Reputation: 133
I'm using OpenCV (3.4.6) to capture video from a Raspberry Pi HQ camera, attached to a Raspberry Pi 4 running Raspbian 10. Bounding boxes suggested by an object detection algorithm are then drawn onto the captured frames, and the output saved to a video file.
Everything works great, except when I try to crop the video to my region of interest.
Using the following line successfully crops the image (as confirmed by cv2.imshow
):
frame = frame[0:720, 280:1000]
But, it also stops the output being saved by out.write
- I just get an empty file, and being new to both Python and OpenCV, I'm struggling to work out why.
I've trimmed down the code to the relevant part below (apologies - it's still a bit long - I'm not entirely sure what might be important):
# begin video capture
try:
vid = cv2.VideoCapture(int(video_path), cv2.CAP_V4L2)
except:
vid = cv2.VideoCapture(video_path)
out = None
# set capture resolution
inputcodec = cv2.VideoWriter_fourcc('M', 'J', 'P', 'G') # forces MJPG codec
vid.set(6, inputcodec) # sets inputcodec for video capture
vid.set(3, 1280) # sets video capture width
vid.set(4, 720) # sets video capture height
if FLAGS.output:
# by default VideoCapture returns float instead of int
width = int(vid.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(vid.get(cv2.CAP_PROP_FRAME_HEIGHT))
fps = int(vid.get(cv2.CAP_PROP_FPS))
codec = cv2.VideoWriter_fourcc(*FLAGS.output_format)
out = cv2.VideoWriter(FLAGS.output, codec, fps, (width, height))
while True:
return_value, frame = vid.read()
frame = frame[0:720, 280:1000] # crop video to square region of interest
if return_value:
frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
image = Image.fromarray(frame)
else:
print('Video has ended or failed, try a different video format!')
break
frame_size = frame.shape[:2]
image_data = cv2.resize(frame, (input_size, input_size))
image_data = image_data / 255.
image_data = image_data[np.newaxis, ...].astype(np.float32)
start_time = time.time()
if FLAGS.framework == 'tflite':
interpreter.set_tensor(input_details[0]['index'], image_data)
interpreter.invoke()
pred = [interpreter.get_tensor(output_details[i]['index']) for i in range(len(output_details))]
if FLAGS.model == 'yolov3' and FLAGS.tiny == True:
boxes, pred_conf = filter_boxes(pred[1], pred[0], score_threshold=0.25,
input_shape=tf.constant([input_size, input_size]))
else:
boxes, pred_conf = filter_boxes(pred[0], pred[1], score_threshold=0.25,
input_shape=tf.constant([input_size, input_size]))
else:
batch_data = tf.constant(image_data)
pred_bbox = infer(batch_data)
for key, value in pred_bbox.items():
boxes = value[:, :, 0:4]
pred_conf = value[:, :, 4:]
boxes, scores, classes, valid_detections = tf.image.combined_non_max_suppression(
boxes=tf.reshape(boxes, (tf.shape(boxes)[0], -1, 1, 4)),
scores=tf.reshape(
pred_conf, (tf.shape(pred_conf)[0], -1, tf.shape(pred_conf)[-1])),
max_output_size_per_class=50,
max_total_size=50,
iou_threshold=FLAGS.iou,
score_threshold=FLAGS.score
)
pred_bbox = [boxes.numpy(), scores.numpy(), classes.numpy(), valid_detections.numpy()]
image = utils.draw_bbox(frame, pred_bbox)
fps = 1.0 / (time.time() - start_time)
print("FPS: %.2f" % fps)
result = np.asarray(image)
cv2.namedWindow("result", cv2.WINDOW_AUTOSIZE)
result = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
if not FLAGS.dont_show:
cv2.imshow("result", result)
if FLAGS.output:
out.write(result)
if cv2.waitKey(1) & 0xFF == ord('q'): break
cv2.destroyAllWindows()
Any suggestions for how to resolve this would be greatly appreciated.
Upvotes: 1
Views: 323
Reputation: 15407
You created the VideoWriter and set its width and height to 1280 by 720. Later you told it to save 720 by 720 crops. That is the problem. frame = frame[0:720, 280:1000]
is a 720 by 720 array.
Decide what dimensions your video should have, and use them consistently. A video stream can't change resolution.
Upvotes: 1