I can't find anything similar to Savitzky Golay Polynomial Fit on opencv. This is a standard smoothing operation though, so it seems like something they should have. Does anybody know of anything they have? Using C++ for what its worth.
Thanks!
-Tim
It is not clear what you need to do: fit or smooth, you mentioned both. But if you need to smooth using OpenCV you can try Kalman filter (fit in its way and smooth), smooth-2D (using your 1D-data) or your own convolution smooth kernel 1D+1D using 1D-data for kernelX only convolution (the fastest way to smooth).
OpenCV is near to real time image and video processing library, and it contains most common task solvers for this, where is no polynomial fitting among them yet. But if you really need fitting (not just smoothing) you can use polynomial fitting matrix equation and calculate your answer by yourself in a simple way thanks to Mat objects in OpenCV which has inv()(inverse) and t()(transpose) functions.
Related
I need to implement the same logic and values we get in MATLAB using interp1 with 'pchip' or 'cubic' flags and I'm unable to find a fitting replacement in OpenCV other than implementing by myself the cubic interpolation in Numerical Recipes (as noted by another question, this is based on De Boor's algorithm, which is used by MATLAB).
We need to interpolate values of a 1D doubles vector based on our sample points. Using Linear interpolation did not yield sufficient results as it is not smooth enough and results in non-continuity on the gradient in the joint points.
I came across this on OpenCV's website. But this seems to only be bicubic and work on an image, whilst I need a cubic interpolation.
Does anyone know any other function or simple solution on OpenCV's side for this issue? Any suggestion would help, thanks.
Side note: we are using OpenCV 4.3.0.
I need to compute the convolution between an input image and the derivative of a Gaussian kernel. This question has been already asked (How to apply a partial derivative Gaussian kernel to an image with OpenCV?), but it's not clear if the proposed solution refers to a simple Gaussian kernel or to its derivative.
I'm not a great expert of image filtering, but searching on the web I found out that the Sobel operator is an approximation of a gaussian kernel derivative. So, can I directly use it on my input image to get my convolution? Or, do I need to apply it on a gaussian kernel? Or is there still an even better way to proceed?
Thank you in advance
Thresholded Image
BGR Image
Fitted Thresholded Image
Hi all. I'm working on a project about computer vision using OpenCV for C++ interface. My purpose is to track a moving deformable object that is marked with a colored tape. By processing each frame of the video I'm able to effectively isolate the color (as you can see in the thresholded image) and track its trajectory, movement and shape into the BGR image.
My problem is that I need to extrapolate an equation or polynomial that can describe the current shape assumed by my tracked object.
Is there an effective way to do this? I've no idea on how to address the problem.
Thanks in advance,
Cheers!
If your final goal is to detect your shape in various forms i think you want to read about Active shape model: https://en.wikipedia.org/wiki/Active_shape_model
If you just want to get a polynomial fit of the shape in each instance of time i would use the suggestion of Cherkesgiller Tural and read about 2D curve fitting.
If I understood correctly:
I would start to fit a polygon on your shape. A common method for that is alpha-shapes.
You can also try an optimization approach which is enormously powerful because you can basically design your cost-function and constrains however you want. But it is computationally very costly (depending on the algorithm).
Have a look at this thread: It might help you.
I have a 3D image data obtained from a 3D OCT scan. The data can be represented as I(x,y,z) which means there is an intensity value at each voxel.
I am writing an algorithm which involves finding the image's gradient in x,y and z directions in C++. I've already written a code in C++ using OpenCV for 2D and want to extend it to 3D with minimal changes in my existing code for 2D.
I am familiar with 2D gradients using Sobel or Scharr operators. My search brought me to this post, answers to which recommend ITK and Point Cloud Library. However, these libraries have a lot more functionalities which might not be required. Since I am not very experienced with C++, these libraries require a bit of reading, which time doesn't permit me. Moreover, these libraries don't use cv::Mat object. If I use anything other than cv::Mat, my whole code might have to be changed.
Can anyone help me with this please?
Update 1: Possible solution using kernel separability
Based on #Photon's answer, I'm updating the question.
From what #Photon says, I get an idea of how to construct a Sobel kernel in 3D. However, even if I construct a 3x3x3 cube, how to implement it in OpenCV? The convolution operations in OpenCV using filter2d are only for 2D.
There can be one way. Since the Sobel kernel is separable, it means that we can break the 3D convolution into convolution in lower dimensions. Comments 20 and 21 of this link also tell the same thing. Now, we can separate the 3D kernel but even then filter2D cannot be used since the image is still in 3D. Is there a way to break down the image as well? There is an interesting post which hints at something like this. Any further ideas on this?
Since the Sobel operator is separable, it's easy to envision how to add a 3rd dimension.
For example, when you look at the filter definition for Gx in the link you posted, you see that is multiplies the surrounding pixels by coefficients that have a sign dependent on the relative X position, and magnitude relative to the offset in Y.
When you extend to 3D, the Gx gradient should be calculated the same way, but you need to work on a 3x3x3 cube, and the coefficient sign remains the same definition, and the magnitude now depends on change in either Y or Z or both.
The other gradients (Gy, Gz) are the same but around their axis.
I am trying to use two Gaussian mixtures with EM algorithm to estimate color distribution of a video frame. For that, I want to use two separate peaks in the color distribution as the two Gaussian means to facilitate the EM calculation. I have several difficulties with the implementation of these in OpenCV.
My first question is: how can I determine the two peaks? I've searched about peak estimation in OpenCV, but still couldn't find any seperate function. So I am going to determine two regions, then find their maximum values as peaks. Is this way correct?
My second question is: how to perform Gaussian mixture model with EM in OpenCV? As far as I know, the "cv::EM::predict" function could give me the index of the most probable mixture component. But I have difficulties with training EM. I've searched and found some other codes, but finding the correct parameters is too much difficult for. Could someone provide me any example code for this? Thank you in advance.
#ederman, try {OpenCV library location}\opencv\samples\cpp\em.cpp instead of the web link. I think the sample code in the link is out of date now. I have successfully compiled the sample code in OpenCV 2.3.1. It shouldn't be a problem for 2.4.2.
Good luck:)
My first question is: how can I determine the two peaks?
I would iterate through the range of sample values possible, and test when the does EM.predict(sample)[0] peaks.