def get_frame(self):
self.curr_time = time.time()
ret, frame = self.cap.read()
if ret:
ret, jpeg = cv2.imencode('.jpg', frame)
boxes = self.yolo.predict(frame)
if (len(boxes) > 0):
frame2 = draw_boxes(frame, boxes,
self.config['model']['labels'])
ret, jpeg = cv2.imencode('.jpg', frame2)
if self.curr_time - self.last_recorded_time >= self.report_interval and boxes[
0].get_score() >= self.conf_threshold:
print(boxes[0].get_score())
submit_form()
self.last_recorded_time = self.curr_time
return jpeg.tobytes()
return jpeg.tobytes()
else:
return None
def detect(img, data, ctpn, sess):
blobs, im_scales, img, scale = data.get_blobs(img, None)
boxes,scores = ctpn.predict(blobs, im_scales, img, sess)
boxes = ctpn.detect(boxes, scores[:, np.newaxis], img.shape[:2])
# text_recs, detected_img = draw_boxes_(img, boxes, scale)
text_recs, detected_img = draw_boxes(img, boxes, caption='im_name', wait=True, is_display=True)
return text_recs, detected_img, img
def toad_game_custom():
x1 = x2 = y1 = y2 = 0
cap = cv2.VideoCapture(CAMERA)
td = ToadGenerator()
i = 0
boxes = []
while True:
flag, img = cap.read()
if i % 25 == 0:
boxes = get_hand_prediction(img)
boxes = non_max_suppression_fast(boxes)
if len(boxes) > 0:
x1, y1, x2, y2 = boxes[0]
img = td.step(img, (x1, y1, x2, y2))
try:
img = draw_boxes(img, boxes)
cv2.imshow('Toad collection', img)
except:
cap.release()
raise
i += 1
ch = cv2.waitKey(1)
if ch == 27:
break
cap.release()
cv2.destroyAllWindows()
def detect_image(self, image):
if self.model_image_size != (None, None):
assert self.model_image_size[
0] % 32 == 0, 'Multiples of 32 required'
assert self.model_image_size[
1] % 32 == 0, 'Multiples of 32 required'
image_data = preprocess_image(image, self.model_image_size)
#
image_shape = np.array([image.size[1], image.size[0]])
image_shape = np.expand_dims(image_shape, 0)
start = time.time()
out_boxes, out_classes, out_scores = self.predict(
image_data, image_shape)
end = time.time()
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
print("Inference time: {:.8f}s".format(end - start))
# draw result on input image
image_array = np.array(image, dtype='uint8')
image_array = draw_boxes(image_array, out_boxes, out_classes,
out_scores, self.class_names, self.colors)
out_classnames = [self.class_names[c] for c in out_classes]
return Image.fromarray(
image_array), out_boxes, out_classnames, out_scores
def predict_image(sess, image_in_file, image_out_file):
image, image_data = preprocess_image(image_in_file,
model_image_size=(608, 608))
out_scores, out_boxes, out_classes = sess.run([scores, boxes, classes],
feed_dict={
yolo_model.input:
image_data,
K.learning_phase(): 0
})
print('Found {} boxes for {}'.format(len(out_boxes), image_in_file))
# Generate colors for drawing bounding boxes.
colors = generate_colors(class_names)
# Draw bounding boxes on the image file
draw_boxes(image, out_scores, out_boxes, out_classes, class_names, colors)
# Save the predicted bounding box on the image
image.save(image_out_file, quality=90)
# Display the results in the notebook
output_image = scipy.misc.imread(image_out_file)
imshow(output_image)
pylab.show()
return out_scores, out_boxes, out_classes
def predict_video(sess, video_file, video_out_file):
video_in = imageio.get_reader(video_file)
frames = []
for i, image in enumerate(tqdm(video_in)):
imageio.imwrite("data/cache.jpg", image)
image_in, image_data = preprocess_image("data/cache.jpg",
model_image_size=(608, 608),
image_shape=(720, 1280),
type="video")
out_scores, out_boxes, out_classes = sess.run([scores, boxes, classes],
feed_dict={
yolo_model.input:
image_data,
K.learning_phase(): 0
})
colors = generate_colors(class_names)
draw_boxes(image_in, out_scores, out_boxes, out_classes, class_names,
colors)
image_in.save(os.path.join("data", "cache_out.jpg"), quality=90)
frames.append(imageio.imread("data/cache_out.jpg"))
imageio.mimsave(video_out_file, frames)
def main():
try:
FLAG = process_config()
except:
print("missing or invalid arguments")
exit(0)
if FLAG.GPU_options:
