Python scipy.concatenate方法代码示例

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def extend(self, extracted_features):
    # This method reads the pkl files in a folder and adds them to the 
    # existing data for processing in the TCData class.


    (data, labels, feature_string, width, height, winsize, nbins) = extracted_features
    npixels = width * height

    xlabel = 'Grayscale intensity'
    ylabel = 'Probability'
    xvals  = scipy.arange(self.data.shape[0]).reshape(-1,1)
    self.data       = N.concatenate((self.data, data),axis=1) 
    self.width      = N.append(self.width, width)
    self.height     = N.append(self.height, height)
    self.xvals      = N.append(self.xvals, xvals)
    self.labels.extend(labels)
    
    self.img_label_split.extend([len(self.labels)])
    self.data_split.extend([self.data.shape[1]]) 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def find_log_ticks(start, stop):
    """
    finds tick values for linear axis
    """
    if (start < stop):
        min, max = start, stop
    else:
        min, max = stop, start
    # lists for ticks
    tick_0_list = []
    tick_1_list = []
    tick_2_list = []
    max_decade = math.ceil(math.log10(max))
    min_decade = math.floor(math.log10(min))
    start_ax = None
    stop_ax = None
    for decade in scipy.arange(min_decade, max_decade + 1, 1):
        # for number in scipy.concatenate((scipy.arange(1,2,0.2),scipy.arange(2,3,0.5),scipy.arange(3,10,1))):
        for number in [1, 1.2, 1.4, 1.6, 1.8, 2.0, 2.5, 3, 4, 5, 6, 7, 8, 9]:
            u = number * 10.0 ** decade
            if u >= min and u <= max:
                if start_ax == None:
                    start_ax = number
                stop_ax = number
                if number == 1:
                    tick_0_list.append(u)
                if number in [2, 3, 4, 5, 6, 7, 8, 9]:
                    tick_1_list.append(u)
                if number in [1.2, 1.4, 1.6, 1.8, 2.5]:
                    tick_2_list.append(u)
    # print tick_0_list
    # print tick_1_list
    # print tick_2_list
    return tick_0_list, tick_1_list, tick_2_list, start_ax, stop_ax 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def _compose(orig, recon):
    _imgo = []
    _imgr = []
    for i in range(orig.shape[0]):
        _imgo.append(orig[i])
    for i in range(orig.shape[0]):
        _imgr.append(recon[i])
    _imgo = sp.concatenate(_imgo, 1)
    _imgr = sp.concatenate(_imgr, 1)
    _rv = sp.concatenate([_imgo, _imgr], 0)
    _rv = sp.clip(_rv, 0, 1)
    return _rv 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def _compose_multi(imgs):
    _imgs = []
    for i in range(len(imgs)):
        _imgs.append([])
        for j in range(imgs[i].shape[0]):
            _imgs[i].append(imgs[i][j])
        _imgs[i] = sp.concatenate(_imgs[i], 1)
    _rv = sp.concatenate(_imgs, 0)
    _rv = sp.clip(_rv, 0, 1)
    return _rv 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def si_read_ppm(self, rawfilename, filename):
    # This function reads the ppm/jpg file and extracts the features if the 
    # features pkl file doesn't exist. It is also compatible for extension 
    # of the feauture vector and doesn't compute the already computed features

    new_feature_string = []
    updated_feature = 0
    data = N.array([], dtype=int)
    if os.path.exists(filename):
      pkl_f = open(filename, 'r')
      (data, labels, feature_string, width, height, winsize, nbins)= pickle.load(pkl_f)
      self.winsize = winsize
      self.nbins = nbins
      new_feature_string = list(feature_string)
      pkl_f.close()      

    if not new_feature_string.count('sift'):
      updated_feature = 1
      (sift_features, labels, width, height) = self.extract_sift(rawfilename, self.winsize, self.nbins)

