I am looking at the code for Hough transformation in image segmentation. The following code is from Computer Vision by Linda Shapiro. Can somebody tell me what is quantize_angle and how can I compute it?
The Hough transform looks for straight lines (or other features) in an image and represents these features as points in a different 2D coordinate system, where one axis represents the angle θ of a detected line, and the other represents the distance δ from this line to the centre of the image.
Source: Wikipedia
To produce a Hough transform of finite dimensions, both θ and δ have to be quantized. For example, if θ lies in the range (0 ≤ θ < 2π), then you could map it to the range 0–255 by a function such as the following:
int quantize_angle(float theta) {
int q = floor(theta * 128.0 / 3.141592654 + 0.5);
return q % 256;
}
This will result in a Hough transform that is 256 pixels wide.
Related
In the SURF technique, and more precisely within the feature description stage, the authors have stated (if I understand correctly) that the description will be performed in a area of 20 times sigma. Sigma represents the scale on which the keypoint was detected.
Sigma = 0.4 x L where L = 2^Octave x level+1. If we use the OpenCV implementation, the DetectAndCompute function computes, with the value of Keypoint.size, the radius of the circle surrounding the keypoint.
My question is : How could we get the value of sigma from the radius value ?
According to these lines:
KeyPoint& kp = (*keypoints)[k];
float size = kp.size;
Point2f center = kp.pt;
/* The sampling intervals and wavelet sized for selecting an orientation
and building the keypoint descriptor are defined relative to 's' */
float s = size*1.2f/9.0f;
This value s = size*1.2f/9.0f is not montioned in the bay's article scale= L*0.4 or
scale= L* 1.2/3 any one can explain me this part??
What is the algorithm used in OpenCV function convexityDefects() to calculate the convexity defects of a contour?
Please, describe and illustrate the high-level operation of the algorithm, along with its inputs and outputs.
Based on the documentation, the input are two lists of coordinates:
contour defining the original contour (red on the image below)
convexhull defining the convex hull corresponding to that contour (blue on the image below)
The algorithm works in the following manner:
If the contour or the hull contain 3 or less points, then the contour is always convex, and no more processing is needed. The algorithm assures that both the contour and the hull are accessed in the same orientation.
N.B.: In further explanation I assume they are in the same orientation, and ignore the details regarding representation of the floating point depth as an integer.
Then for each pair of adjacent hull points (H[i], H[i+1]), defining one edge of the convex hull, calculate the distance from the edge for each point on the contour C[n] that lies between H[i] and H[i+1] (excluding C[n] == H[i+1]). If the distance is greater than zero, then a defect is present. When a defect is present, record i, i+1, the maximum distance and the index (n) of the contour point where the maximum located.
Distance is calculated in the following manner:
dx0 = H[i+1].x - H[i].x
dy0 = H[i+1].y - H[i].y
if (dx0 is 0) and (dy0 is 0) then
scale = 0
else
scale = 1 / sqrt(dx0 * dx0 + dy0 * dy0)
dx = C[n].x - H[i].x
dy = C[n].y - H[i].y
distance = abs(-dy0 * dx + dx0 * dy) * scale
It may be easier to visualize in terms of vectors:
C: defect vector from H[i] to C[n]
H: hull edge vector from H[i] to H[i+1]
H_rot: hull edge vector H rotated 90 degrees
U_rot: unit vector in direction of H_rot
H components are [dx0, dy0], so rotating 90 degrees gives [-dy0, dx0].
scale is used to find U_rot from H_rot, but because divisions are more computationally expensive than multiplications, the inverse is used as an optimization. It's also pre-calculated before the loop over C[n] to avoid recomputing each iteration.
|H| = sqrt(dx0 * dx0 + dy0 * dy0)
U_rot = H_rot / |H| = H_rot * scale
Then, a dot product between C and U_rot gives the perpendicular distance from the defect point to the hull edge, and abs() is used to get a positive magnitude in any orientation.
distance = abs(U_rot.C) = abs(-dy0 * dx + dx0 * dy) * scale
In the scenario depicted on the above image, in first iteration, the edge is defined by H[0] and H[1]. The contour points tho examine for this edge are C[0], C[1], and C[2] (since C[3] == H[1]).
There are defects at C[1] and C[2]. The defect at C[1] is the deepest, so the algorithm will record (0, 1, 1, 50).
The next edge is defined by H[1] and H[2], and corresponding contour point C[3]. No defect is present, so nothing is recorded.
The next edge is defined by H[2] and H[3], and corresponding contour point C[4]. No defect is present, so nothing is recorded.
Since C[5] == H[3], the last contour point can be ignored -- there can't be a defect there.
In OpenCV how do you calculate the average gradient strength in a Mat and the average gradient direction?
I have sourced the below methods by googling but I want to confirm I am actually doing this correctly before moving onto the next step.
Is this correct?
Mat img = imread('foo.png', CV_8UC); // read image as grayscale single channel
// Calculate the mean intensity and the std deviation
// Any errors here or am I doing this correctly?
Scalar sMean, sStdDev;
meanStdDev(src, sMean, sStdDev);
double mean = sMean[0];
double stddev = sStdDev[0];
// Calculate the average gradient magnitude/strength across the image
// Any errors here or am I doing this correctly?
Mat dX, dY, magnitude;
Sobel(src, dX, CV_32F, 1, 0, 1);
Sobel(src, dY, CV_32F, 0, 1, 1);
magnitude(dX, dY, magnitude);
Scalar sMMean, sMStdDev;
meanStdDev(magnitude, sMMean, sMStdDev);
double magnitudeMean = sMMean[0];
double magnitudeStdDev = sMStdDev[0];
// Calculate the average gradient direction across the image
// Any errors here or am I doing this correctly?
Scalar avgHorizDir = mean(dX);
Scalar avgVertDir = mean(dY);
double avgDir = atan2(-avgVertDir[0], avgHorizDir[0]);
float blurriness = cv::videostab::calcBlurriness(src); // low values = sharper. High values = blurry
Technically those are the correct ways of obtaining the two averages.
The way you compute mean direction uses weighted directional statistics, meaning that pixels without a strong gradient have less influence on the average.
However, for most images this average direction is not very meaningful, as there exist edges in all directions and cancel out.
If your image is of a single edge, then this will work great.
If your image has lines in it, containing edges in opposite directions, this will not work. In this case, you want to average the double angle (average orientations). The obvious way of doing this is to compute the direction per pixel as an angle, double them, then use directional statistics to average (ie convert back to vectors and average those). Doubling the angle causes opposite directions to be mapped to the same value, thus averaging doesn’t cancel these out.
Another simple way to average orientations is to take the average of the tensor field obtained by the outer product of the gradient field with itself, and determine the direction of the eigenvector corresponding to the largest eigenvalue. The tensor field is obtained as follows:
Mat Sxx = dX * dX;
Mat Syy = dY * dY;
Mat Sxy = dX * dY;
This should then be averaged:
Scalar mSxx = mean(sXX);
Scalar mSyy = mean(sYY);
Scalar mSxy = mean(sXY);
These values form a 2x2 real-valued symmetric matrix:
| mSxx mSxy |
| mSxy mSyy |
It is relatively straight-forward to determine its eigendecomposition, and can be done analytically. I don’t have the equations on hand right now, so I’ll leave it as an exercise to the reader. :)
How can I calculate distance from camera to a point on a ground plane from an image?
I have the intrinsic parameters of the camera and the position (height, pitch).
Is there any OpenCV function that can estimate that distance?
You can use undistortPoints to compute the rays backprojecting the pixels, but that API is rather hard to use for your purpose. It may be easier to do the calculation "by hand" in your code. Doing it at least once will also help you understand what exactly that API is doing.
Express your "position (height, pitch)" of the camera as a rotation matrix R and a translation vector t, representing the coordinate transform from the origin of the ground plane to the camera. That is, given a point in ground plane coordinates Pg = [Xg, Yg, Zg], its coordinates in camera frame are given by
Pc = R * Pg + t
The camera center is Cc = [0, 0, 0] in camera coordinates. In ground coordinates it is then:
Cg = inv(R) * (-t) = -R' * t
where inv(R) is the inverse of R, R' is its transpose, and the last equality is due to R being an orthogonal matrix.
Let's assume, for simplicity, that the the ground plane is Zg = 0.