session_config = tf.ConfigProto()
session_config.gpu_options.per_process_gpu_memory_fraction = 0.9
session_config.gpu_options.allow_growth = True
sess = tf.Session(config=session_config)
else:
sess = tf.Session()
model = yolov3(FLAG)
model.build()
model.init_saver()
model.load(sess)
image_test = Image.open('images/timg.jpg')
resized_image = image_test.resize(size=(416, 416))
image_data = np.array(resized_image, dtype='float32') / 255.0
img_hw = tf.placeholder(dtype=tf.float32, shape=[2])
boxes, scores, classes = model.pedict(img_hw,
iou_threshold=0.5,
score_threshold=0.5)
begin_time = time.time()
boxes_, scores_, classes_, conv0 = sess.run(
[boxes, scores, classes, model.feature_extractor.conv0],
feed_dict={
img_hw: [image_test.size[1], image_test.size[0]],
model.x: [image_data]
})
end_time = time.time()
print(end_time - begin_time)
# print conv0
image_draw = draw_boxes(np.array(image_test, dtype=np.float32) / 255,
boxes_,
classes_,
FLAG.names,
scores=scores_)
fig = plt.figure(frameon=False)
ax = plt.Axes(fig, [0, 0, 1, 1])
ax.set_axis_off()
fig.add_axes(ax)
plt.imshow(image_draw)
fig.savefig('prediction.jpg')
plt.show()
sess.close()
def main():
ctpn = CTPN(cfg)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
ctpn.load_ckpt(sess)
if cfg.ADJUST_ANGLE:
angle_detector = VGG(cfg)
angle_detector.load_weights()
data = DataLoader(cfg)
img = data.load_data('images/xuanye.jpg')
t = time.time()
if cfg.ADJUST_ANGLE:
angle = angle_detector.predict(img=np.copy(img))
print('The angel of this character is:', angle)
im = Image.fromarray(img)
print('Rotate the array of this img!')
if angle == 90:
im = im.transpose(Image.ROTATE_270)
elif angle == 180:
im = im.transpose(Image.ROTATE_180)
elif angle == 270:
im = im.transpose(Image.ROTATE_90)
img = np.array(im)
# img = cv2.resize(img, (2000,3000),interpolation=cv2.INTER_CUBIC)
blobs, im_scales, resized_img, scale = data.get_blobs(img, None)
boxes, scores = ctpn.predict(blobs, im_scales, resized_img, sess)
boxes = ctpn.detect(boxes, scores[:, np.newaxis], resized_img.shape[:2])
text_recs, im = draw_boxes(resized_img,
boxes,
caption='im_name',
wait=True,
is_display=True)
# text_recs = sort_box(text_recs)
print("It takes time:{}s".format(time.time() - t))
# cv2.imshow('img',im)
# cv2.waitKey(0)
cv2.imwrite('images/result.jpg', im)
boxes = decode_netout(
netout[0],
obj_threshold=
0.25, #predicted box_score should be greater than this threshold
nms_threshold=
0.28, #threshold limiting the percentage operlap between the predicted bounding boxes
anchors=ANCHORS,
nb_class=CLASS)
# Boxes has x,y,w,h
boxes = scale_boxes(image, boxes)
print('Got ', len(boxes), ' cells!')
#if gorund truth boxes are required to be shown in image pass gt=gt below
image_box, final_boxes = draw_boxes(image,
boxes,
labels=LABELS,
h_threshold=0.2,
w_threshold=0.2,
h_min=0.05,
w_min=0.05,
gt=[])
cv2.imwrite(fileOut + f, image)
with open(annoOut + annot_fname, 'w') as pred_file:
for eachpred in final_boxes:
pred_file.write(
'%.2f %.2f %.2f %.2f\n' %
(eachpred[0], eachpred[1], eachpred[2], eachpred[3]))
def _main():
# parse command line arguments
parser = argparse.ArgumentParser()
requiredNamed = parser.add_argument_group('required named arguments')
requiredNamed.add_argument(
'--path_to_input_image',
type=str,
required=True,
help=
'The path to the input image on which object detection will be performed on.\n\
This argument is required.')
parser.add_argument(
'--path_to_trained_model',
default='model_weights/coco_pretrained_weights.ckpt',
type=str,
help=
"The path to the location of pretrained model weights, which will be loaded into\n\
the model and then used for object detection. The default pretrained weights path is\n\
'model_weights/coco_pretrained_weights.ckpt', which contains weights trained on\n\
the coco dataset.")
parser.add_argument(
'--save_as',
type=str,
default=None,
help=
'The filename for the image on which object detection was performed. If no filename\n\
is provided, the image will be saved as "[original_name] + _yolo_v3.jpg".'
