Last active
January 20, 2020 17:06
-
-
Save Namburger/b64502ca017399c9427ccd909a5b914e to your computer and use it in GitHub Desktop.
An example of doing object detection with open cv on the dev board
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Copyright 2019 Google LLC | |
# | |
# Changes made by Nam Vu 01/20/2020 | |
# | |
# Licensed under the Apache License, Version 2.0 (the "License"); | |
# you may not use this file except in compliance with the License. | |
# You may obtain a copy of the License at | |
# | |
# https://www.apache.org/licenses/LICENSE-2.0 | |
# | |
# Unless required by applicable law or agreed to in writing, software | |
# distributed under the License is distributed on an "AS IS" BASIS, | |
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | |
# See the License for the specific language governing permissions and | |
# limitations under the License. | |
r"""A demo for object detection. | |
Model and Label File taken from here: | |
https://github.com/google-coral/edgetpu/tree/master/test_data | |
$ python3 object_detection.py --model mobilenet_ssd_v2_coco_quant_postprocess_edgetpu.tflite --label coco_labels.txt | |
""" | |
import argparse | |
import platform | |
import subprocess | |
from edgetpu.detection.engine import DetectionEngine | |
from edgetpu.utils import dataset_utils | |
import cv2 | |
import numpy as np | |
from PIL import Image | |
from PIL import ImageDraw | |
def do_detection(engine, img, k=3): | |
# Run inference. | |
ans = engine.detect_with_image( | |
img, | |
threshold=0.05, | |
keep_aspect_ratio=False, | |
relative_coord=False, | |
top_k=k) | |
return ans | |
def main(): | |
parser = argparse.ArgumentParser() | |
parser.add_argument( | |
'--model', | |
help='Path of the detection model, it must be a SSD model with postprocessing operator.', | |
required=True) | |
parser.add_argument('--label', help='Path of the labels file.') | |
parser.add_argument('--top_k', help='Top number to results to return.') | |
args = parser.parse_args() | |
# Initialize engine. | |
engine = DetectionEngine(args.model) | |
labels = dataset_utils.read_label_file(args.label) if args.label else None | |
top_k = args.top_k if args.top_k else 3 | |
# capturing videos with cv2 | |
cap = cv2.VideoCapture(0) | |
# looping through all frames | |
frame_num = 0 | |
while True: | |
print('<<<----------------------------------------->>>') | |
print('Frame number: ', frame_num) | |
ret, cv2img = cap.read() | |
cv2img = cv2.cvtColor(cv2img, cv2.COLOR_BGR2RGB) | |
img = Image.fromarray(cv2img) | |
draw = ImageDraw.Draw(img) | |
ans = do_detection(engine, img) | |
# Save result. | |
if ans: | |
for obj in ans: | |
print('-----------------------------------------') | |
if labels: | |
print(labels[obj.label_id]) | |
print('score = ', obj.score) | |
box = obj.bounding_box.flatten().tolist() | |
print('box=', box) | |
draw.rectangle(box, outline='red') | |
else: | |
print('No object detected!') | |
new_cv2img = np.array(img.convert('RGB')) | |
new_cv2img = new_cv2img[:, :, ::-1].copy() # RGB to BGR | |
cv2.imshow('Object Detection', new_cv2img) | |
if cv2.waitKey(1) & 0xFF == ord('q'): | |
break | |
frame_num += 1 | |
print('\n') | |
cap.release() | |
cv2.destroyAllWindows() | |
if __name__ == '__main__': | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment