Conical gradient in Qt (without QConicalGradient) - c++

I have to draw a conical gradient in Qt C++ but I can not use the QConicalGradient. I did have a linear gradient, but I do not know how to make a conical gradient. I do not want the finished code, but I ask for a simple algorithm.
for(int y = 0; y < image.height(); y++){
QRgb *line = (QRgb *)image.scanLine(y);
for(int x = 0; x < image.width(); x++){
QPoint currentPoint(x, y);
QPoint relativeToCenter = currentPoint - centerPoint;
float angle = atan2(relativeToCenter.y(), relativeToCenter.x);
// I have a problem in this line because I don't know how to set a color:
float hue = map(-M_PI, angle, M_PI, 0, 255);
line[x] = (red << 16) + (grn << 8) + blue;
}
}
Can you help me?

Here is some pseudo code:
Given some area to paint on, and a defined center for your gradient...
For each point that you are painting on in the area, calculate the angle to the center of your gradient.
// QPoint currentPoint; // created/populated with a x, y value by two for loops
QPoint relativeToCenter = currentPoint - centerPoint;
angle = atan2(relativeToCenter.y(), relativeToCenter.x());
Then map that angle to a color using your linear gradient, or some sort of mapping function.
float hue = map(-PI, angle, PI, 0, 255); // convert angle in radians to value
// between 0 and 255
Paint that pixel, and repeat for every pixel in your area.
EDIT: Depending on the pattern of the gradient, you will want to create a different QColor pixel. For example if you had a "rainbow" gradient, just going from one hue to the next, you could use a linear mapping function like this:
float map(float x1, float x, float x2, float y1, float y2)
{
if(true){
if(x<x1)
x = x1;
if(x>x2)
x = x2;
}
return y1 + (y2-y1)/(x2-x1)*(x-x1);
}
Then you create a QColor object using the outputted value:
float hue = map(-PI, angle, PI, 0, 255); // convert angle in radians to value
// between 0 and 255
QColor c;
c.setHsl( (int) hue, 255, 255);
Then use this QColor object with your QPainter or QBrush or QPen that you are using. Or if you are putting a qRgb value back in:
line[x] = c.rgb();
http://qt-project.org/doc/qt-4.8/qcolor.html
Hope that helps.

Related

I have a device reporting left handed coordinate angle and magnitude, how do I represent that as a line on the screen from the center?

The device I am using generates vectors like this;
How do I translate polar (angle and magnitude) from a left handed cordinate to a cartesian line, drawn on a screen where the origin point is the middle of a screen?
I am displaying the line on a wt32-sc01 screen using c++. There is a tft.drawline function but its references are normal pixel locations. In which case 0,0 is the upper left corner of the screen.
This is what I have so far (abbreviated)
....
int screen_height = tft.height();
int screen_width = tft.width();
// Device can read to 12m and reports in mm
float zoom_factor = (screen_width / 2.0) / 12000.0;
int originY = (int)(screen_height / 2);
int originX = (int)(screen_width / 2);
// Offset is for screen scrolling. No screen offset to start
int offsetX = 0;
int offsetY = 0;
...
// ld06 holds the reported angles and distances.
Coord coord = polarToCartesian(ld06.angles[i], ld06.distances[i]);
drawVector(coord, WHITE);
Coord polarToCartesian(float theta, float r) {
// cos() and sin() take radians
float rad = theta * 0.017453292519;
Coord converted = {
(int)(r * cos(rad)),
(int)(r * sin(rad))
};
return converted;
}
void drawVector(Coord coord, int color) {
// Cartesian relative the center of the screen factoring zoom and pan
int destX = (int)(zoom_factor * coord.x) + originX + offsetX;
int destY = originY - (int)(zoom_factor * coord.y) + offsetY;
// From the middle of the screen (origin X, origin Y) to destination x,y
tft.drawLine( originX, originY, destX, destY, color);
}
I have something drawing on the screen, but now I have to translate between a left handed coordinate system and the whole plane is rotated 90 degrees. How do I do that?
If I understood correctly, your coordinate system is with x pointing to the right and the y to the bottom and you used the formula for the standard math coordinate system where y is pointing up so multiplying your sin by -1 should do the trick (if it doesn't, try multiplying random things by -1, it often works for this kind of problems).
I assuming (from your image) your coordinate system has x going right y going up angle going from y axis clockwise and (0,0) is also center of your polar coordinates and your goniometrics accept radians then:
#include <math.h>
float x,y,ang,r;
const float deg = M_PI/180.0;
// ang = <0,360> // your angle
// r >= 0 // your radius (magnitude)
x = r*sin(ang*deg);
y = r*cos(ang*deg);

SDL2 function to draw a filled circle

I am looking for a function that draws a filled circle using SDL2 without using a renderer at all. I currently have this:
void Circle(int center_x, int center_y, int radius, SDL_Color color) {
eraseOldCircle();
uint32_t *pixels = (uint32_t *) windowSurface->pixels;
SDL_PixelFormat *windowFormat = windowSurface->format;
SDL_LockSurface(windowSurface); // Lock surface for direct pixel access capability
int radiussqrd = radius * radius;
for(int x=center_x-radius; x<=center_x+radius; x++) {
int dx = center_x - x;
for(int y=center_y-radius; y<=center_y+radius; y++) {
int dy = center_y - y;
if((dy * dy + dx * dx) <= radiussqrd) {
pixels[(y * WIDTH + x)] = SDL_MapRGB(windowFormat, color.r, color.g, color.b);
}
}
}
SDL_UnlockSurface(windowSurface);
SDL_UpdateWindowSurface(window);
}
which has been adapted from another function I found here, it draws the pixels directly to the windowSurface after calling eraseOldCircle (which puts the game's background image back to the previous position of the circle, effectively erasing it from there.) but it is still too slow for what I need (probably the maths?). What would be the fastest way to draw a circle using direct pixel access? I need it to be high speed so I can use it in a 2D game. I haven't been able to find anything until now, everything I see uses SDL_Renderer, but I should strictly never use it.
Here is eraseOldCircle() in case it helps:
void eraseOldCircle() {
//Calculate previous position of ball
SDL_Rect pos = {circlePosition.x-(radius+steps), circlePosition.y-(radius+steps), radius*radius, radius*2+steps};
SDL_BlitSurface(backgroundImage, &pos, windowSurface, &pos);
}
I'm not too sure how to do it with surfaces and memory management and all that, but if this helps, here is a version using an SDL_Renderer that runs pretty quickly:
void draw_circle(SDL_Renderer *renderer, int x, int y, int radius, SDL_Color color)
{
SDL_SetRenderDrawColor(renderer, color.r, color.g, color.b, color.a);
for (int w = 0; w < radius * 2; w++)
{
for (int h = 0; h < radius * 2; h++)
{
int dx = radius - w; // horizontal offset
int dy = radius - h; // vertical offset
if ((dx*dx + dy*dy) <= (radius * radius))
{
SDL_RenderDrawPoint(renderer, x + dx, y + dy);
}
}
}
}
If you draw many circles, I would guess SDL_UpdateWindowSurface is where you spend the most time. Try this instead
SDL_LockSurface
// erase and draw all circles (possibly >1000)
SDL_UnlockSurface
SDL_UpdateWindowSurface
You can optimize your circle drawing code a bit, but it is probably fast enough. I also think that SDL_Renderer is probably fast enough.
The documentation for SDL_UpdateWindowSurface says it will copy the surface to the screen. You only need to do this once per frame.

How to Rotate QPixmap around center without cutting off parts of the image

Let's say I have a QPixmap that has the dimensions of (20 x 100). How can I create a copy of this QPixmap that's rotated a specific amount and also has new dimensions to allocate the new dimensions of the rotated pixmap?
I've found multiple examples on how to rotate using QPainter and QTransform, but none seem to provide a proper manner to keep the QPixmap from cutting off.
The best example I've found so far is:
// original = Original QPixmap
QSize size = original.size();
QPixmap newPixmap(size);
newPixmap.fill(QColor::fromRgb(0, 0, 0, 0));
QPainter p(&newPixmap);
p.translate(size.height() / 2, size.height() / 2);
p.rotate(35); // Any rotation, for this example 35 degrees
p.translate(size.height() / -2, size.height() / -2);
p.drawPixmap(0, 0, original);
p.end();
This rotates a QPixmap, and places it on a new QPixmap of the same dimensions. However, I am at a loss on how to modify this to work with new dimensions.
I've even tried simply modifying the initial size of the new pixmap, but that just causes the image to be off center (and still cut off for some reason?)
Any support would be appreciated!
One way to do this would be to calculate the minimum bounding rect for your rotated image and to create a new pixmap with these dimensions onto which you can render your rotated image which is now guarenteed to fit. To do this you could take each corner point of your image rectangle and rotate them around the center. The resulting points can then be used to calculate your minimum bounding rectangle by looking at each point and finding both the minimum and maximum x and y values.
For example in the following hypothetical example we have a 100x100 rectangle. If we use a simple algorithm to rotate each corner point of the rectangle around the center by our angle (in this case 45 degrees) we get the four new corner points (50, -20), (-20, 50), (120, 120) and (50, 120). From these points we can see the minimum x value is -20, the minimum y value is -20, the maximum x value is 120 and the maximum y value is 120, so the minimum bounding rect can be described by topLeft:(-20, -20) and bottomRight:(120, 120).
To help you with this here is a function taken from another stackoverflow post for rotating a point around another point:
QPointF getRotatedPoint( QPointF p, QPointF center, qreal angleRads )
{
qreal x = p.x();
qreal y = p.y();
float s = qSin( angleRads );
float c = qCos( angleRads );
// translate point back to origin:
x -= center.x();
y -= center.y();
// rotate point
float xnew = x * c - y * s;
float ynew = x * s + y * c;
// translate point back:
x = xnew + center.x();
y = ynew + center.y();
return QPointF( x, y );
}
And here is a function I wrote that uses it to calculate the minimum bounding rect for some rectangle rotated by some angle...
QRectF getMinimumBoundingRect( QRect r, qreal angleRads )
{
QPointF topLeft = getRotatedPoint( r.topLeft(), r.center(), angleRads );
QPointF bottomRight = getRotatedPoint( r.bottomRight(), r.center(), angleRads );
QPointF topRight = getRotatedPoint( r.topRight(), r.center(), angleRads );
QPointF bottomLeft = getRotatedPoint( r.bottomLeft(), r.center(), angleRads );
// getMin and getMax just return the min / max of their arguments
qreal minX = getMin( topLeft.x(), bottomRight.x(), topRight.x(), bottomLeft.x() );
qreal minY = getMin( topLeft.y(), bottomRight.y(), topRight.y(), bottomLeft.y() );
qreal maxX = getMax( topLeft.x(), bottomRight.x(), topRight.x(), bottomLeft.x() );
qreal maxY = getMax( topLeft.y(), bottomRight.y(), topRight.y(), bottomLeft.y() );
return QRectF( QPointF( minX, minY ), QPointF( maxX, maxY ) );
}
So now we have the minimum bounding rectangle for our rotated image we can create a new pixmap with its width and height and render our rotated image to it at the center. This is tricky because of the transformation involved which makes it a bit more confusing as to what your source and target rects might be. It's actually not as hard as it might seem. You perform your translation / rotation to rotate the paint device around the center, then you can simply render your source image onto your destination image exactly as you would if you were rendering the source to the center of the destination.
For example:
QPixmap originalPixmap; // Load this from somewhere
QRectF minimumBoundingRect = getMinimumBoundingRect( originalPixmap.rect(), angleRads);
QPixmap rotatedPixmap( minimumBoundingRect.width(), minimumBoundingRect.height() );
QPainter p( &rotatedPixmap );
p.save();
// Rotate the rotated pixmap paint device around the center...
p.translate( 0.5 * rotatedPixmap.width(), 0.5 * rotatedPixmap.height() );
p.rotate( angleDegrees );
p.translate( -0.5 * rotatedPixmap.width(), -0.5 * rotatedPixmap.height() );
// The render rectangle is simply the originalPixmap rectangle as it would be if placed at the center of the rotatedPixmap rectangle...
QRectF renderRect( 0.5 * rotatedRect.width() - 0.5 * originalPixmap.width(),
0.5 * rotatedRect.height() - 0.5 * originalPixmap.height(),
originalPixmap.width(),
originalPixmap.height() );
p.drawPixmap( renderRect, originalPixmap, originalPixmap.rect() );
p.restore();
And voila, a nicely rotated image with no corners chopped off.

Fill circle with gradient

I want fill circle with gradient color, like I show on bottom. I can't figure out easy way, how to do that.
I can make more circles, but transitions are visible.
cv::circle(img, center, circle_radius * 1.5, cv::Scalar(1.0, 1.0, 0.3), CV_FILLED);
cv::circle(img, center, circle_radius * 1.2, cv::Scalar(1.0, 1.0, 0.6), CV_FILLED);
cv::circle(img, center, circle_radius, cv::Scalar(1.0, 1.0, 1.0), CV_FILLED);
All you need to do is create a function which takes in a central point and a new point, calculates the distance, and returns a grayscale value for that point. Alternatively you could just return the distance, store the distance at that point, and then scale the whole thing later with cv::normalize().
So let's say you have the central point as (50, 50) in a (100, 100) image. Here's pseudocode for what you'd want to do:
function euclideanDistance(center, point) # returns a float
return sqrt( (center.x - point.x)^2 + (center.y - point.y)^2 )
center = (50, 50)
rows = 100
cols = 100
gradient = new Mat(rows, cols) # should be of type float
for row < rows:
for col < cols:
point = (col, row)
gradient[row, col] = euclideanDistance(center, point)
normalize(gradient, 0, 255, NORM_MINMAX, uint8)
gradient = 255 - gradient
Note the steps here:
Create the Euclidean distance function to calculate distance
Create a floating point matrix to hold the distance values
Loop through all rows and columns and assign a distance value
Normalize to the range you want (you could stick with a float here instead of casting to uint8, but you do you)
Flip the binary gradient, since distances farther away will be brighter---but you want the opposite.
Now for your exact example image, there's a gradient in a circle, whereas this method just creates the whole image as a gradient. In your case, if you want a specific radius, just modify the function which calculates the Euclidean distance, and if it's beyond some distance, set it to 0 (the value at the center of the circle, which will be flipped eventually to white):
function euclideanDistance(center, point, radius) # returns a float
distance = sqrt( (center.x - point.x)^2 + (center.y - point.y)^2 )
if distance > radius:
return 0
else
return distance
Here is the above in actual C++ code:
#include <opencv2/opencv.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <cmath>
float euclidean_distance(cv::Point center, cv::Point point, int radius){
float distance = std::sqrt(
std::pow(center.x - point.x, 2) + std::pow(center.y - point.y, 2));
if (distance > radius) return 0;
return distance;
}
int main(){
int h = 400;
int w = 400;
int radius = 100;
cv::Mat gradient = cv::Mat::zeros(h, w, CV_32F);
cv::Point center(150, 200);
cv::Point point;
for(int row=0; row<h; ++row){
for(int col=0; col<w; ++col){
point.x = col;
point.y = row;
gradient.at<float>(row, col) = euclidean_distance(center, point, radius);
}
}
cv::normalize(gradient, gradient, 0, 255, cv::NORM_MINMAX, CV_8U);
cv::bitwise_not(gradient, gradient);
cv::imshow("gradient", gradient);
cv::waitKey();
}
A completely different method (though doing the same thing) would be to use the distanceTransform(). This function maps the distance from the center of a white blob to the nearest black value to a grayscale value, like we were doing above. This code is more concise and does the same thing. However, it can work on arbitrary shapes, not just circles, so that's cool.
#include <opencv2/opencv.hpp>
#include <opencv2/highgui/highgui.hpp>
int main(){
int h = 400;
int w = 400;
int radius = 100;
cv::Point center(150, 200);
cv::Mat gradient = cv::Mat::zeros(h, w, CV_8U);
cv::rectangle(gradient, cv::Point(115, 100), cv::Point(270, 350), cv::Scalar(255), -1, 8 );
cv::Mat gradient_padding;
cv::bitwise_not(gradient, gradient_padding);
cv::distanceTransform(gradient, gradient, CV_DIST_L2, CV_DIST_MASK_PRECISE);
cv::normalize(gradient, gradient, 0, 255, cv::NORM_MINMAX, CV_8U);
cv::bitwise_or(gradient, gradient_padding, gradient);
cv::imshow("gradient-distxform.png", gradient);
cv::waitKey();
}
You have to draw many circles. Color of each circle depends on distance from center. Here is some simple example:
void printGradient(cv::Mat &_input,const cv::Point &_center, const double radius)
{
cv::circle(_input, _center, radius, cv::Scalar(0, 0, 0), -1);
for(double i=1; i<radius; i=i++)
{
const int color = 255-int(i/radius * 255); //or some another color calculation
cv::circle(_input,_center,i,cv::Scalar(color, color, color),2);
}
}
And result:
Another approach not mentioned yet is to precompute a circle gradient image (with one of the mentioned approaches like the accepted solution) and use affine warping with linear interpolation to create other such circles (different sizes). This can be faster, if warping and interpolation are optimized and maybe accelerated by hardware.
Result might be a bit worse than perfect.
I once used this to create a single individual vignetting mask circle for each frame innendoscopic imaging. Was faster than to compute the distances "manually".

