Related
I'm using OpenCV2.4.8.2 on Mac OS 10.9.5.
I have the following snippet of code:
static void compute_weights(const vector<Mat>& images, vector<Mat>& weights)
{
weights.clear();
for (int i = 0; i < images.size(); i++) {
Mat image = images[i];
Mat mask = Mat::zeros(image.size(), CV_32F);
int x_start = (i == 0) ? 0 : image.cols/2;
int y_start = 0;
int width = image.cols/2;
int height = image.rows;
Mat roi = mask(Rect(x_start,y_start,width,height)); // Set Roi
roi.setTo(1);
weights.push_back(mask);
}
}
static void blend(const vector<Mat>& inputImages, Mat& outputImage)
{
int maxPyrIndex = 6;
vector<Mat> weights;
compute_weights(inputImages, weights);
// Find the fused pyramid:
vector<Mat> fused_pyramid;
for (int i = 0; i < inputImages.size(); i++) {
Mat image = inputImages[i];
// Build Gaussian Pyramid for Weights
vector<Mat> weight_gaussian_pyramid;
buildPyramid(weights[i], weight_gaussian_pyramid, maxPyrIndex);
// Build Laplacian Pyramid for original image
Mat float_image;
inputImages[i].convertTo(float_image, CV_32FC3, 1.0/255.0);
vector<Mat> orig_guassian_pyramid;
vector<Mat> orig_laplacian_pyramid;
buildPyramid(float_image, orig_guassian_pyramid, maxPyrIndex);
for (int j = 0; j < orig_guassian_pyramid.size() - 1; j++) {
Mat sized_up;
pyrUp(orig_guassian_pyramid[j+1], sized_up, Size(orig_guassian_pyramid[j].cols, orig_guassian_pyramid[j].rows));
orig_laplacian_pyramid.push_back(orig_guassian_pyramid[j] - sized_up);
}
// Last Lapalcian layer is the same as the Gaussian layer
orig_laplacian_pyramid.push_back(orig_guassian_pyramid[orig_guassian_pyramid.size()-1]);
// Convolve laplacian original with guassian weights
vector<Mat> convolved;
for (int j = 0; j < maxPyrIndex + 1; j++) {
// Create 3 channels for weight gaussian pyramid as well
vector<Mat> gaussian_3d_vec;
for (int k = 0; k < 3; k++) {
gaussian_3d_vec.push_back(weight_gaussian_pyramid[j]);
}
Mat gaussian_3d;
merge(gaussian_3d_vec, gaussian_3d);
//Mat convolved_result = weight_gaussian_pyramid[j].clone();
Mat convolved_result = gaussian_3d.clone();
multiply(gaussian_3d, orig_laplacian_pyramid[j], convolved_result);
convolved.push_back(convolved_result);
}
if (i == 0) {
fused_pyramid = convolved;
} else {
for (int j = 0; j < maxPyrIndex + 1; j++) {
fused_pyramid[j] += convolved[j];
}
}
}
// Blending
for (int i = (int)fused_pyramid.size()-1; i > 0; i--) {
Mat sized_up;
pyrUp(fused_pyramid[i], sized_up, Size(fused_pyramid[i-1].cols, fused_pyramid[i-1].rows));
fused_pyramid[i-1] += sized_up;
}
Mat final_color_bgr;
fused_pyramid[0].convertTo(final_color_bgr, CV_32F, 255);
final_color_bgr.copyTo(outputImage);
imshow("final", outputImage);
waitKey(0);
imwrite(outputImagePath, outputImage);
}
This code is doing some basic pyramid blending for 2 images. The key issues are related to imshow and imwrite in the last line. They gave me drastically different results. I apologize for displaying such a long/messy code, but I am afraid this difference is coming from some other parts of the code that can subsequently affect the imshow and imwrite.
The first image shows the result from imwrite and the second image shows the result from imshow, based on the code given. I'm quite confused about why this is the case.
