I am trying to detect blur rate of the face images with below code.
cv::Mat greyMat;
cv::Mat laplacianImage;
cv::Mat imageClone = LapMat.clone();
cv::resize(imageClone, imageClone, cv::Size(150, 150), 0, 0, cv::INTER_CUBIC);
cv::cvtColor(imageClone, greyMat, CV_BGR2GRAY);
Laplacian(greyMat, laplacianImage, CV_64F);
cv::Scalar mean, stddev; // 0:1st channel, 1:2nd channel and 2:3rd channel
meanStdDev(laplacianImage, mean, stddev, cv::Mat());
double variance = stddev.val[0] * stddev.val[0];
cv::Mat M = (cv::Mat_(3, 1) << -1, 2, -1);
cv::Mat G = cv::getGaussianKernel(3, -1, CV_64F);
cv::Mat Lx;
cv::sepFilter2D(LapMat, Lx, CV_64F, M, G);
cv::Mat Ly;
cv::sepFilter2D(LapMat, Ly, CV_64F, G, M);
cv::Mat FM = cv::abs(Lx) + cv::abs(Ly);
double focusMeasure = cv::mean(FM).val[0];
return focusMeasure;
it some times gives not good results as attached picture.
Is there a best practice way to detect blurry faces ?
I attached an example image which is high scored with above code which is false.
Best
I'm not sure how are you interpreting your results. To measure blur, you usually take the output of the Blur Detector (a number) and compare it against a threshold value, then determine if the input is, in fact, blurry or not. I don't see such a comparison in your code.
There are several ways to measure "blurriness", or rather, sharpness. Let's take a look at one. It involves computing the variance of the Laplacian and then comparing it to an expected value. This is the code:
//read the image and convert it to grayscale:
cv::Mat inputImage = cv::imread( "dog.png" );
cv::Mat gray;
cv::cvtColor( inputImage, gray, cv::COLOR_RGB2GRAY );
//Cool, let's compute the laplacian of the gray image:
cv::Mat laplacianImage;
cv::Laplacian( gray, laplacianImage, CV_64F );
//Prepare to compute the mean and standard deviation of the laplacian:
cv::Scalar mean, stddev;
cv::meanStdDev( laplacianImage, mean, stddev, cv::Mat() );
//Let’s compute the variance:
double variance = stddev.val[0] * stddev.val[0];
Up until this point, we've effectively calculated the variance of the Laplacian, but we still need to compare against a threshold:
double blurThreshold = 300;
if ( variance <= blurThreshold ) {
std::cout<<"Input image is blurry!"<<std::endl;
} else {
std::cout<<"Input image is sharp"<<std::endl;
}
Let’s check out the results. These are my test images. I've printed the variance value in the lower-left corner of the images. The threshold value is 300, blue text is within limits, red text is below.
Related
Problem : Watershed algorithm
I started app project, for image processing, using OpenCv 4.5.3 and Swift ( with C++ ). I'm fighting with watershaded alg. for a really long time... And i have no clue what did i do wrong. Just don't know...
Error :
libc++abi.dylib: terminating with uncaught exception of type cv::Exception: OpenCV(4.5.3)
/Volumes/build-storage/build/master_iOS-mac/opencv/modules/imgproc/src/segmentation.cpp:161:
error: (-215:Assertion failed) src.type()
== CV_8UC3 && dst.type() == CV_32SC1 in function 'watershed'
terminating with uncaught exception of type cv::Exception: OpenCV(4.5.3)
/Volumes/build-storage/build/master_iOS-mac/opencv/modules/imgproc/src/segmentation.cpp:161: error:
(-215:Assertion failed) src.type()
== CV_8UC3 && dst.type() == CV_32SC1 in function 'watershed'
In the definition of openCv's watershed we can find :
#param image Input 8-bit 3-channel image.
#param markers Input/output 32-bit single-channel image (map) of markers. It should have the same size as image .