C++ / OpenGL - 2D - How to clip a circle in a rectangle boundary box

I was just wondering how would I go about clipping a circle in a rectangular boundary box? I am currently using the Cohen–Sutherland algorithm for line clipping in my program and so far I've managed to get rectangles and polygons to clip. However, for circle clipping, I have no idea how I would accomplish this. I'm using the following to construct my circle:
glBegin(GL_POLYGON);
double radius = 50;
for(int angle = 0; angle <= 360; angle++ ){
float const curve = 2 * PI * (float)angle / (float)360;
glVertex2f(point.x + sin(curve) * radius, point.y + cos(curve) * radius);
}
glEnd();
My clipping algorithm is the same as the one here: http://en.wikipedia.org/wiki/Cohen%E2%80%93Sutherland_algorithm. However, it returns 2 points representing a new line to later be used to draw the clipped shape. So basically I've tried to do this:
line Lines[360] // an array with size 360 with data type line, which is a struct holding two points (x1, y1, x2, y2) of the new line returned by my clipping function.
double radius = 50;
for(int angle = 0; angle < 360; angle++){
float const currentCurve = 2 * PI * (float)angle / (float)360;
float const nextCurve = 2 * PI * (float)(angle+1) / (float)360;
int x1 = (int)(point[i].x + sin(currentCurve) * radius); // point is another struct holding only a single point.
y1 = (int)(point[i].y + cos(currentCurve) * radius);
x2 = (int)(point[i+1].x+ sin(nextCurve) * radius);
y2 = (int)(point[i+1].y + cos(nextCurve) * radius);=
// Clip the points with the clipping algorithm:
Lines[i] = Clipper(x1, y1, x2, y2);
}
// Once all lines have been clipped or not, draw:
glBegin(GL_POLYGON);
for(int i = 0; i < 360; i++){
glVertex2f(Lines[i].x1, Lines[i].y1);
glVertex2f(Lines[i].x2, Lines[i].y2);
}
glEnd();
Note that, I've drawn a circle on the screen with a mouse and and stored each 360 points into a struct array called point, which is apart of a linked list. So I have like 1 node representing one circle on the screen.
Anyway, with the above, my circle is not drawing clipped (or drawing at all for that matter) and my application crashes after a few mouse clicks.
Use the scissor test - read up on glScissor(): http://www.opengl.org/sdk/docs/man/xhtml/glScissor.xml