Investigate how to filter out moving background #2

Closed
opened 2021-07-13 09:27:13 +02:00 by Yun · 1 comment
Contributor

There may still be noise caused by bright pixels moving in the background.

One idea is to use opencv to calculate dense optical flow for each frame and use the portions with close to no movement as an additional mask that gets applied to the image.
Another idea would be to use background subtraction against the previous frame.
calculation of these masks will probably need to be done before the main multiprocessing part.

There may still be noise caused by bright pixels moving in the background. One idea is to use opencv to calculate dense [optical flow](https://docs.opencv.org/master/d4/dee/tutorial_optical_flow.html) for each frame and use the portions with close to no movement as an additional mask that gets applied to the image. Another idea would be to use [background subtraction](https://docs.opencv.org/3.4/d1/dc5/tutorial_background_subtraction.html) against the previous frame. calculation of these masks will probably need to be done before the main multiprocessing part.
Author
Contributor

Doesn't seem like this is that big of an issue with PaddleOCR

Doesn't seem like this is that big of an issue with PaddleOCR
Yun closed this issue 2021-07-16 20:09:19 +02:00
Sign in to join this conversation.
No Label
No Milestone
No project
No Assignees
1 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: pradana.aumars/videocr#2
No description provided.