def test_haardetector():
"""
Excersize the Haar feature detector.
Asserts that some basic test cases are correct.
"""
path = os.path.dirname(__file__)
image = plt.imread('%s/../examples/data/cameraman.png' % path)
options = {}
options['levels'] = 5
options['threshold'] = 0.2
options['locality'] = 5
features = detect(image, HaarDetector, options, debug=True)
assert len(features['points'].items()) > 0
def test_haardetector():
"""
Asserts that at-least one feature is detected in the "camera-man" image,
which should contain many features.
"""
path = os.path.dirname(__file__)
image = plt.imread('{0}/../examples/data/cameraman.png'.format(path))
options = {
'levels': 5,
'threshold': 0.2,
'locality': 5
}
features = detector.detect(image, detector.HaarDetector, options)
# Asserts that there are features present:
assert len(features['points'].items()) > 0
"""
Detects haar salient features in an image -
"""
import scipy.ndimage as nd
from matplotlib.pyplot import imread, plot, imshow, show
from register.features.detector import detect, HaarDetector
# Load the image.
image = imread('data/cameraman.png')
#image = nd.zoom(image, 0.50)
options = {}
options['levels'] = 5 # number of wavelet levels
options['threshold'] = 0.2 # threshold between 0.0 and 1.0 to filter out weak features (0.0 includes all features)
options['locality'] = 5 # minimum (approx) distance between two features
features = detect(image, HaarDetector, options)
imshow(image, cmap='gray')
for id, point in features['points'].items():
plot(point[1], point[0], 'or')
show()