      if data.size:
        data = scipy.concatenate((data.transpose(), sift_features.transpose()), 1).transpose()
      else:
        data = sift_features
      new_feature_string.append('sift')

    if updated_feature:
      outf = open(filename, 'w')
      pickle.dump((data, labels, new_feature_string, width, height, self.winsize, self.nbins),outf)
      outf.close()
      print 'Saved data to %s.' % filename
    
    return (data, labels, new_feature_string, width, height, self.winsize, self.nbins) 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def testpoly():
    [x, y] = scipy.mgrid[0:10, 0:10]
    #print 'X'
    #print x
    #print 'Y'
    #print y
    u = scipy.zeros((10, 10))
    v = scipy.zeros((10, 10))
    # Random polynomials
    a0 = scipy.random.rand(1)
    a1 = 0.1*(scipy.random.rand(2)-0.5)
    a2 = 0.01*(scipy.random.rand(3)-0.5)
    a = scipy.concatenate((a0, a1))
    a = scipy.concatenate((a, a2))
    a[2] = 0.01*a[2]
    print('A coefficients')
    print(a)
    b0 = scipy.random.rand(1)
    b1 = 0.1*(scipy.random.rand(2)-0.5)
    b2 = 0.01*(scipy.random.rand(3)-0.5)
    b = scipy.concatenate((b0, b1))
    b = scipy.concatenate((b, b2))
    b[1] = 0.01*b[1]
    print('B coeffcicients')
    print(b)
    for i in range(10):
        for j in range(10):
            u[i, j] = poly(a, x[i, j], y[i, j], 2) #+ scipy.random.normal(0.0, 0.01)
            v[i, j] = poly(b, x[i, j], y[i, j], 2)  #+ scipy.random.normal(0.0, 0.01)
    #print z
    s1 = polyFit2(u, x, y, 2)
    s2 = polyFit2(v, x, y, 2)
    print('S1', s1)
    print('S2', s2)
    uc = poly(s1, x, y, 2)
    vc = poly(s2, x, y, 2)

    P.figure(1)
    P.clf()
    P.grid(True)
    P.plot(u, v, 'gx')
    P.plot(uc, vc, 'r+') 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def  extract_sift(cls, rawfilename, winsize, nbins):
    """read_ppm(rawfilename, filename)

    Read in raw pixel data from rawfilename (.ppm).
    Create a histogram around each pixel to become
    the feature vector for that obsevation (pixel).
    Pickle the result and save it to filename.
    Note: does NOT update object fields.
    Follow this with a call to readin().
    """
    if cls._VL_SIFT_:
      # VLSIFT matlab 

      im  = Image.open(rawfilename)
      (width, height) = im.size

      mlab.bb_sift(N.array(im), 'temp.mat')
      sift_features = scipy.io.loadmat('temp.mat')
      kp = sift_features['f_']
      sift_features = sift_features['d_']
      sift_features  = scipy.concatenate((sift_features.transpose(), kp[2:4].transpose()), 1).transpose()

      labels = [];
      for ikp in kp.transpose():
        (x,y) = ikp[0:2]
        labels    += ['(%d,%d)' % (y,x)]
    else:
      #Opencv SIFT 
      img = cv2.imread(rawfilename)
      gray= cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
      height, width = gray.shape

      # Computing SIFT
      sift = cv2.SIFT(edgeThreshold = 3)
      kp, des = sift.detectAndCompute(gray,None)

      labels  = []
      sift_features = N.transpose(des)
      scale_angle = []

      for ikp in kp:
        (x,y) = ikp.pt
        scale_angle.append([ikp.size/12, ikp.angle])
        labels    += ['(%d,%d)' % (y,x)]
    
      scale_angle = N.array(scale_angle)
      sift_features  = scipy.concatenate((sift_features.transpose(), scale_angle), 1).transpose()

    return (sift_features, labels, width, height) 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def  extract_hist(cls, rawfilename, winsize, nbins):
    # This function extracts the histogram features from the image

    im  = Image.open(rawfilename)
    