I also noticed that when I do these:
Mat float_image;
inputImages[i].convertTo(float_image, CV_32FC3, 1.0/255.0);
imshow("float image", float_image);
imshow("orig image", image);
They show exactly the same thing, that is they both show the same picture in the original rgb image (in image).
IMWRITE functionality
By default, imwrite, converts the input image into Only 8-bit (or 16-bit unsigned (CV_16U) in case of PNG, JPEG 2000, and TIFF) single-channel or 3-channel (with ‘BGR’ channel order) images can be saved using this function.
So whatever format you feed in for imwrite, it blindly converts into CV_8U with a range 0(black) - 255(white) in BGR format.
IMSHOW - problem
So when noticed your function, fused_pyramid[0].convertTo(final_color_bgr, CV_32F, 255); fused_pyramid is already under mat type 21 (floating point CV_32F). You tried to convert into floating point with a scale factor 255. This scaling factor 255 caused the problem # imshow. Instead to visualize, you can directly feed in fused_pyramid without conversion as already it is scaled to floating point between 0.0(black) - 1.0(white).
Hope it helps.
I am a beginner in openCV.
I want to plot the intensity profile for R, G and B for the image given below.
I am like to plot R, G and B values w.r.t to pixel location in three different graphs.
So far I have learnt how to read an Image and display. for example using imread();
Mat img = imread("Apple.bmp");
and then showing it on the screen using imshow(" Window", img);.
Now I would like to put all R , G and B values in 3 separate buffers; buf1, buf2, buf3 and plot these values.
Kindly provide me some hint or a sample code snippet to help me understand this.
You can separate R, G and B into separate Mats using cv::split()
std::vector<Mat> planes(3);
cv::split(img, planes);
cv::Mat R = planes[2];
cv::Mat G = planes[1];
cv::Mat B = planes[0];
But you only need to separate them like this if you have code that is expecting a Mat with a single color channnel.
Don't use at<>() as the supposed duplicate suggest - it is really slow if you are sequentially scanning an image (but it is good for random access).
You can scan the image efficiently like this
for(int i = 0; i < img.rows; ++i)
{
// get pointers to each row
cv::Vec3b* row = img.ptr<cv::Vec3b>(i);
// now scan the row
for(int j = 0; j < img.cols; ++j)
{
cv::Vec3b pixel = row[j];
uchar r = pixel[2];
uchar g = pixel[1];
uchar b = pixel[0];
process(r, g, b);
}
}
Lastly if you do want to make a histogram, you can use this code. It is fairly old so I suppose it still works.
void show_histogram_image(cv::Mat src, cv::Mat &hist_image)
{ // based on http://docs.opencv.org/2.4.4/modules/imgproc/doc/histograms.html?highlight=histogram#calchist
int sbins = 256;
int histSize[] = {sbins};
float sranges[] = { 0, 256 };
const float* ranges[] = { sranges };
cv::MatND hist;
int channels[] = {0};
cv::calcHist( &src, 1, channels, cv::Mat(), // do not use mask
hist, 1, histSize, ranges,
true, // the histogram is uniform
false );
double maxVal=0;
minMaxLoc(hist, 0, &maxVal, 0, 0);
int xscale = 10;
int yscale = 10;
//hist_image.create(
hist_image = cv::Mat::zeros(256, sbins*xscale, CV_8UC3);
for( int s = 0; s < sbins; s++ )
{
float binVal = hist.at<float>(s, 0);
int intensity = cvRound(binVal*255/maxVal);
rectangle( hist_image, cv::Point(s*xscale, 0),
cv::Point( (s+1)*xscale - 1, intensity),
cv::Scalar::all(255),
CV_FILLED );
}
}
I am trying to calculate the skew of text in an image so I can correct it for the best OCR results.