Code
+(UIImage *) watershed:(UIImage *)src{
cv::Mat img, mask;
UIImageToMat(src, img);
// Change the background from white to black, since that will help later to extract
// better results during the use of Distance Transform
cv::inRange(img, cv::Scalar(255,255,255), cv::Scalar(255,255,255), mask);
img.setTo(cv::Scalar(0,0,0), mask);
// Create a kernel that we will use to sharpen our image
// an approximation of second derivative, a quite strong kernel
cv::Mat kernel = (cv::Mat_<float>(3,3) <<
1, 1, 1,
1, -8, 1,
1, 1, 1);
// do the laplacian filtering as it is
// well, we need to convert everything in something more deeper then CV_8U
// because the kernel has some negative values,
// and we can expect in general to have a Laplacian image with negative values
// BUT a 8bits unsigned int (the one we are working with) can contain values from 0 to 255
// so the possible negative number will be truncated
cv::Mat lapl;
cv::filter2D(img, lapl, CV_32F, kernel);
cv::Mat sharp;
img.convertTo(sharp, CV_32F);
cv::Mat result = sharp - lapl;
// convert back to 8bits gray scale
result.convertTo(result, CV_8UC3);
lapl.convertTo(lapl, CV_8UC3);
cv::Mat bw;
cv::cvtColor(result, bw, cv::COLOR_BGR2GRAY);
cv::threshold(bw, bw, 40, 255, cv::THRESH_BINARY | cv::THRESH_OTSU);
// Perform the distance transform algorithm
cv::Mat dist;
cv::distanceTransform(bw, dist, cv::DIST_L2, cv::DIST_MASK_3);
// Normalize the distance image for range = {0.0, 1.0}
// so we can visualize and threshold it
cv::normalize(dist, dist, 0, 1.0, cv::NORM_MINMAX);
// Threshold to obtain the peaks
// This will be the markers for the foreground objects
cv::threshold(dist, dist, 0.4, 1.0, cv::THRESH_BINARY);
// Dilate a bit the dist image
cv::Mat kernel1 = cv::Mat::ones(3, 3, CV_8U);
dilate(dist, dist, kernel1);
// Create the CV_8U version of the distance image
// It is needed for findContours()
cv::Mat dist_8u;
dist.convertTo(dist_8u, CV_8U);
// Find total markers
std::vector<std::vector<cv::Point> > contours;
findContours(dist_8u, contours, cv::RETR_EXTERNAL, cv::CHAIN_APPROX_SIMPLE);
// Create the marker image for the watershed algorithm
cv::Mat markers = cv::Mat::zeros(dist.size(), CV_32S);
// Draw the foreground markers
for (size_t i = 0; i < contours.size(); i++)
{
drawContours(markers, contours, static_cast<int>(i), cv::Scalar(static_cast<int>(i)+1), -1);
}
// Draw the background marker
circle(markers, cv::Point(5,5), 3, cv::Scalar(255), -1);
cv::Mat markers8u;
markers.convertTo(markers8u, CV_8U, 10);
// Perform the watershed algorithm
watershed(result, markers);
return MatToUIImage(result);
}
You can clearly see, that variables has proper type, as in descr. of function:
result.convertTo(result, CV_8UC3);
cv::Mat markers = cv::Mat::zeros(dist.size(), CV_32S);
The convertTo can not add channels as well can not reduce/convert image to image with smaller amount of channels.
The key in this case is to use :
cvtColor(src, src, COLOR_BGRA2BGR); // change 4 to 3 channels
I have a weird problem where my average gradient magnitude result is different if I use a mask as opposed to creating a new Mat of that small ROI. I'll explain the 2 different ways I do this and 2 different average gradient magnitude results I get. I thought I should get the same average gradient magnitude result?
Scenario: Image A is my source/original image of a landscape. I want to get the average gradient magnitude in the region A (10,100), (100,100), (100,150), (10,150).
Technique 1:
- Create a ROI Mat that just shows region A. So its dimensions are 90 by 50.
- Perform cv::Sobel(), cv::magnitude() then cv::meanStdDev()
- My average gradient magnitude result is 11.34.
Technique 2:
- Create a new Mat that is a mask. The mat is the same dimensions as Image A and has a white area where Region A is. Then create a new Mat that just shows that region of Image A and the rest of the Mat is black - hopefully this makes sense.
