Ask Your Question

Tetragramm's profile - activity

2018-05-20 17:41:07 -0600 answered a question Covariance matrices for Kalman Filter

There are three covariance matrices. Each with different meanings. I'm going to use a hypothetical scenario of trackin

2018-05-20 17:13:29 -0600 answered a question Calculate Mean: different result for masked image vs ROI

What's happening is that your borders are different. In technique 1, Sobel uses BORDER_REFLECT101 to reflect the edges

2018-05-08 18:23:59 -0600 commented answer 'Mapping' Aruco markers?

You can test by first, ensuring that if you do it twice, you end up with what you started with. Second, use a known rot

2018-05-08 17:18:47 -0600 commented answer 'Mapping' Aruco markers?

Very close, but number 4 is the matrix product, and the output should be another 3x1 matrix.

2018-05-01 19:16:56 -0600 commented question Parking distance calculation for a car

You probably want to use the background segmentation functions. They will definitely help with detecting the car, and p

2018-04-18 17:25:15 -0600 commented answer Obtaining Euler angles from Rodrigues Rotation Matrix

Absolutely not. I mean R_resultant = Rx1Ry1Rz1 = Rz2Ry2Rx2 = ... Rx1 will not equal Rx2 and so forth. You simply cho

2018-04-16 20:06:22 -0600 answered a question Obtaining Euler angles from Rodrigues Rotation Matrix

Good news! A rotation matrix is a rotation matrix is a rotation matrix. A rotation can be represented by several diffe

2018-04-02 22:59:13 -0600 commented question About SURF descriptors vector information

There are multiple types of information. Do you want a vector of the locations, or the descriptors, or something else?

2018-04-02 16:48:19 -0600 commented question Any easier way to do this?

Also note that what you were doing with the .at<uchar> and .at<int> is not safe and could not work. Make su

2018-04-02 16:44:40 -0600 commented answer Movidius or a GPU would be better

OpenCV has plenty of GPU support, but OpenCV is not a deep learning library. The DNN module(s) do not support the GPU,

2018-03-26 18:16:04 -0600 commented question Edge based alignment

Initial thoughts on sub-pixel. Sobel X and Y, the use the cartToPolar function to get gradient magnitude and orientatio

2018-03-26 18:01:13 -0600 commented question Phantom Intel IPP Asynchronous C/C++ Converters

I think this is a question best asked on the Intel forums. As it's about an Intel product, and the problem doesn't lie

2018-03-25 17:28:24 -0600 commented question Phantom Intel IPP Asynchronous C/C++ Converters

Do you have the full IPP? There is a small subset provided with OpenCV, but it is not the full set of functions. See H

2018-03-23 14:36:02 -0600 commented question Edge based alignment

First, that Gaussian Blur might account for most of the difference. Not sure on subpixel edges. You can calculate mo

2018-03-22 15:50:41 -0600 commented question Edge based alignment

Well, the threshold should chop off the same amount on either side, so it wouldn't bias the center too much. The rota

2018-03-22 15:48:51 -0600 commented answer SolvePnp results to Quaternion, euler flipping.

I don't know what to say. Look at the rotation matrices. They're basically identical. You can, if you're careful, a

2018-03-21 20:01:01 -0600 answered a question SolvePnp results to Quaternion, euler flipping.

So you aren't having a problem. It's just different rotational representations. Both of your examples have essentially

2018-03-20 21:18:35 -0600 commented question Appropriate data format for 3D transformations.

If you keep a vector of Point3f or Point3d, or a cv::Mat with 3 channels, you can apply the transform function to them.

2018-03-20 21:15:42 -0600 commented question Edge based alignment

Better solution. You can threshold the marker reliably, yes? Do that, and perform connectedComponentsWithStats. It sh

2018-03-20 21:11:31 -0600 commented question Edge based alignment

Idle thought before I finished reading your whole post. Not totally worthless, but... Try Phase Correlation. It's got

2018-03-20 17:40:22 -0600 answered a question Using cv::gpu::FAST_GPU with cv::gpu::PyrLKOpticalFlow in OpenCV 2.4.13.6.

Which version of OpenCV is this? 2.4. something? In 3.4.1, there's a built-in convert function. Aaand, in 2.4.13 it's

2018-03-20 07:55:18 -0600 commented answer Example of SparsePyrLKOpticalFlow

Check out the static function convert. Also, it's best to start a new question rather than resurrect old ones.

2018-03-19 21:11:51 -0600 answered a question Track ArUco marker in gray scale OpenCV image?

Why yes. Yes you can. https://docs.opencv.org/3.4.1/d5/dae/tutorial_aruco_detection.html

2018-03-18 18:29:00 -0600 commented question SolvePnp results to Quaternion, euler flipping.

Inverse != negating x,y,z, and w.

