I want to calculate perimeter of a white blob in a 512*512 dimension binary image. Image will have only one blob. I used following code earlier in OpenCV 3 but somehow it doesn't work in OpenCV 4.2. IplImage
is deprecated in latest version. And I cannot pass Mat object directly to cvFindContours function. I am new to opencv and I don't know how does it work. Other related questions regarding perimeter are still unanswered.
To summaries, following works in opencv 3 but does not work in current opencv version (4.2).
int getPerimeter(unsigned char* inImagePtr, int inW, int inH)
{
int sumEven = 0; int sumOdd = 0;
int sumCorner = 0; int prevCode = 0;
//create a mat input Image
cv::Mat inImage(inH, inW, CV_8UC1, inImagePtr);
//create four connected structuring element
cv::Mat element = cv::Mat::zeros(3, 3, CV_8UC1);
element.data[1] = 1; element.data[3] = 1;
element.data[4] = 1; element.data[5] = 1;
element.data[7] = 1;
//erode input image
cv::Mat erodeImage;
erode(inImage, erodeImage, element);
//Invert eroded Image
cv::threshold(erodeImage, erodeImage, 0, 255, THRESH_BINARY_INV);
//multiply with original binary Image to get the edge Image
cv::Mat edge = erodeImage.mul(inImage);
//Get chain code of the blob
CvChain* chain = 0;
CvMemStorage* storage = 0;
storage = cvCreateMemStorage(0);
auto temp = new IplImage(edge);
cvFindContours(temp, storage, (CvSeq**)(&chain), sizeof(*chain), CV_RETR_EXTERNAL, CV_CHAIN_CODE);
delete temp;
for (; chain != NULL; chain = (CvChain*)chain->h_next)
{
CvSeqReader reader;
int i, total = chain->total;
cvStartReadSeq((CvSeq*)chain, &reader, 0);
for (i = 0; i < total; i++)
{
char code;
CV_READ_SEQ_ELEM(code, reader);
if (code % 2 == 0)
sumEven++;
else
sumOdd++;
if (i > 0) {
if (code != prevCode)
sumCorner++;
}
prevCode = code;
}
}
float perimeter = (float)sumEven*0.980 + (float)sumOdd*1.406 - (float)sumCorner*0.091;
return (roundf(perimeter));
}
This worked just fine for me!
int getPerimeter(unsigned char* inImagePtr, int inW, int inH) {
// create a mat input Image
cv::Mat inImage(inH, inW, CV_8UC1, inImagePtr);
// create four connected structuring element
cv::Mat element = cv::Mat::zeros(3, 3, CV_8UC1);
element.data[1] = 1;
element.data[3] = 1;
element.data[4] = 1;
element.data[5] = 1;
element.data[7] = 1;
// erode input image
cv::Mat erodeImage;
erode(inImage, erodeImage, element);
// Invert eroded Image
cv::threshold(erodeImage, erodeImage, 0, 255, THRESH_BINARY_INV);
// multiply with original binary Image to get the edge Image
cv::Mat edge = erodeImage.mul(inImage);
vector<vector<Point>> contours;
findContours(edge, contours, RETR_EXTERNAL, CHAIN_APPROX_SIMPLE); // Retrieve only external contour
int preValue[2];
int nextValue[2];
int sumEven = 0;
int sumOdd = 0;
//vector<Point>::iterator itr;
for (int ii = 0; ii < contours[0].size(); ii++) {
Point pt = contours[0].at(ii);
preValue[0] = pt.x;
preValue[1] = pt.y;
if (ii != contours[0].size() - 1) {
Point pt_next = contours[0].at(ii + 1);
nextValue[0] = pt_next.x;
nextValue[1] = pt_next.y;
} else {
Point pt_next = contours[0].at(0);
nextValue[0] = pt_next.x;
nextValue[1] = pt_next.y;
}
if ((preValue[0] == nextValue[0]) or (preValue[1] == nextValue[1])) {
sumEven = sumEven + abs(nextValue[0] - preValue[0]) + abs(nextValue[1] - preValue[1]);
} else {
sumOdd = sumOdd + abs(nextValue[0] - preValue[0]);
}
}
int sumCorner = contours[0].size() - 1;
float perimeter = round(sumEven * 0.980 + sumOdd * 1.406 - sumCorner * 0.091);
return (roundf(perimeter));
}
I have some code to draw a line between two points on an image which are selected by mouse, and then to display a histogram.