- Perform cv::Sobel(), cv::magnitude() (but use the mask) then cv::meanStdDev()
- My average gradient magnitude result is 43.76.
Why the different result?
Below is my code:
static Mat backupSrc;
static Mat curSrc;
// Technique 1
void inspectRegion(const Point& strt, const Point& end) {
curSrc = Mat(backupSrc.size(), CV_8UC3);
cvtColor(backupSrc, curSrc, CV_GRAY2RGB);
Rect region = Rect(strt, end);
Mat regionImg = Mat(curSrc, region);
// Calculate the average gradient magnitude/strength across the image
Mat dX, dY, mag;
Sobel(regionImg, dX, CV_32F, 1, 0);
Sobel(regionImg, dY, CV_32F, 0, 1);
magnitude(dX, dY, mag);
Scalar sMMean, sMStdDev;
meanStdDev(mag, sMMean, sMStdDev);
double magnitudeMean = sMMean[0];
double magnitudeStdDev = sMStdDev[0];
rectangle(curSrc, region, { 0 }, 1);
printf("[Gradient Magnitude Mean: %.3f, Gradient Magnitude Std Dev: %.3f]\n", magnitudeMean, magnitudeStdDev);
}
// Technique 2
void inspectRegion(const std::vector<Point>& pnts) {
curSrc = Mat(backupSrc.size(), CV_8UC3);
cvtColor(backupSrc, curSrc, CV_GRAY2RGB);
std::vector<std::vector<Point>> cPnts;
cPnts.push_back(pnts);
Mat mask = Mat::zeros(curSrc.rows, curSrc.cols, CV_8UC1);
fillPoly(mask, cPnts, { 255 });
Mat regionImg;
curSrc.copyTo(regionImg, mask);
// Calculate the average gradient magnitude/strength across the image
Mat dX, dY, mag;
Sobel(regionImg, dX, CV_32F, 1, 0);
Sobel(regionImg, dY, CV_32F, 0, 1);
magnitude(dX, dY, mag);
Scalar sMMean, sMStdDev;
meanStdDev(mag, sMMean, sMStdDev, mask);
double magnitudeMean = sMMean[0];
double magnitudeStdDev = sMStdDev[0];
polylines(curSrc, pnts, true, { 255 }, 3);
printf("[Gradient Magnitude Mean: %.3f, Gradient Magnitude Std Dev: %.3f]\n", magnitudeMean, magnitudeStdDev);
}
In technique 2 the gradients around the boarders of your rectangle will be very high and will corrupt the calculation.
Consider dilating your mask before computing the gradients so that this spike is outside of the non-dilated mask that you send into the meanStdDev function.
I am writing my thesis and one part of the task is to interpolate between images to create intermediate images. The work has to be done in c++ using openCV 2.4.13.
The best solution I've found so far is computing optical flow and remapping. But this solution has two problems that I am unable to solve on my own:
There are pixels that should go out of view (bottom of image for example), but they do not.
Some pixels do not move, creating a distorted result (upper right part of the couch)
What has made the flow&remap approach better:
Equalizing the intensity. This i'm allowed to do. You can check the result by comparing the couch form (centre of remapped image and original).
Reducing size of image. This i'm NOT allowed to do, as I need the same size output. Is there a way to rescale the optical flow result to get the bigger remapped image?
Other approaches tried and failed:
cuda::interpolateFrames. Creates incredible ghosting.
blending images with cv::addWeighted. Even worse ghosting.
Below is the code I am using at the moment. And images: dropbox link with input and result images
int main(){
cv::Mat second, second_gray, cutout, cutout_gray, flow_n;
second = cv::imread( "/home/zuze/Desktop/forstack/second_L.jpg", 1 );
cutout = cv::imread("/home/zuze/Desktop/forstack/cutout_L.png", 1);
cvtColor(second, second_gray, CV_BGR2GRAY);
cvtColor(cutout, cutout_gray, CV_RGB2GRAY );
///----------COMPUTE OPTICAL FLOW AND REMAP -----------///
cv::calcOpticalFlowFarneback( second_gray, cutout_gray, flow_n, 0.5, 3, 15, 3, 5, 1.2, 0 );
cv::Mat remap_n; //looks like it's drunk.
