def search_windows(img, windows, clf, scaler, color_space='RGB',
spatial_size=(32, 32), hist_bins=32,
hist_range=(0, 256), orient=9,
pix_per_cell=8, cell_per_block=2,
hog_channel=0, spatial_feat=True,
hist_feat=True, hog_feat=True):
#1) Create an empty list to receive positive detection windows
on_windows = []
#2) Iterate over all windows in the list
for window in windows:
#3) Extract the test window from original image
test_img = cv2.resize(img[window[0][1]:window[1][1], window[0][0]:window[1][0]], (64, 64))
#4) Extract features for that window using single_img_features()
features = single_img_features(test_img, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
#5) Scale extracted features to be fed to classifier
test_features = scaler.transform(np.array(features).reshape(1, -1))
#6) Predict using your classifier
prediction = clf.predict(test_features)
#7) If positive (prediction == 1) then save the window
if prediction == 1:
on_windows.append(window)
#8) Return windows for positive detections
return on_windows
def test_02_trainingse_nont(self):
images = glob.glob('train_imgs/non-vehicles/*/*.png')
(clf, scaler) = class_train.load()
counter = 0
for i in images:
img = cnst.img_read_f(i)
test_img = cv2.resize(img, (64, 64))
features = single_img_features(test_img,
color_space=cnst.color_space,
spatial_size=cnst.spatial_size,
hist_bins=cnst.hist_bins,
orient=cnst.orient,
pix_per_cell=cnst.pix_per_cell,
cell_per_block=cnst.cell_per_block,
hog_channel=cnst.hog_channel,
spatial_feat=cnst.spatial_feat,
hist_feat=cnst.hist_feat,
hog_feat=cnst.hog_feat)
test_features = scaler.transform(np.array(features).reshape(1, -1))
#test_features = features
prediction = clf.predict(test_features)
if (prediction > 0.0):
print("Prediction " + str(prediction) + " for " + i)
counter = counter + 1
print((len(images) - counter) / len(images))
def get_features_dataset(dataset,
color_space='RGB',
spatial_size=(32, 32),
hist_bins=32,
orient=9,
pix_per_cell=8,
cell_per_block=2,
hog_channel=2):
"""
:param dataset:
:param color_space:
:param spatial_size:
:param hist_bins:
:param orient:
:param pix_per_cell:
:param cell_per_block:
:param hog_channel:
:return:
"""
features = []
t0 = time.time()
# Extract features from each image in the dataset and append
for file in dataset:
img = mpimg.imread(file)
features_img = single_img_features(img, color_space, spatial_size,
hist_bins, orient, pix_per_cell,
cell_per_block, hog_channel)
features.append(features_img)
t1 = time.time()
print('Extracted %s features in %s seconds.' %
(len(features_img), t1 - t0))
return features
def test_01_trainingset(self):
images = glob.glob('train_imgs/vehicles/GTI_Far/image0000.png')
(clf, scaler) = class_train.load()
for i in images:
img = cnst.img_read_f(i)
test_img = cv2.resize(img, (64, 64))
features = single_img_features(test_img,
color_space=cnst.color_space,
spatial_size=cnst.spatial_size,
hist_bins=cnst.hist_bins,
orient=cnst.orient,
pix_per_cell=cnst.pix_per_cell,
cell_per_block=cnst.cell_per_block,
hog_channel=cnst.hog_channel,
spatial_feat=cnst.spatial_feat,
hist_feat=cnst.hist_feat,
hog_feat=cnst.hog_feat)
print(features.shape)
print(features)
print(len(features))
test_features = scaler.transform(np.array(features).reshape(1, -1))
#test_features = features
prediction = clf.predict(test_features)
print(str(prediction) + " : " + i)