However, when I press q as required by code I get an error saying R6010 abort() has been called and saying VC++ run time error.
Please advise me how I can find this error.
#include <vector>
#include "opencv2/highgui/highgui.hpp"
#include <opencv\cv.h>
#include <iostream>
#include<conio.h>
using namespace cv;
using namespace std;
struct Data_point
{
int x;
unsigned short int y;
};
int PlotMeNow(unsigned short int *values, unsigned int nSamples)
{
std::vector<Data_point> graph(nSamples);
for (unsigned int i = 0; i < nSamples; i++)
{
graph[i].x = i;
graph[i].y = values[i];
}
cv::Size imageSize(5000, 500); // your window size
cv::Mat image(imageSize, CV_8UC1);
if (image.empty()) //check whether the image is valid or not
{
std::cout << "Error : Image cannot be created..!!" << std::endl;
system("pause"); //wait for a key press
return 0;
}
else
{
std::cout << "Good job : Image created successfully..!!" << std::endl;
}
// tru to do some ofesseting so the graph do not hide on x or y axis
Data_point dataOffset;
dataOffset.x = 20;
// we have to mirror the y axis!
dataOffset.y = 5000;
for (unsigned int i = 0; i<nSamples; ++i)
{
graph[i].x = (graph[i].x + dataOffset.x) * 3;
graph[i].y = (graph[i].y + dataOffset.y) / 200;
}
// draw the samples
for (unsigned int i = 0; i<nSamples - 1; ++i)
{
cv::Point2f p1;
p1.x = graph[i].x;
p1.y = graph[i].y;
cv::Point2f p2;
p2.x = graph[i + 1].x;
p2.y = graph[i + 1].y;
cv::line(image, p1, p2, 'r', 1, 4, 0);
}
cv::namedWindow("MyWindow1", CV_WINDOW_AUTOSIZE); //create a window with the name "MyWindow"
cv::imshow("MyWindow1", image); //display the image which is stored in the 'img' in the "MyWindow" window
while (true)
{
char c = cv::waitKey(10);
if (c == 'q')
break;
}
destroyWindow("MyWindow1");
destroyWindow("MyWindow"); //destroy the window with the name, "MyWindow"
return 0;
}
void IterateLine(const Mat& image, vector<ushort>& linePixels, Point p2, Point p1, int* count1)
{
LineIterator it(image, p2, p1, 8);
for (int i = 0; i < it.count; i++, it++)
{
linePixels.push_back(image.at<ushort>(it.pos())); //doubt
}
*count1 = it.count;
}
//working line with mouse
void onMouse(int evt, int x, int y, int flags, void* param)
{
if (evt == CV_EVENT_LBUTTONDOWN)
{
std::vector<cv::Point>* ptPtr = (std::vector<cv::Point>*)param;
ptPtr->push_back(cv::Point(x, y));
}
}
void drawline(Mat image, std::vector<Point>& points)
{
cv::namedWindow("Output Window");
cv::setMouseCallback("Output Window", onMouse, (void*)&points);
int X1 = 0, Y1 = 0, X2 = 0, Y2 = 0;
while (1)
{
cv::imshow("Output Window", image);
if (points.size() > 1) //we have 2 points
{
for (auto it = points.begin(); it != points.end(); ++it)
{
}
break;
}
waitKey(10);
}
//just for testing that we are getting pixel values
X1 = points[0].x;
X2 = points[1].x;
Y1 = points[0].y;
Y2 = points[1].y;
// Draw a line
line(image, Point(X1, Y1), Point(X2, Y2), 'r', 2, 8);
cv::imshow("Output Window", image);
//exit image window
while (true)
{
char c = cv::waitKey(10);
if (c == 'q')
break;
}
destroyWindow("Output Window");
}
void show_histogram_image(Mat img1)
{
int sbins = 65536;
int histSize[] = { sbins };
float sranges[] = { 0, 65536 };
const float* ranges[] = { sranges };
cv::MatND hist;
int channels[] = { 0 };
cv::calcHist(&img1, 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;
cv::Mat hist_image;
hist_image = cv::Mat::zeros(65536, sbins*xscale, CV_16UC1);
for int s = 0; s < sbins; s++)
{
float binVal = hist.at<float>(s, 0);
int intensity = cvRound(binVal * 65535 / maxVal);
