def process(self, cascade, nested_cascade, img_file):
cascade = cv2.CascadeClassifier(self.current_dir + cascade)
nested = cv2.CascadeClassifier(self.current_dir + nested_cascade)
gray = cv2.imread(self.current_dir + img_file, 0)
# gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# gray = cv2.equalizeHist(gray)
face_rects = self.detect(gray, cascade)
vis = gray.copy()
self.draw_rects(vis, face_rects, (0, 255, 0))
eye_count = 0
if not nested.empty():
for x1, y1, x2, y2 in face_rects:
face_roi = gray[y1:y2, x1:x2]
face_vis = vis[y1:y2, x1:x2]
eyes_rects = self.detect(face_roi.copy(), nested)
self.draw_rects(face_vis, eyes_rects, (255, 0, 0))
for x1, y1, x2, y2 in eyes_rects:
eyes_roi = face_roi[y1:y2, x1:x2]
edge_eyes_roi = edge_detect.edge_detect(eyes_roi)
helper.save_image(
edge_eyes_roi, self.current_dir +
'/../resources/eyes_roi%s.png' % (eye_count))
# edge_eyes_roi = self.noise_removal.median(edge_eyes_roi)
edge_eyes_roi = self.noise_removal.opening(edge_eyes_roi)
helper.save_image(
edge_eyes_roi, self.current_dir +
'/../resources/eyes_roi_noise_removal%s.png' %
(eye_count))
eye_count += 1
edge_eyes_roi = np.array(edge_eyes_roi, dtype=np.uint8)
eyes_vis = face_vis[y1:y2, x1:x2]
circles = cv2.HoughCircles(edge_eyes_roi,
cv2.cv.CV_HOUGH_GRADIENT,
1,
20,
param1=50,
param2=30,
minRadius=0,
maxRadius=0)
if circles is not None:
circles = np.uint16(np.around(circles))
for i in circles[0, :]:
cv2.circle(eyes_vis, (i[0], i[1]), i[2],
(0, 255, 0), 2)
cv2.imshow('face detect', vis)
def process(self, cascade, nested_cascade, img_file):
cascade = cv2.CascadeClassifier(self.current_dir + cascade)
nested = cv2.CascadeClassifier(self.current_dir + nested_cascade)
gray = cv2.imread(self.current_dir + img_file, 0)
# gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# gray = cv2.equalizeHist(gray)
face_rects = self.detect(gray, cascade)
vis = gray.copy()
self.draw_rects(vis, face_rects, (0, 255, 0))
eye_count = 0
if not nested.empty():
for x1, y1, x2, y2 in face_rects:
face_roi = gray[y1:y2, x1:x2]
face_vis = vis[y1:y2, x1:x2]
eyes_rects = self.detect(face_roi.copy(), nested)
self.draw_rects(face_vis, eyes_rects, (255, 0, 0))
for x1, y1, x2, y2 in eyes_rects:
eyes_roi = face_roi[y1:y2, x1:x2]
edge_eyes_roi = edge_detect.edge_detect(eyes_roi)
helper.save_image(edge_eyes_roi, self.current_dir + "/../resources/eyes_roi%s.png" % (eye_count))
# edge_eyes_roi = self.noise_removal.median(edge_eyes_roi)
edge_eyes_roi = self.noise_removal.opening(edge_eyes_roi)
helper.save_image(
edge_eyes_roi, self.current_dir + "/../resources/eyes_roi_noise_removal%s.png" % (eye_count)
)
eye_count += 1
edge_eyes_roi = np.array(edge_eyes_roi, dtype=np.uint8)
eyes_vis = face_vis[y1:y2, x1:x2]
circles = cv2.HoughCircles(
edge_eyes_roi, cv2.cv.CV_HOUGH_GRADIENT, 1, 20, param1=50, param2=30, minRadius=0, maxRadius=0
)
if circles is not None:
circles = np.uint16(np.around(circles))
for i in circles[0, :]:
cv2.circle(eyes_vis, (i[0], i[1]), i[2], (0, 255, 0), 2)
cv2.imshow("face detect", vis)
os.makedirs(epoch_dir)
helper.write_loss_in_txt(g_loss, epoch)
saver.save(sess, config.TRAIN.out_dir + "/model.ckpt")
except:
time.sleep(1)
## ------------ validation loop -------------------------
for ind in range(len(val_file_list)):
A_file_name_val = config.base_dir + config.VAL.A_data_dir + '/' + val_file_list[ind]
if not os.path.isfile(A_file_name_val): # test label
continue
A_image_val = helper.read_image(A_file_name_val) # training image A
# B_image_val = helper.read_image(B_file_name_val) # training image A
output = sess.run(generator_test, feed_dict={input_A_test: A_image_val})
output = np.concatenate([A_image_val, output, B_image], axis=2)
helper.save_image(output, config.TRAIN.out_dir + "/%04d/" % epoch + val_file_list[ind].replace('.jpg', '_out.jpg'))
# ---------------------------------------------------
# test
# ---------------------------------------------------
if config.TEST.is_test:
test_file_list = os.listdir(config.base_dir + config.TEST.A_data_dir)
if not os.path.isdir(config.TEST.out_dir + config.TEST.out_dir_postfix):
os.makedirs(config.TEST.out_dir + config.TEST.out_dir_postfix)
time_list = np.zeros(len(test_file_list), dtype=float)
for ind in range(len(test_file_list)):
A_file_name_val = config.base_dir + config.TEST.A_data_dir + '/' + test_file_list[ind]
if not os.path.isfile(A_file_name_val):
continue
def _prewitt(self):
return ([[-1, 0, 1], [-1, 0, 1], [-1, 0, 1]],
[[-1, -1, -1], [0, 0, 0], [1, 1, 1]])
def bounding_box(img, range_w=None, range_h=None):
img = histogram_equalization.adaptive_mean(img)
cv2.imshow('binary', img)
contours, hierarchy = cv2.findContours(img, cv2.RETR_LIST, cv2.CHAIN_APPROX_TC89_KCOS)
boundings = []
for cnt in contours:
x, y, w, h = cv2.boundingRect(cnt)
if (not range_w or (w >= range_w[0] and w <= range_w[1])) and \
(not range_h or (h >= range_h[0] and h <= range_h[1])):
boundings.append((x, y, w, h))
cv2.rectangle(img, (x, y), (x + w, y + h), (255), 2)
return img, boundings
if __name__ == '__main__':
pix = helper.read_image('../resources/fp01.jpg')
pix = helper.convert_gray(pix)
helper.save_image('../resources/gray.png', True)
pix = basic(pix)
# noise_removal = NoiseRemoval()
# pix = noise_removal.opening(pix)
helper.save_image(pix, '../resources/edge.png')