Currently this is the function I am using:
double compute_skew(Mat &img)
{
// Binarize
cv::threshold(img, img, 225, 255, cv::THRESH_BINARY);
// Invert colors
cv::bitwise_not(img, img);
cv::Mat element = cv::getStructuringElement(cv::MORPH_RECT, cv::Size(5, 3));
cv::erode(img, img, element);
std::vector<cv::Point> points;
cv::Mat_<uchar>::iterator it = img.begin<uchar>();
cv::Mat_<uchar>::iterator end = img.end<uchar>();
for (; it != end; ++it)
if (*it)
points.push_back(it.pos());
cv::RotatedRect box = cv::minAreaRect(cv::Mat(points));
double angle = box.angle;
if (angle < -45.)
angle += 90.;
cv::Point2f vertices[4];
box.points(vertices);
for(int i = 0; i < 4; ++i)
cv::line(img, vertices[i], vertices[(i + 1) % 4], cv::Scalar(255, 0, 0), 1, CV_AA);
return angle;
}
When I look at then angle in debug I get 0.000000
However when I give it this image I get proper results of a skew of about 16 degrees:
How can I properly detect the skew in the first image?
there are a few other ways to get the skew degree, 1) by hough transform 2) by horizontal projection profile. rotate the image in different angle bins and calculate horizontal projection. the angle with the greatest horizontal histogram value is the deskewed angle.
i have provided below implementation of 1). i believe this to be superior to the boxing method you are using because it requires that you completely clean the image of any noise,which just isnt possible in most of the time.
you should know that the method doesnt work well if there's too much noise. you can reduce noise in different ways depending on what type of "line" you want to treat as the most dominant in the image. i have provided two methods for this. be sure to play with parameters and threshold etc.
results (all run using preprocess2, all run using same parameter set)
code
#include <opencv2/opencv.hpp>
using namespace cv;
using namespace std;
void hough_transform(Mat& im,Mat& orig,double* skew)
{
double max_r=sqrt(pow(.5*im.cols,2)+pow(.5*im.rows,2));
int angleBins = 180;
Mat acc = Mat::zeros(Size(2*max_r,angleBins),CV_32SC1);
int cenx = im.cols/2;
int ceny = im.rows/2;
for(int x=1;x<im.cols-1;x++)
{
for(int y=1;y<im.rows-1;y++)
{
if(im.at<uchar>(y,x)==255)
{
for(int t=0;t<angleBins;t++)
{
double r =(x-cenx)*cos((double)t/angleBins*CV_PI)+(y-ceny)*sin((double)t /angleBins*CV_PI);
r+=max_r;
acc.at<int>(t,int(r))++;
}
}
}
}
Mat thresh;
normalize(acc,acc,255,0,NORM_MINMAX);
convertScaleAbs(acc,acc);
/*debug
Mat cmap;
applyColorMap(acc,cmap,COLORMAP_JET);
imshow("cmap",cmap);
imshow("acc",acc);*/
Point maxLoc;
minMaxLoc(acc,0,0,0,&maxLoc);
double theta = (double)maxLoc.y/angleBins*CV_PI;
double rho = maxLoc.x-max_r;
if(abs(sin(theta))<0.000001)//check vertical
{
//when vertical, line equation becomes
//x = rho
double m = -cos(theta)/sin(theta);
Point2d p1 = Point2d(rho+im.cols/2,0);
Point2d p2 = Point2d(rho+im.cols/2,im.rows);
line(orig,p1,p2,Scalar(0,0,255),1);
*skew=90;
cout<<"skew angle "<<" 90"<<endl;
}else
{