)
parser.add_argument('--tensorboard_save_path',
default='tensorboard/tensorboard_detect/',
help="")
parser.add_argument(
'--class_path',
default='utils/coco_classes.txt',
type=str,
help=
'The path that points towards where the class names for the dataset are stored.\n\
The default path is "utils/coco_classes.txt".')
parser.add_argument(
'--anchors_path',
default='utils/anchors.txt',
type=str,
help=
'The path that points towards where the anchor values for the model are stored.\n\
The default path is "utils/anchors.txt", which contains anchors trained on the coco dataset.'
)
parser.add_argument(
'--input_height',
default=416,
type=int,
help=
'The input height of the yolov3 model. The height must be a multiple of 32.\n\
The default height is 416.')
parser.add_argument(
'--input_width',
default=416,
type=int,
help=
'The input width of the yolov3 model. The width must be a mutliple of 32.\n\
The default width is 416.')
args = vars(parser.parse_args())
h = args['input_height']
w = args['input_width']
anchors = get_anchors(args['anchors_path'])
classes = get_classes(args['class_path'])
save_as = args['save_as']
if save_as is None:
filename_w_ext = os.path.basename(args['path_to_input_image'])
filename, file_extension = os.path.splitext(filename_w_ext)
save_as = filename + '_yolo_v3' + file_extension
image, original_im = process_image(args['path_to_input_image'], h, w)
tf.reset_default_graph()
# build graph
with tf.variable_scope('x_input'):
X = tf.placeholder(dtype=tf.float32, shape=[None, h, w, 3])
yolo_outputs = yolo_v3(inputs=X,
num_classes=len(classes),
anchors=anchors,
h=h,
w=w,
training=False) # output
with tf.variable_scope('obj_detections'):
raw_outputs = tf.concat(yolo_outputs, axis=1)
# pass image through model
with tf.Session() as sess:
writer = tf.summary.FileWriter(args['tensorboard_save_path'],
sess.graph)
writer.close()
saver = tf.train.Saver()
print('restoring model weights...')
saver.restore(sess, save_path=args['path_to_trained_model'])
print('feeding image found at filepath: ', args['path_to_input_image'])
start = time.time()
ro = sess.run(raw_outputs,
feed_dict={X: [np.array(image, dtype=np.float32)]})
end = time.time()
total_time = end - start
print("total inference time was: " + str(round(total_time, 2)) +
" seconds (that's " + str(round(60.0 / total_time, 2)) +
" fps!)")
# convert box coordinates, apply nms, and draw boxes
boxes = convert_box_coordinates(ro)
filtered_boxes = non_max_suppression(boxes,
confidence_threshold=0.5,
iou_threshold=0.4)
draw_boxes(save_as, args['class_path'], filtered_boxes, original_im, image)
print('image with detections saved as: ', save_as)
detected_boxes = sess.run(boxes, feed_dict={inputs: [img_processed]})
t1 = time.time()
print("Amount of seconds to predict:", t1 - t0)
# non max supression
filtered_boxes = non_max_suppression(
detected_boxes,
confidence_threshold=_CONF_THRESHOLD,
iou_threshold=_IOU_THRESHOLD
)
draw_boxes(
filtered_boxes,
img,
classes,
_INPUT_SIZE,
True,
width=2
)
open_cv_image = np.array(img)
open_cv_image = cv2.cvtColor(open_cv_image, cv2.COLOR_RGB2BGR)
# show frame to user
cv2.imshow('frame', open_cv_image)
# close windows when pressing 'q'
if cv2.waitKey(1) & 0xFF == ord('q'):
retrieving_frames = False
cv2.destroyAllWindows()
tmp = np.zeros(self.img.shape[:2], dtype='uint8')
tmp = draw_lines(tmp, rowboxes + colboxes, color=255, lineW=2)
labels = measure.label(tmp < 255, connectivity=2) #8连通区域标记
regions = measure.regionprops(labels)
ceilboxes = minAreaRectbox(regions, False, tmp.shape[1],
tmp.shape[0], True, True)
ceilboxes = np.array(ceilboxes)
ceilboxes[:, [0, 2, 4, 6]] += xmin
ceilboxes[:, [1, 3, 5, 7]] += ymin
self.tableCeilBoxes.extend(ceilboxes)
self.childImgs.append(childImg)
def table_ocr(self):
pass
if __name__ == '__main__':
import time
from utils.utils import draw_boxes
p = 'img/table-detect.jpg'
img = cv2.imread(p)
t = time.time()
tableDetect = table(img)
tableCeilBoxes = tableDetect.tableCeilBoxes
img = tableDetect.img
tmp = np.zeros_like(img)
img = draw_boxes(tmp, tableDetect.tableCeilBoxes, color=(255, 255, 255))
print(time.time() - t)
cv2.imwrite('img/table-ceil.png', img)