rectangle(hist_image, cv::Point(s*xscale, hist_image.rows),
cv::Point((s + 1)*xscale - 1, hist_image.rows - intensity),
cv::Scalar::all(65535), 1);
}
imshow("Histogram", hist_image);
waitKey(0);
}
int main()
{
vector<Point> points1;
vector<ushort>linePixels;
Mat img = cvLoadImage("desert.jpg");
if (img.empty()) //check whether the image is valid or not
{
cout << "Error : Image cannot be read..!!" << endl;
system("pause"); //wait for a key press
return -1;
}
//Draw the line
drawline(img, points1);
//now check the collected points
Mat img1 = cvLoadImage("desert.jpg");
if (img1.empty()) //check whether the image is valid or not
{
cout << "Error : Image cannot be read..!!" << endl;
system("pause"); //wait for a key press
return -1;
}
int *t = new int;
IterateLine( img1, linePixels, points1[1], points1[0], t );
PlotMeNow(&linePixels[0], t[0]);
show_histogram_image(img);
delete t;
_getch();
return 0;
}
This is one of the bad smells in your code:
void IterateLine(const Mat& image, vector<ushort>& linePixels, Point p2, Point p1, int* count1)
{
...
linePixels.push_back(image.at<ushort>(it.pos())); //doubt
Now image is a CV_8UC3 image (from Mat img1 = cvLoadImage("desert.jpg");, but you are accessing here like it is CV_16UC1, so what gets put in linePixels is garbage. This will almost certainly cause PlotMeNow() to draw outside its image and corrupt something, which is probably why your code is crashing.
Sine it is very unclear what your code is trying to do, I can't suggest what you should have here instead.
I have just managed to do this, you only have to put "-1" to your loop limit:
for (unsigned int i = 0; i < nSamples-1; i++)
{
graph[i].x = i;
graph[i].y = values[i];
}
I want to calculate optical flow using cvcalcopticalflowBM function in opencv 2.4.7
When I complied the belowed code. The error message is "Sizes of input arguments do not macth() in cvcalcopticalflowbm
I do not understand why it is. Please help me. Thank you advance.
#define BS 5
IplImage *imgA = NULL, *imgB = NULL;
IplImage *grayA = NULL, *grayB = NULL;
IplImage *velx = NULL, *vely = NULL;
IplImage *result = NULL;
imgA = cvLoadImage("00.jpg", 1);
imgB = cvLoadImage("01.jpg", 1);
grayA = cvCreateImage(cvGetSize(imgA), IPL_DEPTH_8U, 1);
grayB = cvCreateImage(cvGetSize(imgA), IPL_DEPTH_8U, 1);
cvCvtColor(imgA, grayA, CV_BGR2GRAY);
cvCvtColor(imgB, grayB, CV_BGR2GRAY);
CvSize size = cvGetSize(imgA);
size.width /= BS;
size.height /= BS;
result = cvCreateImage(size, IPL_DEPTH_8U, 1);
for (int i=0; i<size.height; i++) {
for (int j=0; j<size.width; j++) {
cvSet(result, CV_RGB(255,255,255), NULL);
}
}
velx = cvCreateImage(size, IPL_DEPTH_32F, 1);
vely = cvCreateImage(size, IPL_DEPTH_32F, 1);
cvCalcOpticalFlowBM(grayB, grayA, cvSize(BS, BS), cvSize(1, 1), cvSize(1, 1), 0, velx, vely);
//
cvNamedWindow("HorFlowBM", CV_WINDOW_AUTOSIZE);
cvShowImage("HorFlowBM", velx);
cvNamedWindow("VerFlowBM", CV_WINDOW_AUTOSIZE);
cvShowImage("VerFlowBM", vely);
for (int i=0; i<size.height; i+=2) {
for (int j=0; j<size.width; j+=2) {
int dx = (int)cvGetReal2D(velx, i, j);
int dy = (int)cvGetReal2D(vely, i, j);
cvLine(result, cvPoint(j, i), cvPoint(j+dx, i+dy), CV_RGB(0,0,0), 1, 8, 0);
}
}
cvNamedWindow("OpticalFlow", CV_WINDOW_AUTOSIZE);
cvShowImage("OpticalFlow", result);
cvWaitKey(0);
Are you sure that the input images are getting load. Try to show them after loading them i.e. cvShowImage("input1", imgA);. Also, try to print the size of both the images to check that the size of both the images is same.