//convert normal form back to slope intercept form
//y = mx + b
double m = -cos(theta)/sin(theta);
double b = rho/sin(theta)+im.rows/2.-m*im.cols/2.;
Point2d p1 = Point2d(0,b);
Point2d p2 = Point2d(im.cols,im.cols*m+b);
line(orig,p1,p2,Scalar(0,0,255),1);
double skewangle;
skewangle= p1.x-p2.x>0? (atan2(p1.y-p2.y,p1.x-p2.x)*180./CV_PI):(atan2(p2.y-p1.y,p2. x-p1.x)*180./CV_PI);
*skew=skewangle;
cout<<"skew angle "<<skewangle<<endl;
}
imshow("orig",orig);
}
Mat preprocess1(Mat& im)
{
Mat ret = Mat::zeros(im.size(),CV_32SC1);
for(int x=1;x<im.cols-1;x++)
{
for(int y=1;y<im.rows-1;y++)
{
int gy = (im.at<uchar>(y-1,x+1)-im.at<uchar>(y-1,x-1))
+2*(im.at<uchar>(y,x+1)-im.at<uchar>(y,x-1))
+(im.at<uchar>(y+1,x+1)-im.at<uchar>(y+1,x-1));
int gx = (im.at<uchar>(y+1,x-1) -im.at<uchar>(y-1,x-1))
+2*(im.at<uchar>(y+1,x)-im.at<uchar>(y-1,x))
+(im.at<uchar>(y+1,x+1)-im.at<uchar>(y-1,x+1));
int g2 = (gy*gy + gx*gx);
ret.at<int>(y,x)=g2;
}
}
normalize(ret,ret,255,0,NORM_MINMAX);
ret.convertTo(ret,CV_8UC1);
threshold(ret,ret,50,255,THRESH_BINARY);
return ret;
}
Mat preprocess2(Mat& im)
{
// 1) assume white on black and does local thresholding
// 2) only allow voting top is white and buttom is black(buttom text line)
Mat thresh;
//thresh=255-im;
thresh=im.clone();
adaptiveThreshold(thresh,thresh,255,CV_ADAPTIVE_THRESH_GAUSSIAN_C,THRESH_BINARY,15,-2);
Mat ret = Mat::zeros(im.size(),CV_8UC1);
for(int x=1;x<thresh.cols-1;x++)
{
for(int y=1;y<thresh.rows-1;y++)
{
bool toprowblack = thresh.at<uchar>(y-1,x)==0 || thresh.at<uchar>(y-1,x-1)==0 || thresh.at<uchar>(y-1,x+1)==0;
bool belowrowblack = thresh.at<uchar>(y+1,x)==0 || thresh.at<uchar>(y+1, x-1)==0 || thresh.at<uchar>(y+1,x+1)==0;
uchar pix=thresh.at<uchar>(y,x);
if((!toprowblack && pix==255 && belowrowblack))
{
ret.at<uchar>(y,x) = 255;
}
}
}
return ret;
}
Mat rot(Mat& im,double thetaRad)
{
cv::Mat rotated;
double rskew = thetaRad* CV_PI/180;
double nw = abs(sin(thetaRad))*im.rows+abs(cos(thetaRad))*im.cols;
double nh = abs(cos(thetaRad))*im.rows+abs(sin(thetaRad))*im.cols;
cv::Mat rot_mat = cv::getRotationMatrix2D(Point2d(nw*.5,nh*.5), thetaRad*180/CV_PI, 1);
Mat pos = Mat::zeros(Size(1,3),CV_64FC1);
pos.at<double>(0)=(nw-im.cols)*.5;
pos.at<double>(1)=(nh-im.rows)*.5;
Mat res = rot_mat*pos;
rot_mat.at<double>(0,2) += res.at<double>(0);
rot_mat.at<double>(1,2) += res.at<double>(1);
cv::warpAffine(im, rotated, rot_mat,Size(nw,nh), cv::INTER_LANCZOS4);
return rotated;
}
int main(int argc, char** argv)
{
string src="C:/data/skew.png";
Mat im= imread(src);
Mat gray;
cvtColor(im,gray,CV_BGR2GRAY);
Mat preprocessed = preprocess2(gray);
imshow("preprocessed2",preprocessed);
double skew;
hough_transform(preprocessed,im,&skew);
Mat rotated = rot(im,skew* CV_PI/180);
imshow("corrected",rotated);
waitKey(0);
return 0;
}
the approach you posted has its own "ideal binarization" assumption. the threshold value directly affects the process. utilize otsu threshold, or think about DFT for a generic solution.