Let K be the matrix of intrinsic parameters. Given a pixel q = [u, v], write it in homogeneous image coordinates Q = [u, v, 1]. Its location in camera coordinates is
Qc = Ki * Q
where Ki = inv(K) is the inverse of the intrinsic parameters matrix. The same point in world coordinates is then
Qg = R' * Qc + Cg
All the points Pg = [Xg, Yg, Zg] that belong to the ray from the camera center through that pixel, expressed in ground coordinates, are then on the line
Pg = Cg + lambda * (Qg - Cg)
for lambda going from 0 to positive infinity. This last formula represents three equations in ground XYZ coordinates, and you want to find the values of X, Y, Z and lambda where the ray intersects the ground plane. But that means Zg=0, so you have only 3 unknowns. Solve them (you recover lambda from the 3rd equation, then substitute in the first two), and you get Xg and Yg of the solution to your problem.
I have an OpenGL program (written in Delphi) that lets user draw a polygon. I want to automatically revolve (lathe) it around an axis (say, Y asix) and get a 3D shape.
How can I do this?
For simplicity, you could force at least one point to lie on the axis of rotation. You can do this easily by adding/subtracting the same value to all the x values, and the same value to all the y values, of the points in the polygon. It will retain the original shape.
The rest isn't really that hard. Pick an angle that is fairly small, say one or two degrees, and work out the coordinates of the polygon vertices as it spins around the axis. Then just join up the points with triangle fans and triangle strips.
To rotate a point around an axis is just basic Pythagoras. At 0 degrees rotation you have the points at their 2-d coordinates with a value of 0 in the third dimension.
Lets assume the points are in X and Y and we are rotating around Y. The original 'X' coordinate represents the hypotenuse. At 1 degree of rotation, we have:
sin(1) = z/hypotenuse
cos(1) = x/hypotenuse
(assuming degree-based trig functions)
To rotate a point (x, y) by angle T around the Y axis to produce a 3d point (x', y', z'):
y' = y
x' = x * cos(T)
z' = x * sin(T)
So for each point on the edge of your polygon you produce a circle of 360 points centered on the axis of rotation.
Now make a 3d shape like so:
create a GL 'triangle fan' by using your center point and the first array of rotated points
for each successive array, create a triangle strip using the points in the array and the points in the previous array
finish by creating another triangle fan centered on the center point and using the points in the last array
One thing to note is that usually, the kinds of trig functions I've used measure angles in radians, and OpenGL uses degrees. To convert degrees to radians, the formula is:
degrees = radians / pi * 180
Essentially the strategy is to sweep the profile given by the user around the given axis and generate a series of triangle strips connecting adjacent slices.
Assume that the user has drawn the polygon in the XZ plane. Further, assume that the user intends to sweep around the Z axis (i.e. the line X = 0) to generate the solid of revolution, and that one edge of the polygon lies on that axis (you can generalize later once you have this simplified case working).
For simple enough geometry, you can treat the perimeter of the polygon as a function x = f(z), that is, assume there is a unique X value for every Z value. When we go to 3D, this function becomes r = f(z), that is, the radius is unique over the length of the object.
Now, suppose we want to approximate the solid with M "slices" each spanning 2 * Pi / M radians. We'll use N "stacks" (samples in the Z dimension) as well. For each such slice, we can build a triangle strip connecting the points on one slice (i) with the points on slice (i+1). Here's some pseudo-ish code describing the process:
double dTheta = 2.0 * pi / M;
double dZ = (zMax - zMin) / N;
// Iterate over "slices"
for (int i = 0; i < M; ++i) {
double theta = i * dTheta;
double theta_next = (i+1) * dTheta;
// Iterate over "stacks":
for (int j = 0; j <= N; ++j) {
double z = zMin + i * dZ;
// Get cross-sectional radius at this Z location from your 2D model (was the
// X coordinate in the 2D polygon):
double r = f(z); // See above definition
// Convert 2D to 3D by sweeping by angle represented by this slice:
double x = r * cos(theta);
double y = r * sin(theta);
// Get coordinates of next slice over so we can join them with a triangle strip:
double xNext = r * cos(theta_next);
double yNext = r * sin(theta_next);
// Add these two points to your triangle strip (heavy pseudocode):
strip.AddPoint(x, y, z);
strip.AddPoint(xNext, yNext, z);
}
}
That's the basic idea. As sje697 said, you'll possibly need to add end caps to keep the geometry closed (i.e. a solid object, rather than a shell). But this should give you enough to get you going. This can easily be generalized to toroidal shapes as well (though you won't have a one-to-one r = f(z) function in that case).
If you just want it to rotate, then:
glRotatef(angle,0,1,0);
will rotate it around the Y-axis. If you want a lathe, then this is far more complex.