createNewFrame(remap_n, flow_n, 1, second, cutout );
cv::Mat cflow_n;
cflow_n = cutout_gray;
cvtColor(cflow_n, cflow_n, CV_GRAY2BGR);
drawOptFlowMap(flow_n, cflow_n, 10, CV_RGB(0,255,0));
///--------EQUALIZE INTENSITY, COMPUTE OPTICAL FLOW AND REMAP ----///
cv::Mat cutout_eq, second_eq;
cutout_eq= equalizeIntensity(cutout);
second_eq= equalizeIntensity(second);
cv::Mat flow_eq, cutout_eq_gray, second_eq_gray, cflow_eq;
cvtColor( cutout_eq, cutout_eq_gray, CV_RGB2GRAY );
cvtColor( second_eq, second_eq_gray, CV_RGB2GRAY );
cv::calcOpticalFlowFarneback( second_eq_gray, cutout_eq_gray, flow_eq, 0.5, 3, 15, 3, 5, 1.2, 0 );
cv::Mat remap_eq;
createNewFrame(remap_eq, flow_eq, 1, second, cutout_eq );
cflow_eq = cutout_eq_gray;
cvtColor(cflow_eq, cflow_eq, CV_GRAY2BGR);
drawOptFlowMap(flow_eq, cflow_eq, 10, CV_RGB(0,255,0));
cv::imshow("remap_n", remap_n);
cv::imshow("remap_eq", remap_eq);
cv::imshow("cflow_eq", cflow_eq);
cv::imshow("cflow_n", cflow_n);
cv::imshow("sec_eq", second_eq);
cv::imshow("cutout_eq", cutout_eq);
cv::imshow("cutout", cutout);
cv::imshow("second", second);
cv::waitKey();
return 0;
}
Function for remapping, to be used for intermediate image creation:
void createNewFrame(cv::Mat & frame, const cv::Mat & flow, float shift, cv::Mat & prev, cv::Mat &next){
cv::Mat mapX(flow.size(), CV_32FC1);
cv::Mat mapY(flow.size(), CV_32FC1);
cv::Mat newFrame;
for (int y = 0; y < mapX.rows; y++){
for (int x = 0; x < mapX.cols; x++){
cv::Point2f f = flow.at<cv::Point2f>(y, x);
mapX.at<float>(y, x) = x + f.x*shift;
mapY.at<float>(y, x) = y + f.y*shift;
}
}
remap(next, newFrame, mapX, mapY, cv::INTER_LANCZOS4);
frame = newFrame;
cv::waitKey();
}
Function to display optical flow in vector form:
void drawOptFlowMap (const cv::Mat& flow, cv::Mat& cflowmap, int step, const cv::Scalar& color) {
cv::Point2f sum; //zz
std::vector<float> all_angles;
int count=0; //zz
float angle, sum_angle=0; //zz
for(int y = 0; y < cflowmap.rows; y += step)
for(int x = 0; x < cflowmap.cols; x += step)
{
const cv::Point2f& fxy = flow.at< cv::Point2f>(y, x);
if((fxy.x != fxy.x)||(fxy.y != fxy.y)){ //zz, for SimpleFlow
//std::cout<<"meh"; //do nothing
}
else{
line(cflowmap, cv::Point(x,y), cv::Point(cvRound(x+fxy.x), cvRound(y+fxy.y)),color);
circle(cflowmap, cv::Point(cvRound(x+fxy.x), cvRound(y+fxy.y)), 1, color, -1);
sum +=fxy;//zz
angle = atan2(fxy.y,fxy.x);
sum_angle +=angle;
all_angles.push_back(angle*180/M_PI);
count++; //zz
}
}
}
Function to equalize intensity of images, for better results:
cv::Mat equalizeIntensity(const cv::Mat& inputImage){
if(inputImage.channels() >= 3){
cv::Mat ycrcb;
cvtColor(inputImage,ycrcb,CV_BGR2YCrCb);
std::vector<cv::Mat> channels;
cv::split(ycrcb,channels);
cv::equalizeHist(channels[0], channels[0]);
cv::Mat result;
cv::merge(channels,ycrcb);
cvtColor(ycrcb,result,CV_YCrCb2BGR);
return result;
}
return cv::Mat();
}
So to recap, my questions:
Is it possible to resize Farneback optical flow to apply to 2xbigger image?
