Accessing Ximea API via OpenCV in Python

Question

OpenCV has a great support for XIMEA cameras. However, I cannot find a way to access the XIMEA-specific functions and key words. I found on other sites (and on Stackoverflow), that there are key words like CV_CAP_XIAPI or CV_CAP_PROP_XI_DOWNSAMPLING, but I can't find them in Python.

In detail, the cameras have an auto-exposure feature and I would like to deactivate that. If I set the CV_CAP_PROP_EXPOSURE to a specific value, it changes, however, after a couple of (milli)seconds, the auto-exposure takes over again. Is there a way to force the camera to stick to that value?

Thank you for your help.

Best, anki

Btw, there's a direct XIMEA Python binding called pyximea, but I haven't solved building it on Windows yet, so, I rely on OpenCV (which is great and would be even better when I could directly access the API).

I can see those methods in OpenCV 3.0 by doing `import cv2` and `cv2.CV_CAP_XIAPI` — Ryan, Jul 27 '15 at 07:55
Ok, I am having 2.4.11, probably it's worth installing cv3 then ;) — anki, Jul 27 '15 at 07:57
Might be worth trying `from cv2 import cv` and see if you can see them in `cv` — Ryan, Jul 27 '15 at 08:00
I already did this, but haven't found it. Now I probably have to change to Python 3.x because OpenCV 3 would love to see Python 3... — anki, Jul 27 '15 at 08:03
I'm using OpenCV 3 with python 2.7; you don't need to switch to python 3 — Ryan, Jul 27 '15 at 08:10

score 0 · Accepted Answer · answered Aug 05 '15 at 09:09

0

As outlined by Ryan, the OpenCV version is crucial. Building OpenCV 3.0.0 from Source including the option WITH_XIMEA for Python 2.7.x works perfect. I offer a downloadable compiled package on my blog. Thanks!

answered Aug 05 '15 at 09:09

anki

765
5
14

A short update: I switched now to Python 3.5.x, works beautifully with OpenCV 3.1. And vice versa, OpenCV 3.1 with Python 2.7.x. – anki May 10 '16 at 20:05

Accessing Ximea API via OpenCV in Python

1 Answers1