2018-03-17 12:46:17 -0600 commented question SolvePnp results to Quaternion, euler flipping.

You don't negate just w, you negate the whole quaternion.

2018-03-17 12:36:50 -0600 commented question SolvePnp results to Quaternion, euler flipping.

Remember that OpenCV is default Row Major, while Eigen is default Column Major. That means you may be transposing your

2018-03-16 18:55:32 -0600 commented question Recording and writing from multiple threaded PS3Eyes

Test 1: Can you verify with a single threaded program that you can open all of the cameras and display even one image fr

2018-03-16 18:54:27 -0600 commented question Recording and writing from multiple threaded PS3Eyes

Test 1: Can you verify with a single threaded program that you can open all of the cameras and display even one image fr

2018-03-14 17:40:21 -0600 answered a question ImageMagick / OpenCV equivalent

So apparently -normalize is just -contrast-stretch with specific arguments, so having both is somewhat redundant. Close

2018-03-14 17:27:32 -0600 commented question BufferedOutputStream bos = new BufferedOutputStream(sock.getOutputStream());

I think you're in the wrong place. I don't see any OpenCV code there at all.

2018-03-13 20:09:56 -0600 commented question Distances to surroundings in image

If I'm understanding your question right, you can use reduce to get the average of each column, ie: the average distance

2018-03-13 20:06:54 -0600 answered a question use multiple cameras to measure the distance between objects

It's not official, but I made THIS library. You'll need to calibrate the cameras and get their positions, but this libr

2018-03-08 19:55:47 -0600 answered a question opencv and Pleora PvBuffer

As it happens, yes. This is not really the kind of question this forum answers, but I happen to know. Or, really, you

2018-03-07 18:59:30 -0600 answered a question filter2D Filter out crosses

If you use a 3x3 opening operation with a box kernel, that should do it. It might do things like merge really nearby le

2018-03-01 21:20:11 -0600 answered a question How to find edge pixel coordinates at corners of the bounding box

So, uh, you see that code? You see the part labelled xLeftBottom, yLeftBottom, xRightTop, yRightTop? That's them. The

2018-02-26 18:24:18 -0600 commented question Real world position

I think you are using a different coordinate system than is normal. X and Y are typically camera pixel locations, with

2018-02-26 18:20:27 -0600 edited question Real world position

Real world position I got my camera calibration completed and the image looks very good. Now can I determine what my X,Y

2018-02-26 18:19:55 -0600 edited question Real world position

Real world position I got my camera calibration completed and the image looks very good. Now can I determine what my X,Y

2018-02-22 16:57:19 -0600 commented answer camera calibration

Oh, right. I think it's if the plane is parallel to the focal plane then?

2018-02-21 22:01:00 -0600 answered a question camera calibration

I believe so yes. Any two parallel lines in real life should remain parallel in the image, if it is properly calibrated

2018-02-20 22:09:45 -0600 answered a question Find dots on the surgeon's gloves

Try the niBlackThreshold ximgproc::niBlackThreshold( im, thresh, 255, THRESH_BINARY_INV, 21, -1.0 ); That gives me im

2018-02-20 17:03:30 -0600 commented question How detect very small blobs

For images like this, the Laplacian transform can be remarkably effective. Other useful algorithms might be Spectral Sa

2018-02-19 23:02:03 -0600 commented question How detect very small blobs

Why don't you post the color images, if they are a significantly different color, that would be useful. Also try a loca

2018-02-19 18:32:50 -0600 commented question Camera imager and "X" position location

I'm having some trouble visualizing what you mean. Could you make a diagram, and/or some sample images?

2018-02-18 19:58:37 -0600 commented question inRange behaving differently when called from Objective C++

Have you tried artificial images? Take an image with a series of different brightness squares, then change your thresho

2018-02-17 14:51:38 -0600 answered a question ive made a text game game engine that use's opencv and i whant to copirightt it is that ok?

Yes, so long as you follow the terms of the BSD license. Which isn't hard. You have to acknolwedge that you use OpenCV,

2018-02-16 15:47:17 -0600 commented answer How can i fill the gaps in the foreground and improve foreground extraction?

So, you still aren't really reading what these functions are doing. Take the code from your link and insert it before

2018-02-15 17:47:20 -0600 commented answer How can i fill the gaps in the foreground and improve foreground extraction?

Try copying the frame to a black image, using the fgmask.

2018-02-15 11:14:02 -0600 commented answer How can i fill the gaps in the foreground and improve foreground extraction?

You need to read a bit more carefully. Notice how in the tutorial, Opening is white on a black background. In your cas

2018-02-15 10:41:08 -0600 answered a question How can i fill the gaps in the foreground and improve foreground extraction?

Take a look at this tutorial: Morphological Operations