otsu trial:
int main()
{
Mat input = imread("your text");
cvtColor(input, input, CV_BGR2GRAY);
Mat img;
cv::threshold(input, img, 100, 255, cv::THRESH_OTSU);
cv::bitwise_not(img, img);
imshow("img ", img);
waitKey(0);
vector<Point> points;
findNonZero(img, points);
cv::RotatedRect box = cv::minAreaRect(points);
double angle = box.angle;
if (angle < -45.)
angle += 90.;
cv::Point2f vertices[4];
box.points(vertices);
for(int i = 0; i < 4; ++i)
cv::line(img, vertices[i], vertices[(i + 1) % 4], cv::Scalar(255, 0, 0));
imshow("img ", img);
waitKey(0);
return 0;
}
I am trying to convert this second answer code into c++ , What I did is not giving me appropriate result , here is my code :
{
Mat img = imread("messi5.jpg");
int level_n = 2;
Mat p = Mat::zeros(img.cols*img.rows, 3, CV_32F);
vector<Mat> bgr;
cv::split(img, bgr);
//Divide each pixel color with 127 for level 2
for(int i=0; i<img.cols*img.rows; i++) {
p.at<float>(i,0) = bgr[0].data[i] / 127.0;
p.at<float>(i,1) = bgr[1].data[i] / 127.0;
p.at<float>(i,2) = bgr[2].data[i] / 127.0;
}
vector<Mat> Img2 = p[bgr];
Mat out;
cv::transform(img,out,p);
imshow ("output" , out);
}
What I didn't understand is how I put these colour's which I divided by 127 into Matrix , where I am going wrong?
Other way i am trying is
vector<Mat> bgr;
Mat blue , green , red;
cv::split(img, bgr);
blue = bgr[0]/127.0;
if (blue > 128)
{
blue = 255;
}
else
{
blue = 0;
}
same for red and green
Why don't just do it in place (for level 2):
Mat img = imread("messi5.jpg");
for(int i=0;i<img.rows;i++)
for(int j=0;j<img.cols;j++) {
cv::Vec3b p = img.at<cv::Vec3b>(i,j);
for(int k = 0;k < img.channels();k++)
p[k] = p[k] > 127 ? 255 : 0;
img.at<cv::Vec3b>(i,j) = p;
}
// do whatever you want with processed image img
I am trying to use opencv EM algorithm to do color extraction.I am using the following code based on example in opencv documentation:
cv::Mat capturedFrame ( height, width, CV_8UC3 );
int i, j;
int nsamples = 1000;
cv::Mat samples ( nsamples, 2, CV_32FC1 );
cv::Mat labels;
cv::Mat img = cv::Mat::zeros ( height, height, CV_8UC3 );
img = capturedFrame;
cv::Mat sample ( 1, 2, CV_32FC1 );
CvEM em_model;
CvEMParams params;
samples = samples.reshape ( 2, 0 );
for ( i = 0; i < N; i++ )
{
//from the training samples
cv::Mat samples_part = samples.rowRange ( i*nsamples/N, (i+1)*nsamples/N);
cv::Scalar mean (((i%N)+1)*img.rows/(N1+1),((i/N1)+1)*img.rows/(N1+1));
cv::Scalar sigma (30,30);
cv::randn(samples_part,mean,sigma);
}
samples = samples.reshape ( 1, 0 );
//initialize model parameters
params.covs = NULL;
params.means = NULL;
params.weights = NULL;
params.probs = NULL;
params.nclusters = N;
params.cov_mat_type = CvEM::COV_MAT_SPHERICAL;
params.start_step = CvEM::START_AUTO_STEP;
params.term_crit.max_iter = 300;
params.term_crit.epsilon = 0.1;
params.term_crit.type = CV_TERMCRIT_ITER|CV_TERMCRIT_EPS;
//cluster the data
em_model.train ( samples, Mat(), params, &labels );
cv::Mat probs;
probs = em_model.getProbs();
cv::Mat weights;
weights = em_model.getWeights();
cv::Mat modelIndex = cv::Mat::zeros ( img.rows, img.cols, CV_8UC3 );
for ( i = 0; i < img.rows; i ++ )
{
for ( j = 0; j < img.cols; j ++ )
{
sample.at<float>(0) = (float)j;
sample.at<float>(1) = (float)i;
int response = cvRound ( em_model.predict ( sample ) );
modelIndex.data [ modelIndex.cols*i + j] = response;
}
}
My question here is:
Firstly, I want to extract each model, here totally five, then store those corresponding pixel values in five different matrix. In this case, I could have five different colors seperately. Here I only obtained their indexes, is there any way to achieve their corresponding colors here? To make it easy, I can start from finding the dominant color based on these five GMMs.
