotransform = inDataset1.GetGeoTransform()
if geotransform is not None:
outDataset.SetGeoTransform(geotransform)
projection = inDataset1.GetProjection()
if projection is not None:
outDataset.SetProjection(projection)
outBand = outDataset.GetRasterBand(1)
outBand.WriteArray(cmap, 0, 0)
outBand.FlushCache()
outBand = outDataset.GetRasterBand(2)
outBand.WriteArray(c11, 0, 0)
outBand.FlushCache()
outBand = outDataset.GetRasterBand(3)
outBand.WriteArray(c11, 0, 0)
outBand.FlushCache()
outDataset = None
print 'change map image written to: %s' % outfile
print 'elapsed time: ' + str(time.time() - start)
def main():
gdal.AllRegister()
path = auxil.select_directory('Choose working directory')
if path:
os.chdir(path)
infile = auxil.select_infile(title='Select an image')
if infile:
inDataset = gdal.Open(infile, GA_ReadOnly)
cols = inDataset.RasterXSize
rows = inDataset.RasterYSize
bands = inDataset.RasterCount
else:
return
pos = auxil.select_pos(bands)
if not pos:
return
bands = len(pos)
dims = auxil.select_dims([0, 0, cols, rows])
if dims:
x0, y0, cols, rows = dims
else:
return
class_image = np.zeros((rows, cols), dtype=np.byte)
K = auxil.select_integer(6, 'Number of clusters')
max_scale = auxil.select_integer(2, 'Maximum scaling factor')
max_scale = min((max_scale, 3))
min_scale = auxil.select_integer(0, 'Minimum scaling factor')
min_scale = min((max_scale, min_scale))
T0 = auxil.select_float(0.5, 'Initial annealing temperature')
beta = auxil.select_float(0.5, 'Spatial mixing parameter')
outfile, outfmt = auxil.select_outfilefmt(
'Select output classification file')
if not outfile:
return
probfile, probfmt = auxil.select_outfilefmt(
'Select output probability file (optional)')
print '========================='
print ' EM clustering'
print '========================='
print 'infile: %s' % infile
print 'clusters: %i' % K
print 'T0: %f' % T0
print 'beta: %f' % beta
start = time.time()
# read in image and compress
DWTbands = []
for b in pos:
band = inDataset.GetRasterBand(b)
DWTband = auxil.DWTArray(
band.ReadAsArray(x0, y0, cols, rows).astype(float), cols, rows)
for i in range(max_scale):
DWTband.filter()
DWTbands.append(DWTband)
rows, cols = DWTbands[0].get_quadrant(0).shape
G = np.transpose(
np.array([
DWTbands[i].get_quadrant(0, float=True).ravel()
for i in range(bands)
]))
# initialize membership matrix
n = G.shape[0]
U = np.random.random((K, n))
den = np.sum(U, axis=0)
for j in range(K):
U[j, :] = U[j, :] / den
# cluster at minimum scale
try:
U, Ms, Cs, Ps, pdens = em(G, U, T0, beta, rows, cols)
except:
print 'em failed'
return
# sort clusters wrt partition density
idx = np.argsort(pdens)
idx = idx[::-1]
U = U[idx, :]
# clustering at increasing scales
for i in range(max_scale - min_scale):
# expand U and renormalize
U = np.reshape(U, (K, rows, cols))
rows = rows * 2
cols = cols * 2
U = ndi.zoom(U, (1, 2, 2))
U = np.reshape(U, (K, rows * cols))
idx = np.where(U < 0.0)
U[idx] = 0.0
den = np.sum(U, axis=0)
for j in range(K):
U[j, :] = U[j, :] / den
# expand the image
for i in range(bands):
DWTbands[i].invert()
G = np.transpose(
np.array([
DWTbands[i].get_quadrant(0, float=True).ravel()
for i in range(bands)
]))
# cluster
unfrozen = np.where(np.max(U, axis=0) < 0.90)
try:
U, Ms, Cs, Ps, pdens = em(G,
U,
0.0,
beta,
rows,
cols,
unfrozen=unfrozen)
except:
print 'em failed'
return
print 'Cluster mean vectors'
print Ms
print 'Cluster covariance matrices'
for k in range(K):
print 'cluster: %i' % k
print Cs[k]
# up-sample class memberships if necessary
if min_scale > 0:
U = np.reshape(U, (K, rows, cols))
f = 2**min_scale
rows = rows * f
cols = cols * f
U = ndi.zoom(U, (1, f, f))
U = np.reshape(U, (K, rows * cols))
idx = np.where(U < 0.0)
U[idx] = 0.0
den = np.sum(U, axis=0)
for j in range(K):
U[j, :] = U[j, :] / den
# classify
labels = np.byte(np.argmax(U, axis=0) + 1)
class_image[0:rows, 0:cols] = np.reshape(labels, (rows, cols))
rows1, cols1 = class_image.shape
# write to disk
driver = gdal.GetDriverByName(outfmt)
outDataset = driver.Create(outfile, cols1, rows1, 1, GDT_Byte)
projection = inDataset.GetProjection()
geotransform = inDataset.GetGeoTransform()
if geotransform is not None:
gt = list(geotransform)
gt[0] = gt[0] + x0 * gt[1]
gt[3] = gt[3] + y0 * gt[5]
outDataset.SetGeoTransform(tuple(gt))
if projection is not None:
outDataset.SetProjection(projection)
outBand = outDataset.GetRasterBand(1)
outBand.WriteArray(class_image, 0, 0)
outBand.FlushCache()
outDataset = None
# write class membership probability file if desired
if probfile:
driver = gdal.GetDriverByName(probfmt)
outDataset = driver.Create(probfile, cols, rows, K, GDT_Byte)
if geotransform is not None:
outDataset.SetGeoTransform(tuple(gt))
if projection is not None:
outDataset.SetProjection(projection)
for k in range(K):
probs = np.reshape(U[k, :], (rows, cols))
probs = np.byte(probs * 255)
outBand = outDataset.GetRasterBand(k + 1)
outBand.WriteArray(probs, 0, 0)
outBand.FlushCache()
outDataset = None
print 'class probabilities written to: %s' % probfile
inDataset = None
if (outfmt == 'ENVI') and (K < 19):
# try to make an ENVI classification header file
hdr = header.Header()
headerfile = outfile + '.hdr'
f = open(headerfile)
line = f.readline()
envihdr = ''
while line:
envihdr += line
line = f.readline()
f.close()
hdr.read(envihdr)
hdr['file type'] = 'ENVI Classification'
hdr['classes'] = str(K + 1)
classlookup = '{0'
for i in range(1, 3 * (K + 1)):
classlookup += ', ' + str(str(ctable[i]))
classlookup += '}'
hdr['class lookup'] = classlookup
hdr['class names'] = ['class %i' % i for i in range(K + 1)]
f = open(headerfile, 'w')
f.write(str(hdr))
f.close()
print 'classification written to: ' + outfile
print 'elapsed time: ' + str(time.time() - start)
print '--done------------------------'