I recognized this error.
The size of velx and vely should be
CvSize velSize =
{
(grayA->width - BLOCK_SIZE + SHIFT_SIZE)/SHIFT_SIZE,
(grayA->height - BLOCK_SIZE + SHIFT_SIZE)/SHIFT_SIZE
};
It becomes correctly when complie the program
Good day. I'm new to OpenCV and right now, I'm trying to do fingertip detection using colour tracking and background subtraction methods. I got the colour tracking part working but I have no idea on how to subtract the background and leave only the fingertips.
Here is my code.
#include <opencv2/opencv.hpp>
#include <stdio.h>
#include <iostream>
using namespace std;
IplImage* GetThresholdedImage(IplImage* img, CvScalar& lowerBound, CvScalar& upperBound)
{
// Convert the image into an HSV image
IplImage* imgHSV = cvCreateImage(cvGetSize(img), 8, 3);
cvCvtColor(img, imgHSV, CV_BGR2HSV);
IplImage* imgThreshed = cvCreateImage(cvGetSize(img), 8, 1);
cvInRangeS(imgHSV, lowerBound, upperBound, imgThreshed);
cvReleaseImage(&imgHSV);
return imgThreshed;
}
int main()
{
int lineThickness = 2;
CvScalar lowerBound = cvScalar(20, 100, 100);
CvScalar upperBound = cvScalar(30, 255, 255);
int b,g,r;
lowerBound = cvScalar(0,58,89);
upperBound = cvScalar(25,173,229);
CvCapture* capture = 0;
capture = cvCaptureFromCAM(1);
if(!capture)
{
printf("Could not initialize capturing...\n");
return -1;
}
cvNamedWindow("video");
cvNamedWindow("thresh");
// This image holds the "scribble" data...
// the tracked positions of the object
IplImage* imgScribble = NULL;
while(true)
{
IplImage* frame = 0;
frame = cvQueryFrame(capture);
if(!frame)
break;
// If this is the first frame, we need to initialize it
if(imgScribble == NULL)
{
imgScribble = cvCreateImage(cvGetSize(frame), 8, 3);
}
// Holds the thresholded image (tracked color -> white, the rest -> black)
IplImage* imgThresh = GetThresholdedImage(frame,lowerBound,upperBound);
// Calculate the moments to estimate the position of the object
CvMoments *moments = (CvMoments*)malloc(sizeof(CvMoments));
cvMoments(imgThresh, moments, 1);
// The actual moment values
double moment10 = cvGetSpatialMoment(moments, 1, 0);
double moment01 = cvGetSpatialMoment(moments, 0, 1);
double area = cvGetCentralMoment(moments, 0, 0);
// Holding the last and current positions
static int posX = 0;
static int posY = 0;
int lastX = posX;
int lastY = posY;
posX = moment10/area;
posY = moment01/area;
cout << "position = " << posX << " " << posY << endl;
// We want to draw a line only if its a valid position
if(lastX>0 && lastY>0 && posX>0 && posY>0)
{
// Draw a yellow line from the previous point to the current point
cvLine(imgScribble, cvPoint(posX, posY), cvPoint(lastX, lastY), upperBound, lineThickness);
}
// Add the scribbling image and the frame...
