quadro's profile - activity

Merging low-angleillumination frames to edge-image Hi, I would like to Merge low-angle illumination frames to an edge-i

Hi,

I try to build a project in Visual Studio use OpenCV. I think it's made with VisualStudio 2013 and OpenCV 2.4.9 or earlier version. Unfortunately, the Visual Studio 2013 is no longer available, but the OpenCV 2.4.9 don't support Visual Studio 2015. So, I got this error below, anyone has an idea how can I fix it? (now I use VS2017 and OpenCV 3.2.0)

this was the project; https://github.com/uniform-marker-fie...

Thank you very much!

Yes, this is the problem, but I don't know what is this files;Here load the files, but I just find one in the directory;

# Load Calibrated Parameters
calibrationFile = "C:\Users\gergely\Desktop\kiserlet\python\OpenCV_Examples-master\OpenCV_Examples-master\01_internal_camera_calibration\calibration.yaml"
calibrationParams = cv2.FileStorage(calibrationFile, cv2.FILE_STORAGE_READ)
camera_matrix = calibrationParams.getNode("cameraMatrix").mat()
dist_coeffs = calibrationParams.getNode("distCoeffs").mat()
And I printed out;
"calibrationParams = <FileStorage 026640D0>
camera_matrix = None
dist_coeffs = None

Thanks you!

Hi,

I tried a OpenCV Aruco board project in Python; https://github.com/LongerVision/OpenC...

And I got the error below;

"map1, map2 = cv2.fisheye.initUndistortRectifyMap(camera_matrix, dist_coeffs, r, new_camera_matrix, image_size, cv2.CV_16SC2) cv2.error: C:\projects\opencv-python\opencv\modules\calib3d\src\fisheye.cpp:424: error: (-215) (K.depth() == CV_32F || K.depth() == CV_64F) && (D.depth() == CV_32F || D.depth() == CV_64F) in function cv::fisheye::initUndistortRectifyMap"

And this was the problematic line; map1, map2 = cv2.fisheye.initUndistortRectifyMap(camera_matrix, dist_coeffs, r, new_camera_matrix, image_size, cv2.CV_16SC2)

Anyone know solutions to fix it?

Thank you very much!

First, I think in a single RGB (maybe webcam) without any structured light... but; now I bought a ToF (Time-of-Flight Camera) is it good for this?

Thanks! It can be used with a simple webcam, or just work with deepth camera only? (like kinect)

Hi,

I'm seeking a solution in OpenCV to match a 3D scanned model (e.g an .STL file) with a real object from camera imput, to recognize it. Anyone had an idea, what is the best way to start?

Thank you very much!

Thanks a lot! @berak; I installed OpenCV 2.4.12, but I can't use "import cv", just "cv2" with this this error; import cv ImportError: DLL load failed, so the problem is the same :/ @sturkmen; thanks for the solution, I try to use it to build my own solution, but unfortunately it did not work with webcam as the previous (what I linked)

Thanks for the quick answer! I just want to recognize a rubik's cube with a camera and save the colors of the sides to a matrix

Hi,

I find this great script, https://sourceforge.net/projects/pyru... but it's use the old "import cv" instead of cv2. Is there any simple solution, to upgrade to cv2? especially the "cubefinder" folder is important for me!

Thank you very much!

Hi, sorry I find the solution! Thanks!

Hi, thanks for the quick answer! I just learn how to visualize different solutions to bounding the masked objects, but I stuck here because if we change the circle to rectangle, get the following error: "SystemError: new style getargs format but argument is not a tuple"

Thank you very much, works great!

Hi,

I would like to change the minEnclosingCircle to a BoundingBox from this example in Python; http://www.pyimagesearch.com/2015/09/...

Can someone help me in this?

Thank you very much!

Hi,

Maybe, I have a very stupid question, but how can I cover the whole area of the read area, when I used this color for object tracking? Because,

redLower = (0, 65, 100) redUpper = (20, 250, 250)

in this solution I have the area from H; 0-20, or 170-179, but how to cover this area together in Python? Example, from H; 170 to 20?

Thank you very much!