python类local_binary_pattern()的实例源码

def extract_RGB_LBP_features(image, labels, size=5, P=8, R=2):
    n_sp = np.max(labels)+1
    hs = size//2
    img_superpixel = np.zeros_like(labels, dtype='int')

    # calculate lbp for entire region
    lbp_img = np.empty((3, ), dtype='object')
    for d in range(3):
        lbp_img[d] = local_binary_pattern(image[..., d], P=P, R=R, method='uniform')

    feat_desc_size = P+1

    feat_descs = np.zeros((n_sp, feat_desc_size*3))

    for i in range(n_sp):
        # get centroid of i'th superpixel
        img_superpixel[:] = labels == i
        cy, cx = [np.rint(x).astype('int') for x in regionprops(img_superpixel)[0].centroid]

        # extract lbp values in sizeXsize region centred on the centroid
        x0, y0, x1, y1 = cx-hs, cy-hs, cx+hs+1, cy+hs+1

        # clip to boundaries of image
        x0 = 0 if x0 < 0 else x0
        y0 = 0 if y0 < 0 else y0
        x1 = image.shape[1]-1 if x1 > image.shape[1]-2 else x1
        y1 = image.shape[0]-1 if y1 > image.shape[0]-2 else y1

        # fill in the feature vector for each image channel
        for d in range(3):
            j, k = d*feat_desc_size, (1+d)*feat_desc_size
            patch = lbp_img[d][y0:y1, x0:x1].flat

            fv = np.histogram(patch, bins=np.arange(0, feat_desc_size+1), range=(0, feat_desc_size+1))[0]
            feat_descs[i, j:k] = fv

    return feat_descs

def LBP(train_image,radius, no_points):
    im = cv2.imread(train_image)
    im_gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
    lbp = local_binary_pattern(im_gray, no_points, radius, method='default')
    #(hist, _) = np.histogram(lbp.ravel(), bins=np.arange(0,(no_points*(no_points-1)+4)))
    #hist = hist.astype("float")
    #hist /= (hist.sum() + eps)
    #print hist.sum()
    print lbp
    return lbp

def LBP(train_image,radius, no_points):
    im = cv2.imread(train_image)
    im_gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
    lbp = local_binary_pattern(im_gray, no_points, radius, method='default')
    (hist, _) = np.histogram(lbp[0:63][0:63].ravel(), bins=np.arange(0,(no_points*(no_points-1)+4)))
    hist = hist.astype("float")
    hist /= (hist.sum() + eps)
    #print hist.sum()
    #print lbp
    return hist

def calculatehistogram(self, image, eps=1e-7):
        lbp = feature.local_binary_pattern(image, self.numberOfPoints, self.radius, method="uniform")
        (histogram, _) = np.histogram(lbp.ravel(),
                                      bins=np.arange(0, self.numberOfPoints + 3),
                                      range=(0, self.numberOfPoints + 2))
        #now we need to normalise the histogram so that the total sum=1
        histogram = histogram.astype("float")
        histogram /= (histogram.sum() + eps)
        return histogram

def calculatehistogram(self, image, eps=1e-7):
        lbp = feature.local_binary_pattern(image, self.numberOfPoints, self.radius, method="uniform")
        (histogram, _) = np.histogram(lbp.ravel(),
                                      bins=np.arange(0, self.numberOfPoints + 3),
                                      range=(0, self.numberOfPoints + 2))
        #now we need to normalise the histogram so that the total sum=1
        histogram = histogram.astype("float")
        histogram /= (histogram.sum() + eps)
        return histogram

def calculatehistogram(self, image, eps=1e-7):
        lbp = feature.local_binary_pattern(image, self.numberOfPoints, self.radius, method="uniform")
        (histogram, _) = np.histogram(lbp.ravel(),
                                      bins=np.arange(0, self.numberOfPoints + 3),
                                      range=(0, self.numberOfPoints + 2))
        # now we need to normalise the histogram so that the total sum=1
        histogram = histogram.astype("float")
        histogram /= (histogram.sum() + eps)
        return histogram

def calculatehistogram(self, image, eps=1e-7):
        lbp = feature.local_binary_pattern(image, self.numberOfPoints, self.radius, method="uniform")
        (histogram, _) = np.histogram(lbp.ravel(),
                                      bins=np.arange(0, self.numberOfPoints + 3),
                                      range=(0, self.numberOfPoints + 2))
        # now we need to normalise the histogram so that the total sum=1
        histogram = histogram.astype("float")
        histogram /= (histogram.sum() + eps)
        return histogram

def getLBP(img):
    img2 = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    radius = 1
    n_points = 8 * radius
    lbpImage = (local_binary_pattern(img2, n_points, radius)).astype(int)**(1.0/radius)

    # block processing:
    lbpImages = block_view(lbpImage, ( int(lbpImage.shape[0] / 2), int(lbpImage.shape[1] / 4)))


    count = 0

    LBP = np.array([]); 
    for i in range(lbpImages.shape[0]):         # for each block:
        for j in range(lbpImages.shape[1]):
            count += 1
#           plt.subplot(4,2,count)
#           plt.imshow(lbpImages[i,j,:,:],cmap = cm.Greys_r)
#           plt.subplot(4,2,count+4*2/2)
#           print count*2+1
            LBPt = cv2.calcHist([lbpImages[i,j,:,:].astype('uint8')], [0], None, [8], [0, 256]) 
            LBP = np.append(LBP, LBPt[:,0]);
#           plt.plot(LBPt)
#   plt.show()


    Fnames = ["LBP"+str(i).zfill(2) for i in range(len(LBP))]

    return normalize(LBP).tolist(), Fnames

def generate_lbp_features(filename):
    radius = 3
    n_points = 10 * radius
    image_arr = io.imread(filename, as_grey=True)
    return local_binary_pattern(image_arr, P=n_points, R=radius)

def generate_lbp_features(filename):
    radius = 3
    n_points = 10 * radius
    image_arr = io.imread(filename, as_grey=True)
    return local_binary_pattern(image_arr, P=n_points, R=radius)

def getLBP(img):
    img2 = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    radius = 1
    n_points = 8 * radius
    lbpImage = (local_binary_pattern(img2, n_points, radius)).astype(int)**(1.0/radius)

    # block processing:
    lbpImages = block_view(lbpImage, ( int(lbpImage.shape[0] / 2), int(lbpImage.shape[1] / 4)))


    count = 0

    LBP = np.array([]); 
    for i in range(lbpImages.shape[0]):         # for each block:
        for j in range(lbpImages.shape[1]):
            count += 1
#           plt.subplot(4,2,count)
#           plt.imshow(lbpImages[i,j,:,:],cmap = cm.Greys_r)
#           plt.subplot(4,2,count+4*2/2)
#           print count*2+1
            LBPt = cv2.calcHist([lbpImages[i,j,:,:].astype('uint8')], [0], None, [8], [0, 256]) 
            LBP = np.append(LBP, LBPt[:,0]);
#           plt.plot(LBPt)
#   plt.show()


    Fnames = ["LBP"+str(i).zfill(2) for i in range(len(LBP))]

    return normalize(LBP).tolist(), Fnames