    (width, height) = im.size
    npixels = width * height
    pix = scipy.array(im)

    # Generate one feature vector (histogram) per pixel
    #winsize = 20  # for test.pgm
    #winsize = 0  # for RGB
    halfwin = int(winsize/2)

    bins    = scipy.linspace(0, 255, nbins)

    # Only use windows that are fully populated
    mywidth  = width-winsize
    myheight = height-winsize
    #data     = scipy.zeros((nbins-1, mywidth * myheight))
    #data     = scipy.zeros((3*winsize*winsize, mywidth * myheight))
    data    = []
    labels  = []

    # Pick up all windows, stepping by half of the window size
    for y in range(halfwin, height-halfwin, int(halfwin/2)):
      for x in range(halfwin, width-halfwin, int(halfwin/2)):
        # Read in data in row-major order
        ind = (y-halfwin)*mywidth + (x-halfwin)
        #data[:,ind] = \
        #    scipy.histogram(pix[y-halfwin:y+halfwin,
        #                        x-halfwin:x+halfwin],
        #                        bins)[0]
        # Just RGB
        #data[:,ind] = pix[y,x]
        # RGB window
        #data[:,ind] = pix[y-halfwin:y+halfwin,x-halfwin:x+halfwin].flat
        hist_features = TCData.extract_hist_subimg(pix[y-halfwin:y+halfwin,x-halfwin:x+halfwin])
        if data == []:
          data = hist_features.reshape(-1,1)
        else:
          data = scipy.concatenate((data, hist_features.reshape(-1,1)),1)
        labels    += ['(%d,%d)' % (y,x)]

    return (data, labels, width, height) 

# 需要导入模块: import scipy [as 别名]
# 或者: from scipy import concatenate [as 别名]
def read_ppm(self, rawfilename, filename):
    # This function reads the ppm/jpg file and extracts the features if the 
    # features pkl file doesn't exist. It is also compatible for extension 
    # of the feauture vector and doesn't compute the already computed features

    new_feature_string = []
    updated_feature = 0
    data = N.array([], dtype=int)
    if os.path.exists(filename):
      pkl_f = open(filename, 'r')
      (data, labels, feature_string, width, height, winsize, nbins)= pickle.load(pkl_f)
      self.winsize = winsize
      self.nbins = nbins
      new_feature_string = list(feature_string)
      pkl_f.close()      

    if not new_feature_string.count('dsift'):
      updated_feature = 1
      (sift_features, labels, width, height) = self.extract_dsift(rawfilename, self.winsize, self.nbins)
      if data.size:
        data = scipy.concatenate((data.transpose(), sift_features.transpose()), 1).transpose()
      else:
        data = sift_features
      new_feature_string.append('dsift')

    if not new_feature_string.count('histogram'):
      updated_feature = 1 
      (hist_features, labels, width, height) = self.extract_hist(rawfilename, self.winsize, self.nbins)
      hist_features = hist_features/(self.winsize)
      if data.size:
        data = scipy.concatenate((data.transpose(), hist_features.transpose()), 1).transpose()
      else:
        data = hist_features
      new_feature_string.append('histogram')

    '''
    if not new_feature_string.count('position'):
      updated_feature = 1 
      
      position_features = []
      for label in labels:
        (y,x) = map(int, label.strip('()').split(','))
        position_features.append([x,y]) 
      position_features = N.array(position_features)
    
      if data.size:
        data = scipy.concatenate((data.transpose(), position_features), 1).transpose()
      else:
        data = position_features
      new_feature_string.append('position')
    '''
    if updated_feature:
      outf = open(filename, 'w')
      pickle.dump((data, labels, new_feature_string, width, height, self.winsize, self.nbins),outf)
      outf.close()
      print 'Saved data to %s.' % filename
    
    return (data, labels, new_feature_string, width, height, self.winsize, self.nbins)