def main():
parser = argparse.ArgumentParser()
parser.add_argument('--model',
help='File path of Tflite model.',
required=True)
parser.add_argument('--label',
help='File path of label file.',
required=True)
args = parser.parse_args()
with open(args.label, 'r', encoding="utf-8") as f:
pairs = (l.strip().split(maxsplit=1) for l in f.readlines())
labels = dict((int(k), v) for k, v in pairs)
engine = edgetpu.classification.engine.ClassificationEngine(args.model)
with picamera.PiCamera() as camera:
camera.resolution = (640, 480)
camera.framerate = 30
_, width, height, channels = engine.get_input_tensor_shape()
camera.start_preview()
try:
stream = io.BytesIO()
for foo in camera.capture_continuous(stream,
format='rgb',
use_video_port=True,
resize=(width, height)):
stream.truncate()
stream.seek(0)
input = np.frombuffer(stream.getvalue(), dtype=np.uint8)
start_ms = time.time()
results = engine.ClassifyWithInputTensor(input, top_k=1)
elapsed_ms = time.time() - start_ms
if results:
camera.annotate_text = "%s %.2f\n%.2fms" % (labels[
results[0][0]], results[0][1], elapsed_ms * 1000.0)
finally:
camera.stop_preview()
def authorized_get(self):
if self.path == '/':
self.send_response(301)
self.send_header('Location', '/index.html')
self.end_headers()
elif self.path == '/index.html':
content = PAGE.encode('utf-8')
self.send_response(200)
self.send_header('Content-Type', 'text/html')
self.send_header('Content-Length', len(content))
self.end_headers()
self.wfile.write(content)
elif self.path == '/stream.mjpg':
self.send_response(200)
self.send_header('Age', 0)
self.send_header('Cache-Control', 'no-cache, private')
self.send_header('Pragma', 'no-cache')
self.send_header('Content-Type', 'multipart/x-mixed-replace; boundary=FRAME')
self.end_headers()
try:
stream_video = io.BytesIO()
stream_tpu = io.BytesIO()
_, width, height, channels = engine.get_input_tensor_shape()
while True:
camera.capture(stream_tpu,
format='rgb',
use_video_port=True,
resize=(width, height))
stream_tpu.truncate()
stream_tpu.seek(0)
input = np.frombuffer(stream_tpu.getvalue(), dtype=np.uint8)
start_ms = time.time()
results = engine.ClassifyWithInputTensor(input, top_k=1)
elapsed_ms = time.time() - start_ms
if results:
camera.annotate_text = "%s %.2f\n%.2fms" % (
labels[results[0][0]], results[0][1], elapsed_ms*1000.0)
camera.capture(stream_video, format='jpeg', use_video_port=True)
stream_video.truncate()
stream_video.seek(0)
self.wfile.write(b'--FRAME\r\n')
self.send_header('Content-Type', 'image/jpeg')
self.send_header('Content-Length', len(stream_video.getvalue()))
self.end_headers()
self.wfile.write(stream_video.getvalue())
self.wfile.write(b'\r\n')
except Exception as e:
logging.warning(
'Removed streaming client %s: %s',
self.client_address, str(e))
else:
self.send_error(404)
self.end_headers()
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--model',
help='File path of Tflite model.',
required=True)
parser.add_argument('--label',
help='File path of label file.',
required=True)
args = parser.parse_args()
with open(args.label, 'r', encoding="utf-8") as f:
pairs = (l.strip().split(maxsplit=1) for l in f.readlines())
labels = dict((int(k), v) for k, v in pairs)
engine = edgetpu.classification.engine.ClassificationEngine(args.model)
camera = cv2.VideoCapture(0)
if camera:
camera.set(3, 640)
camera.set(4, 480)
font = cv2.FONT_HERSHEY_SIMPLEX
bottomLeftCornerOfText = (10, 470)
fontScale = 0.6
fontColor = (255, 255, 255)
lineType = 2
annotate_text = ""
annotate_text_time = time.time()
time_to_show_prediction = 3.0
min_confidence = 0.2
_, width, height, channels = engine.get_input_tensor_shape()
try:
while True:
if not camera.isOpened():
continue
ret, img = camera.read()
if not ret:
continue
input = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
input = cv2.resize(input, (width, height))
input = input.reshape((width * height * channels))
start_ms = time.time()
results = engine.ClassifyWithInputTensor(input, top_k=1)
elapsed_ms = time.time() - start_ms
if results:
print("%s %.2f\n%.2fms" %
(labels[results[0][0]], results[0][1],