How to deal with pixels that go out of view like at the bottom of my images (the brown wooden part should disappear).
How to deal with distortion that is created because optical flow wasn't computed for those pixels, while many pixels around there have motion? (couch upper right, & lion figurine has a ghost hand in the remapped image).
With OpenCV's Farneback optical flow, you will only get a rough estimation of pixel displacement, hence the distortions that appear on the result images.
I don't think optical flow is the way to go for what you are trying to achieve IMHO. Instead I'd recommend you to have a look at Image / Pixel Registration for instace here : http://docs.opencv.org/trunk/db/d61/group__reg.html
Image / Pixel Registration is the science of matching pixels of two images. Active research is ongoing about this complex non-trivial subject that is not yet accurately resolved.
How can I apply a notch filter on an image spectrum using OpenCV 2.4 and C++? I want to calculate the DFT of an image, suppress certain frequencies and calculate inverse dft. Can anyone show me some sample code how to apply a notch filter in frequecy domain?
EDIT:
Here is what I tried, but the quadrants of the frequency spectrum are not in order so the origin of the spectrum is not the center of the image. That makes is difficult for me to identify the frequencies to suppress. When swapping quadrants so that the origin is the center, inverse DFT shows wrong results. Can anyone show me how to do inverse dft with swapped quadrants?
I don't understand the number of columns in the frequency images filter1 and filter2 (see code). If I use filter1.cols as u in the for loop, I don't access the right border of the images. Filter1 and filter2 seem to have approx. 5000 columns but the source image has a resolution of 1280x1024 (grayscale). Any thoughts on that?
Any further comments about my code?
Mat img;
img=imread(filename,CV_LOAD_IMAGE_GRAYSCALE);
int M = getOptimalDFTSize( img.rows );
int N = getOptimalDFTSize( img.cols );
Mat padded;
copyMakeBorder(img, padded, 0, M - img.rows, 0, N - img.cols, BORDER_CONSTANT, Scalar::all(0));
Mat planes[] = {Mat_<float>(padded), Mat::zeros(padded.size(), CV_32F)};
Mat complexImg;
merge(planes, 2, complexImg);
dft(complexImg, complexImg,cv::DFT_SCALE|cv::DFT_COMPLEX_OUTPUT);
split(complexImg, planes);
Mat filter1;
planes[0].copyTo(filter1);
Mat filter2;
planes[1].copyTo(filter2);
for( int i = 0; i < filter1.rows; ++i)
{
for(int u=7;u<15;++u)
{
filter1.at<uchar>(i,u)=0;
filter2.at<uchar>(i,u)=0;
}
Mat inverse[] = {filter1,filter2};
Mat filterspec;
merge(inverse, 2, filterspec);
cv::Mat inverseTransform;
cv::dft(filterspec, inverseTransform,cv::DFT_INVERSE|cv::DFT_REAL_OUTPUT);
cv::Mat finalImage;
inverseTransform.convertTo(finalImage, CV_8U);
I have created dft of an image and after some adjustment with filters i want to convert it back to the real image but every time when i do that it gives me wrong result ..seems like its not converting it back.
ForierTransform and createGaussianHighPassFilter are my own functions rest of the code i am using like below for the inversion back to real image.
Mat fft = ForierTransform(HeightPadded,WidthPadded);
Mat ghpf = createGaussianHighPassFilter(Size(WidthPadded, HeightPadded), db);
Mat res;
cv::multiply(fft,ghpf,res);
imshow("fftXhighpass1", res);
idft(res,res,DFT_INVERSE,res.rows);
cv::Mat croped = res(cv::Rect(0, 0, img.cols,img.rows));
//res.convertTo(res,CV_32S);
imshow("fftXhighpass", res);
even if i dont apply the filter i am unable to reverse back dft result ...
here is my dft code is , i could not find any sample to reverse dft back to normal image..