Secondly, here my sample datapoints are "100", and it takes about nearly 3 seconds for them. But I want to do all these things in no more than 30 milliseconds. I know OpenCV background extraction, which is using GMM, performs really fast, below 20ms, that means, there must be a way for me to do all these within 30 ms for all 600x800=480000 pixels. I found predict function is the most time consuming one.
First Question:
In order to do color extraction you first need to train the EM with your input pixels. After that you simply loop over all the input pixels again and use predict() to classify each of them. I've attached a small example that utilizes EM for foreground/background separation based on colors. It shows you how to extract the dominant color (mean) of each gaussian and how to access the original pixel color.
#include <opencv2/opencv.hpp>
int main(int argc, char** argv) {
cv::Mat source = cv::imread("test.jpg");
//ouput images
cv::Mat meanImg(source.rows, source.cols, CV_32FC3);
cv::Mat fgImg(source.rows, source.cols, CV_8UC3);
cv::Mat bgImg(source.rows, source.cols, CV_8UC3);
//convert the input image to float
cv::Mat floatSource;
source.convertTo(floatSource, CV_32F);
//now convert the float image to column vector
cv::Mat samples(source.rows * source.cols, 3, CV_32FC1);
int idx = 0;
for (int y = 0; y < source.rows; y++) {
cv::Vec3f* row = floatSource.ptr<cv::Vec3f > (y);
for (int x = 0; x < source.cols; x++) {
samples.at<cv::Vec3f > (idx++, 0) = row[x];
}
}
//we need just 2 clusters
cv::EMParams params(2);
cv::ExpectationMaximization em(samples, cv::Mat(), params);
//the two dominating colors
cv::Mat means = em.getMeans();
//the weights of the two dominant colors
cv::Mat weights = em.getWeights();
//we define the foreground as the dominant color with the largest weight
const int fgId = weights.at<float>(0) > weights.at<float>(1) ? 0 : 1;
//now classify each of the source pixels
idx = 0;
for (int y = 0; y < source.rows; y++) {
for (int x = 0; x < source.cols; x++) {
//classify
const int result = cvRound(em.predict(samples.row(idx++), NULL));
//get the according mean (dominant color)
const double* ps = means.ptr<double>(result, 0);
//set the according mean value to the mean image
float* pd = meanImg.ptr<float>(y, x);
//float images need to be in [0..1] range
pd[0] = ps[0] / 255.0;
pd[1] = ps[1] / 255.0;
pd[2] = ps[2] / 255.0;
//set either foreground or background
if (result == fgId) {
fgImg.at<cv::Point3_<uchar> >(y, x, 0) = source.at<cv::Point3_<uchar> >(y, x, 0);
} else {
bgImg.at<cv::Point3_<uchar> >(y, x, 0) = source.at<cv::Point3_<uchar> >(y, x, 0);
}
}
}
cv::imshow("Means", meanImg);
cv::imshow("Foreground", fgImg);
cv::imshow("Background", bgImg);
cv::waitKey(0);
return 0;
}
I've tested the code with the following image and it performs quite good.
Second Question:
I've noticed that the maximum number of clusters has a huge impact on the performance. So it's better to set this to a very conservative value instead of leaving it empty or setting it to the number of samples like in your example. Furthermore the documentation mentions an iterative procedure to repeatedly optimize the model with less-constrained parameters. Maybe this gives you some speed-up. To read more please have a look at the docs inside the sample code that is provided for train() here.