elapsed_ms * 1000.0))
finally:
camera.release()
def main():
default_model_dir = "../all_models"
default_model = 'mobilenet_v2_1.0_224_quant_edgetpu.tflite'
default_labels = 'imagenet_labels.txt'
parser = argparse.ArgumentParser()
parser.add_argument('--model',
help='.tflite model path',
default=os.path.join(default_model_dir, default_model))
parser.add_argument('--labels',
help='label file path',
default=os.path.join(default_model_dir,
default_labels))
args = parser.parse_args()
with open(args.label, 'r') as f:
pairs = (l.strip().split(maxsplit=1) for l in f.readlines())
labels = dict((int(k), v) for k, v in pairs)
engine = edgetpu.classification.engine.ClassificationEngine(args.model)
with picamera.PiCamera() as camera:
camera.resolution = (640, 480)
camera.framerate = 30
camera.annotate_text_size = 20
_, width, height, channels = engine.get_input_tensor_shape()
camera.start_preview()
try:
stream = io.BytesIO()
fps = deque(maxlen=20)
fps.append(time.time())
for foo in camera.capture_continuous(stream,
format='rgb',
use_video_port=True,
resize=(width, height)):
stream.truncate()
stream.seek(0)
input = np.frombuffer(stream.getvalue(), dtype=np.uint8)
start_ms = time.time()
results = engine.ClassifyWithInputTensor(input, top_k=3)
inference_ms = (time.time() - start_ms) * 1000.0
fps.append(time.time())
fps_ms = len(fps) / (fps[-1] - fps[0])
camera.annotate_text = "Inference: %5.2fms FPS: %3.1f" % (
inference_ms, fps_ms)
for result in results:
camera.annotate_text += "\n%.0f%% %s" % (100 * result[1],
labels[result[0]])
print(camera.annotate_text)
finally:
camera.stop_preview()
def main():
default_model_dir = "../all_models"
default_model = 'mobilenet_v2_1.0_224_quant_edgetpu.tflite'
default_labels = 'imagenet_labels.txt'
parser = argparse.ArgumentParser()
parser.add_argument('--model',
help='.tflite model path',
default=os.path.join(default_model_dir, default_model))
parser.add_argument('--labels',
help='label file path',
default=os.path.join(default_model_dir,
default_labels))
args = parser.parse_args()
with open(args.labels, 'r') as f:
pairs = (l.strip().split(maxsplit=1) for l in f.readlines())
labels = dict((int(k), v) for k, v in pairs)
engine = edgetpu.classification.engine.ClassificationEngine(args.model)
pygame.init()
pygame.camera.init()
camlist = pygame.camera.list_cameras()
camera = pygame.camera.Camera(camlist[0], (640, 480))
_, width, height, channels = engine.get_input_tensor_shape()
camera.start()
try:
fps = deque(maxlen=20)
fps.append(time.time())
while True:
imagen = camera.get_image()
imagen = pygame.transform.scale(imagen, (width, height))
input = np.frombuffer(imagen.get_buffer(), dtype=np.uint8)
start_ms = time.time()
results = engine.ClassifyWithInputTensor(input, top_k=3)
inference_ms = (time.time() - start_ms) * 1000.0
fps.append(time.time())
fps_ms = len(fps) / (fps[-1] - fps[0])
annotate_text = "Inference: %5.2fms FPS: %3.1f" % (inference_ms,
fps_ms)
for result in results:
annotate_text += "\n%.0f%% %s" % (100 * result[1],
labels[result[0]])
print(annotate_text)
finally:
camera.stop()
def main(model_path):
label_file_path = '../label/for_lt_20190731/label.csv'
labels = get_labels_from_csv(label_file_path)
engine = edgetpu.classification.engine.ClassificationEngine(model_path)
with picamera.PiCamera() as camera:
camera.resolution = (225, 225)
# camera.resolution = (640, 480)
camera.vflip = True
camera.hflip = True
camera.framerate = 30
_, width, height, channels = engine.get_input_tensor_shape()
camera.start_preview()
try:
stream = io.BytesIO()
for foo in camera.capture_continuous(stream,
format='rgb',
use_video_port=True,
resize=(width, height)):
stream.truncate()
stream.seek(0)
input = np.frombuffer(stream.getvalue(), dtype=np.uint8)
start_ms = time.time()
results = engine.ClassifyWithInputTensor(input, top_k=5)
elapsed_ms = time.time() - start_ms
if results:
# camera.annotate_text = "%s %.2f\n%.2fms" % (
# labels[results[0][0]], results[0][1], elapsed_ms*1000.0)
if results[0][1] < 0.2:
label = ""
target_value = 0.0
else:
label = labels[results[0][0]]
target_value = results[0][1]
camera.annotate_text = "%s %.1f\n%.2fms" % (
label, target_value, elapsed_ms*1000.0)
print(results)
finally:
camera.stop_preview()
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
'--model', help='File path of Tflite model.', required=True)
parser.add_argument(
'--label', help='File path of label file.', required=True)
args = parser.parse_args()
with open(args.label, 'r') as f:
pairs = (l.strip().split(maxsplit=1) for l in f.readlines())
labels = dict((int(k), v) for k, v in pairs)
engine = edgetpu.classification.engine.ClassificationEngine(args.model)
with picamera.PiCamera() as camera:
camera.resolution = (640, 480)
camera.framerate = 30
_, width, height, channels = engine.get_input_tensor_shape()
camera.start_preview()
try:
stream = io.BytesIO()
for foo in camera.capture_continuous(stream,
format='rgb',
use_video_port=True,
resize=(width, height)):
stream.truncate()
stream.seek(0)
input = np.frombuffer(stream.getvalue(), dtype=np.uint8)
start_ms = time.time()
results = engine.ClassifyWithInputTensor(input, top_k=1)
elapsed_ms = time.time() - start_ms
if results:
camera.annotate_text = "%s %.2f\n%.2fms" % (
labels[results[0][0]], results[0][1], elapsed_ms*1000.0)
finally:
camera.stop_preview()
import time
from time import sleep
from picamera import PiCamera
import edgetpu.classification.engine
from io import BytesIO
import numpy as np
if __name__ == "__main__":
print('Starting up')
engine = edgetpu.classification.engine.ClassificationEngine('test_data/mobilenet_v2_1.0_224_quant_edgetpu.tflite')
print('Edge TPU initialised')
_, width, height, channels = engine.get_input_tensor_shape()
with open('test_data/imagenet_labels.txt', 'r') as f:
pairs = (l.strip().split(maxsplit=1) for l in f.readlines())
labels = dict((int(k), v) for k, v in pairs)
print('Labels read')
with PiCamera(resolution='800x600', framerate=30, sensor_mode=2) as camera:
print('Camera initialised')
camera.start_preview()
# Camera warm-up time
sleep(2)
print('Starting classify_image_stream')
while True:
stream = BytesIO()
camera.capture(stream,
format='rgb',
use_video_port=True,
resize=(width, height))
def authorized_get(self):
if self.path == '/':
self.send_response(301)
self.send_header('Location', '/index.html')
self.end_headers()
elif self.path == '/index.html':
content = PAGE.encode('utf-8')
self.send_response(200)
self.send_header('Content-Type', 'text/html')
self.send_header('Content-Length', len(content))
self.end_headers()
self.wfile.write(content)
elif self.path == '/stream.mjpg':
self.send_response(200)
self.send_header('Age', 0)
self.send_header('Cache-Control', 'no-cache, private')
self.send_header('Pragma', 'no-cache')
self.send_header('Content-Type',
'multipart/x-mixed-replace; boundary=FRAME')
self.end_headers()
font = cv2.FONT_HERSHEY_SIMPLEX
try:
stream_video = io.BytesIO()
stream_tpu = io.BytesIO()
_, width, height, channels = engine.get_input_tensor_shape()
while True:
ret, color_image = cap.read()
if not ret:
break
prepimg = color_image[:, :, ::-1].copy()
prepimg = cv2.resize(prepimg, (width, height))
prepimg = np.reshape(prepimg, (width * height * 3, ))
start_ms = time.time()
results = engine.ClassifyWithInputTensor(prepimg, top_k=1)
elapsed_ms = time.time() - start_ms
if results:
caption = "%s %.2f\n%.2fms" % (labels[results[0][0]],
results[0][1],
elapsed_ms * 1000.0)
else:
caption = "..."
imgRGB = cv2.cvtColor(color_image, cv2.COLOR_BGR2RGB)
cv2.putText(imgRGB, caption, (0, 30), font, 1,
(255, 255, 255), 2, cv2.LINE_AA)
jpg = Image.fromarray(imgRGB)
jpg.save(stream_video, 'JPEG')
stream_video.truncate()
stream_video.seek(0)
self.wfile.write(b'--FRAME\r\n')
# self.wfile.write(b"--jpgboundary")
self.send_header('Content-Type', 'image/jpeg')
self.send_header('Content-Length',
len(stream_video.getvalue()))
self.end_headers()
self.wfile.write(stream_video.getvalue())
self.wfile.write(b'\r\n')
except Exception as e:
logging.warning('Removed streaming client %s: %s',
self.client_address, str(e))
else:
self.send_error(404)
self.end_headers()
