I am trying to get OpenCV to work with the See3Cam CU135, but as far as I can tell openCV is not compatible with the image format of the camera (Y16) and the manufacturer suggests making a custom openCV installation. I have tried the custom installation to no avail . Does anyone have any ideas how one could aproach the problem? Thanks
Asked
Active
Viewed 650 times
0
-
In what way is it incompatible? It sounds like a greyscale image of `np.uint16` type - but sadly the manufacturer requires you to login to read the documentation for some reason??? Maybe you have a sample image capture you can share - using Google Drive or Dropbox or similar - along with details of its resolution. – Mark Setchell Feb 04 '22 at 16:14
-
1It looks like the questions was [asked before](https://stackoverflow.com/questions/34985388/cannot-read-image-with-see3cam-with-python-3-and-open-cv). I am not sure about the solution. In case you are using Windows, you may try the configuration used [here](https://stackoverflow.com/questions/66909370/thermal-image-processing). – Rotem Feb 04 '22 at 21:56
-
Thanks for the replies. I have no idea why it is classed as a greyscale image, as the output is coloured. I posted the solution that works for me down below. – webs397 Feb 05 '22 at 09:09
2 Answers
1
I found a solution that works for me here for anyone that has the same problem.
The important part is how the camera is connected to windows directshow:
import cv2
device_index = 0
cap = cv2.VideoCapture(device_index+cv2.CAP_DSHOW)
cap.set(cv2.CAP_PROP_FRAME_HEIGHT, 512)
cap.set(cv2.CAP_PROP_FRAME_WIDTH, 640)
cap.set(cv2.CAP_PROP_FOURCC, cv2.VideoWriter.fourcc('Y','1','6',' '))
cap.set(cv2.CAP_PROP_CONVERT_RGB, 0)
Then after that the usual while loop for streaming (I had to change the device_index to 1)

webs397
- 11
- 3
1
See3CAM_CU135 camera supported output is UYVY and MJPEG not Y16. Because See3CAM_CU135 camera has an onboard ISP to perform major image processing functions like debayering, WB control, Color correction and various other image quality improvement features, which also reduces the processing load in the host platform. So you could not set Y16 in See3CAM_CU135 camera.