cvAdd(frame, imgScribble, frame);
cvShowImage("thresh", imgThresh);
cvShowImage("video", frame);
int c = cvWaitKey(10);
if(c==27) //ESC key
{
break;
}
cvReleaseImage(&imgThresh);
delete moments;
}
cvReleaseCapture(&capture);
return 0;
}
I don t know if I understand you right but I think you should need to add the following:
cvErode(imgThreshed, imgThreshed, NULL, 1);
cvDilate(imgThreshed, imgThreshed, NULL, 1);
in GetThresholdedImage and get less noise ! but after all I think it would be better for you to use the cv::Mat object of opencv ;)
Try BGS library, I used it before and like it. You can get it here: http://code.google.com/p/bgslibrary/
I am trying to implement online face recognition using the webcam. I am using this two websites as references
shervinemami.co.cc
cognotics.com
I have few questions:
In face recognition, there are 6 steps:
Grab a frame from the camera
Detect a face within the image
Crop the frame to show just the face
Convert the frame to greyscale
Preprocess the image
Recognize the person in the image.
I am able to do the first five steps. Last step i am not able to do. I am not sure how to link step 5 to step 6.
I have already created the train.txt file and test.txt file which contains the information of the training and testing images. I have already added the functions such as learn(), doPCA() to the code...
But the point is how to use these functions in the main to recognize the image that is already preprocessed.
Need some help on it...
Attached the code below:
// Real-time.cpp : Defines the entry point for the console application.
#include "stdafx.h"
#include <cv.h>
#include <cxcore.h>
#include <highgui.h>
#include <cvaux.h>
IplImage ** faceImgArr = 0; // array of face images
CvMat * personNumTruthMat = 0; // array of person numbers
int nTrainFaces = 0; // the number of training images
int nEigens = 0; // the number of eigenvalues
IplImage * pAvgTrainImg = 0; // the average image
IplImage ** eigenVectArr = 0; // eigenvectors
CvMat * eigenValMat = 0; // eigenvalues
CvMat * projectedTrainFaceMat = 0; // projected training faces
IplImage* getCameraFrame(CvCapture* &camera);
IplImage* detectFaces( IplImage *img ,CvHaarClassifierCascade* facecascade,CvMemStorage* storage );
CvRect detectFaceInImage(IplImage *inputImg, CvHaarClassifierCascade* cascade);
IplImage* preprocess( IplImage* inputImg);
IplImage* resizeImage(const IplImage *origImg, int newWidth,
int newHeight, bool keepAspectRatio);
void learn();
void recognize();
void doPCA();
void storeTrainingData();
int loadTrainingData(CvMat ** pTrainPersonNumMat);
int findNearestNeighbor(float * projectedTestFace);
int loadFaceImgArray(char * filename);
int _tmain(int argc, _TCHAR* argv[])
{
CvCapture* camera = 0; // The camera device.
CvMemStorage *storage;
cvNamedWindow( "Realtime:", CV_WINDOW_AUTOSIZE);
char *faceCascadeFilename = "C:/OpenCV2.1/data/haarcascades/haarcascade_frontalface_alt.xml";
CvHaarClassifierCascade* faceCascade;
faceCascade = (CvHaarClassifierCascade*)cvLoad(faceCascadeFilename, 0, 0, 0);
storage = cvCreateMemStorage( 0 );
learn();
while ( cvWaitKey(10) != 27 ) // Quit on "Escape" key
{
IplImage *frame = getCameraFrame(camera);
//IplImage* resized=cvCreateImage(cvSize(420,240),frame->depth,3);
//cvResizeWindow( "Image:", 640, 480);
//cvResize(frame,resized);
//cvShowImage( "Realtime:", resized );
IplImage *imgA = resizeImage(frame, 420,240, true);
IplImage *frame1 = detectFaces(imgA,faceCascade,storage);
frame1 = preprocess(frame1);
}
// Free the camera.
cvReleaseCapture( &camera );
cvReleaseMemStorage( &storage );
return 0;
}
IplImage* getCameraFrame(CvCapture* &camera)
{
IplImage *frame;
int w, h;
// If the camera hasn't been initialized, then open it.
