amntools.py

import numpy as np
import matplotlib.pyplot as plt
import colorsys
import pomio
import cv2

"""
Miscellaneous Tools
"""

def readImage( fn ):
  img = cv2.imread(fn,-1)
  if img == None:
    raise IOError( "Image " + fn + " not found" )
  if img.ndim == 3:
    # opencv does BGR
    img = img[:,:,::-1]
  return img

def writeImage( fn, img ):
  if img.ndim == 3:
    # opencv does BGR
    img = img.copy()[:,:,::-1]
  cv2.imwrite(fn,img)

def _get_colors(num_colors):
    colors=[]
    for i in np.arange(0., 360., 360. / num_colors):
        hue = i/360.
        lightness = (50 + np.random.rand() * 10)/100.
        saturation = (90 + np.random.rand() * 10)/100.
        colors.append(colorsys.hls_to_rgb(hue, lightness, saturation))
    return colors


def estimateNeighbourRMSPixelDiff(imgRGB, nhoodSz):
    # estimate neighbourhood average diff
    sigsq = 0
    cnt = 0
    rows = imgRGB.shape[0]
    cols = imgRGB.shape[1]
    # vertical diffs
    idiffs = imgRGB[0:rows-1,:,:] - imgRGB[1:rows,:,:]
    sigsq += np.power(idiffs,2.0).sum()
    cnt += (rows-1)*cols
    # horizontal diffs
    idiffs = imgRGB[:,0:cols-1,:] - imgRGB[:,1:cols,:]
    sigsq += np.power(idiffs,2.0).sum()
    cnt += rows*(cols-1)
    
    if nhoodSz == 8:
        # diagonal to right
        idiffs = imgRGB[0:rows-1,0:cols-1,:] - imgRGB[1:rows,1:cols,:]
        sigsq += np.power(idiffs,2.0).sum()
        cnt += (rows-1)*(cols-1)
        # diagonal to left
        idiffs = imgRGB[0:rows-1,1:cols,:] - imgRGB[1:rows,0:cols-1,:]
        sigsq += np.power(idiffs,2.0).sum()
        cnt += (rows-1)*(cols-1)
    
    sigsq /= cnt
    return np.sqrt(sigsq)


def gplotmatrix( X, labels, msize=5, classColours=None, featureNames=None ):
  assert X.ndim == 2
  D = X.shape[1]
  if labels == None:
    labels = np.zeros( (X.shape[0]), dtype=int )

  assert type(labels)==list or labels.ndim == 1
  assert( len(labels) == X.shape[0] )
  ULAB = np.unique(labels)
  # assume labels are contiguous
  lmin = np.min(labels)
  lmax = np.max(labels)
  plt.clf()
  if classColours == None:
    classColours = _get_colors(lmax+1)

  idx = 1
  for r in range(D):
    for c in range(D):
      plt.subplot(D,D,idx)
      idx += 1
      x1 = X[:,r]
      x2 = X[:,c]
      if r==c:
          # histogram
          plt.hist( x1 )
      else:
          for l in ULAB:
              plt.plot( x2[labels==l], x1[labels==l], '.', \
                            color=classColours[l],\
                            markersize=msize)
              plt.hold(1)
          plt.hold(0)
      plt.grid(1)
      if c==0:
        sl = str(r)
        if featureNames != None:
          sl += ': ' + featureNames[r]
        plt.ylabel( sl )
      if r==D-1:
        sl = str(c)
        if featureNames != None:
          sl += ': ' + featureNames[c]
        plt.xlabel( sl )