Mat ForierTransform(int M,int N)
{
Mat img = imread("thumb1-small-test.jpg", CV_LOAD_IMAGE_GRAYSCALE);
Mat padded;
copyMakeBorder(img, padded, 0, M - img.rows, 0, N - img.cols, BORDER_CONSTANT, Scalar::all(0));
Mat planes[] = {Mat_<float>(padded), Mat::zeros(padded.size(), CV_32F)};
Mat complexImg;
merge(planes, 2, complexImg);
dft(complexImg, complexImg);
split(complexImg, planes);
magnitude(planes[0], planes[1], planes[0]);
Mat mag = planes[0];
mag += Scalar::all(1);
log(mag, mag);
// crop the spectrum, if it has an odd number of rows or columns
mag = mag(Rect(0, 0, mag.cols & -2, mag.rows & -2));
normalize(mag, mag, 0, 1, CV_MINMAX);
return mag;
}
kindly help
[EDIT: After I found the solution with the help of mevatron] (below is the correct code)
Mat ForierTransform(int M,int N)
{
Mat img = imread("thumb1-small-test.jpg", CV_LOAD_IMAGE_GRAYSCALE);
Mat padded;
copyMakeBorder(img, padded, 0, M - img.rows, 0, N - img.cols, BORDER_CONSTANT, Scalar::all(0));
Mat planes[] = {Mat_<float>(padded), Mat::zeros(padded.size(), CV_32F)};
Mat complexImg;
merge(planes, 2, complexImg);
dft(complexImg, complexImg);
return complexImg;
}
Mat img = imread("thumb1-small-test.jpg",CV_LOAD_IMAGE_GRAYSCALE);
int WidthPadded=0,HeightPadded=0;
WidthPadded=img.cols*2;
HeightPadded=img.rows*2;
int M = getOptimalDFTSize( img.rows );
//Create a Gaussian Highpass filter 5% the height of the Fourier transform
double db = 0.05 * HeightPadded;
Mat fft = ForierTransform(HeightPadded,WidthPadded);
Mat ghpf = createGaussianHighPassFilter(Size(WidthPadded, HeightPadded), db);
Mat res;
cv::mulSpectrums(fft,ghpf,res,DFT_COMPLEX_OUTPUT);
idft(res,res,DFT_COMPLEX_OUTPUT,img.rows);
Mat padded;
copyMakeBorder(img, padded, 0, img.rows, 0, img.cols, BORDER_CONSTANT, Scalar::all(0));
Mat planes[] = {Mat_<float>(padded), Mat::zeros(padded.size(), CV_32F)};
split(res, planes);
magnitude(planes[0], planes[1], planes[0]);
Mat mag = planes[0];
mag += Scalar::all(1);
log(mag, mag);
// crop the spectrum, if it has an odd number of rows or columns
mag = mag(Rect(0, 0, mag.cols & -2, mag.rows & -2));
int cx = mag.cols/2;
int cy = mag.rows/2;
normalize(mag, mag, 1, 0, CV_MINMAX);
cv::Mat croped = mag(cv::Rect(cx, cy, img.cols,img.rows));
cv::threshold(croped , croped , 0.56, 1, cv::THRESH_BINARY);
imshow("fftPLUShpf", mag);
imshow("cropedBinary", croped);
It now can able to display ridges valley of finger , and can be more optimize with respect to threshold as well
I see a few problems going on here.
First, you need to use the mulSpectrums function to convolve two FFTs, and not multiply.
Second, the createGaussianHighPassFilter is only outputting a single channel non-complex filter. You'll probably need to just set the complex channel to Mat::zeros like you did for your input image.
Third, don't convert the output of the FFT to log-magnitude spectrum. It will not combine correctly with the filter, and you won't get the same thing when performing the inverse. So, just return complexImg right after the DFT is executed. Log-magnitude spectrum is useful for a human to look at the data, but not for what you are trying to do.
Finally, make sure you pay attention to the difference to between the full-complex output of dft and the Complex Conjugate Symmetric (CCS) packed output. Intel has a good page on how this data is formatted here. In your case, for simplicity I would keep everything in full-complex mode to make your life easier.
Hope that helps!