http://www.mediawiki.org/wiki/Special:Code/pywikipedia/11590

Revision: 11590 Author: drtrigon Date: 2013-05-25 22:20:46 +0000 (Sat, 25 May 2013) Log Message: ----------- bug fix; hard crashes in C code causing core dumps (>200MB) (scipy is NOT stable use numpy instead whenever possible)

Modified Paths: -------------- trunk/pywikipedia/catimages.py

Modified: trunk/pywikipedia/catimages.py =================================================================== --- trunk/pywikipedia/catimages.py 2013-05-25 21:19:21 UTC (rev 11589) +++ trunk/pywikipedia/catimages.py 2013-05-25 22:20:46 UTC (rev 11590) @@ -60,7 +60,7 @@ # additional python packages (non-default but common) try: import numpy as np - from scipy import ndimage, fftpack, linalg#, signal + from scipy import ndimage, fftpack#, signal import cv # TS: nonofficial cv2.so backport of the testing-version of # python-opencv because of missing build-host, done by DaB @@ -707,31 +707,28 @@ ##Image.fromarray(fftpack.ifftn(fft).real).show() ##Image.fromarray(fftpack.ifftn(fftpack.ifftshift(fft)).real).show() ##Image.fromarray(fftpack.ifftn(fftpack.ifftshift(fft.real)).real).show() - try: - U, S, Vh = linalg.svd(np.matrix(fft)) - ma = 0.01*max(S) - count = sum([int(c > ma) for c in S]) + #U, S, Vh = scipy.linalg.svd(np.matrix(fft)) # unstable, crashes with C core dump + U, S, Vh = np.linalg.svd(np.matrix(fft)) + ma = 0.01*max(S) + count = sum([int(c > ma) for c in S])

- #SS = np.zeros(s) - #ss = min(s) - #for i in range(0, len(S)-1, max( int(len(S)/100.), 1 )): # (len(S)==ss) -> else; problem! - # #SS = np.zeros(s) - # #SS[:(ss-i),:(ss-i)] = np.diag(S[:(ss-i)]) - # SS[:(i+1),:(i+1)] = np.diag(S[:(i+1)]) - # #Image.fromarray((np.dot(np.dot(U, SS), Vh) - fft).real).show() - # #Image.fromarray(fftpack.ifftn(fftpack.ifftshift(np.dot(np.dot(U, SS), Vh))).real - gray).show() - # print i, ((np.dot(np.dot(U, SS), Vh) - fft).real).max() - # print i, (fftpack.ifftn(fftpack.ifftshift(np.dot(np.dot(U, SS), Vh))).real - gray).max() - # #if ((np.dot(np.dot(U, SS), Vh) - fft).max() < (255/4.)): - # # break - #data['SVD_Comp'] = float(i)/ss - #data['SVD_Min'] = S[:(i+1)].min() + #SS = np.zeros(s) + #ss = min(s) + #for i in range(0, len(S)-1, max( int(len(S)/100.), 1 )): # (len(S)==ss) -> else; problem! + # #SS = np.zeros(s) + # #SS[:(ss-i),:(ss-i)] = np.diag(S[:(ss-i)]) + # SS[:(i+1),:(i+1)] = np.diag(S[:(i+1)]) + # #Image.fromarray((np.dot(np.dot(U, SS), Vh) - fft).real).show() + # #Image.fromarray(fftpack.ifftn(fftpack.ifftshift(np.dot(np.dot(U, SS), Vh))).real - gray).show() + # print i, ((np.dot(np.dot(U, SS), Vh) - fft).real).max() + # print i, (fftpack.ifftn(fftpack.ifftshift(np.dot(np.dot(U, SS), Vh))).real - gray).max() + # #if ((np.dot(np.dot(U, SS), Vh) - fft).max() < (255/4.)): + # # break + #data['SVD_Comp'] = float(i)/ss + #data['SVD_Min'] = S[:(i+1)].min()

- data['FFT_Peaks'] = float(count)/len(S) - #pywikibot.output( u'FFT_Peaks: %s' % data['FFT_Peaks'] ) - except linalg.LinAlgError: - # SVD did not converge; in fact this should NEVER happen...(?!?) - pass + data['FFT_Peaks'] = float(count)/len(S) + #pywikibot.output( u'FFT_Peaks: %s' % data['FFT_Peaks'] ) # use wavelet transformation (FWT) from e.g. pywt, scipy signal or mlpy # (may be other) in addition to FFT and compare the spectra with FFT... # confer; "A Practical Guide to Wavelet Analysis" (http://journals.ametsoc.org/doi/pdf/10.1175/1520-0477%281998%29079%3C0061%3A...) @@ -1759,6 +1756,7 @@ #rvec, tvec = cv2.solvePnP(objectPoints, corners, cameraMatrix, None) # http://www.opencv.org.cn/opencvdoc/2.3.2/html/modules/calib3d/doc/camera_cal... # -> what about using POSIT ??? (see docs on enwiki) + # http://opencv.willowgarage.com/wiki/Posit #(cv2.findFundamentalMat, cv2.findHomography or from 'pose', cv2.estimateAffine3D)

# (todo) draw the rotated 3D object (projected down to 2D) @@ -1784,7 +1782,7 @@ rot = rvec mat, perp = self._util_getD2coords_calc(np.eye(3), cameraMatrix, rvec, tvec) ortho = mat[:2,2] - ortho = ortho/linalg.norm(ortho) + ortho = ortho/np.linalg.norm(ortho) #self._util_drawAxes(mat, 250, 350, im) #self._util_drawAxes(mat, 50, 50, im)

@@ -1816,7 +1814,7 @@ # D2norm[0] *= sign # usual 2D coords <-> pixel/picture coords # D2norm *= sign # invert all # mat[:,i] = D2norm -# matnorm[:,i] = linalg.norm(D2norm) +# matnorm[:,i] = np.linalg.norm(D2norm) # mat = mat/max(matnorm[0]) # return (mat, D3coords)

@@ -1837,13 +1835,13 @@ #rmat[:,0:3] = np.eye(3) rmat[:,3] = tvec[:,0] origin = np.dot(rmat, cv2.convertPointsToHomogeneous(np.zeros((3,3)).astype('float32')).transpose()[:,0,:]) - origin2D = np.dot((cm), origin) # linalg.inv(cm) + origin2D = np.dot((cm), origin) # np.linalg.inv(cm) #coords = np.dot(cv2.Rodrigues(rvec)[0], D3coords) coords = np.dot(rmat, cv2.convertPointsToHomogeneous(D3coords.astype('float32')).transpose()[:,0,:]) coords2D = np.dot((cm), coords) perp = coords - origin mat = coords2D - origin2D - mat = mat/max([linalg.norm(mat[:,i]) for i in range(3)]) + mat = mat/max([np.linalg.norm(mat[:,i]) for i in range(3)]) return (mat, perp)

# def _util_drawAxes(self, mat, x, y, im):