if (!camera) {
printf("Acessing the camera ...\n");
camera = cvCreateCameraCapture( 0 );
if (!camera) {
printf("Couldn't access the camera.\n");
exit(1);
}
// Try to set the camera resolution to 320 x 240.
cvSetCaptureProperty(camera, CV_CAP_PROP_FRAME_WIDTH, 320);
cvSetCaptureProperty(camera, CV_CAP_PROP_FRAME_HEIGHT, 240);
// Get the first frame, to make sure the camera is initialized.
frame = cvQueryFrame( camera );
if (frame) {
w = frame->width;
h = frame->height;
printf("Got the camera at %dx%d resolution.\n", w, h);
}
// Wait a little, so that the camera can auto-adjust its brightness.
Sleep(1000); // (in milliseconds)
}
// Wait until the next camera frame is ready, then grab it.
frame = cvQueryFrame( camera );
if (!frame) {
printf("Couldn't grab a camera frame.\n");
exit(1);
}
return frame;
}
CvRect detectFaceInImage(IplImage *inputImg, CvHaarClassifierCascade* cascade)
{
// Smallest face size.
CvSize minFeatureSize = cvSize(20, 20);
// Only search for 1 face.
int flags = CV_HAAR_FIND_BIGGEST_OBJECT | CV_HAAR_DO_ROUGH_SEARCH;
// How detailed should the search be.
float search_scale_factor = 1.1f;
IplImage *detectImg;
IplImage *greyImg = 0;
CvMemStorage* storage;
CvRect rc;
double t;
CvSeq* rects;
CvSize size;
int i, ms, nFaces;
storage = cvCreateMemStorage(0);
cvClearMemStorage( storage );
// If the image is color, use a greyscale copy of the image.
detectImg = (IplImage*)inputImg;
if (inputImg->nChannels > 1) {
size = cvSize(inputImg->width, inputImg->height);
greyImg = cvCreateImage(size, IPL_DEPTH_8U, 1 );
cvCvtColor( inputImg, greyImg, CV_BGR2GRAY );
detectImg = greyImg; // Use the greyscale image.
}
// Detect all the faces in the greyscale image.
t = (double)cvGetTickCount();
rects = cvHaarDetectObjects( detectImg, cascade, storage,
search_scale_factor, 3, flags, minFeatureSize);
t = (double)cvGetTickCount() - t;
ms = cvRound( t / ((double)cvGetTickFrequency() * 1000.0) );
nFaces = rects->total;
printf("Face Detection took %d ms and found %d objects\n", ms, nFaces);
// Get the first detected face (the biggest).
if (nFaces > 0)
rc = *(CvRect*)cvGetSeqElem( rects, 0 );
else
rc = cvRect(-1,-1,-1,-1); // Couldn't find the face.
if (greyImg)
cvReleaseImage( &greyImg );
cvReleaseMemStorage( &storage );
//cvReleaseHaarClassifierCascade( &cascade );
return rc; // Return the biggest face found, or (-1,-1,-1,-1).
}
IplImage* detectFaces( IplImage *img ,CvHaarClassifierCascade* facecascade,CvMemStorage* storage )
{
int i;
CvRect *r;
CvSeq *faces = cvHaarDetectObjects(
img,
facecascade,
storage,
1.1,
3,
0 /*CV_HAAR_DO_CANNY_PRUNNING*/,
cvSize( 40, 40 ) );
int padding_width = 30; // pixels
int padding_height = 30; // pixels
for( i = 0 ; i < ( faces ? faces->total : 0 ) ; i++ ) {
r = ( CvRect* )cvGetSeqElem( faces, i );
cvRectangle( img,
cvPoint( r->x, r->y ),
cvPoint( r->x + r->width, r->y + r->height ),
CV_RGB( 255, 0, 0 ), 1, 8, 0 );
}
cvShowImage( "Realtime:", img );
//cropping the face
cvSetImageROI(img, cvRect(r->x,r->y,r->width,r->height));
IplImage *img2 = cvCreateImage(cvGetSize(img),
img->depth,
img->nChannels);
cvCopy(img, img2, NULL);
cvResetImageROI(img);
return img;
}
IplImage* preprocess( IplImage* inputImg){
IplImage *detectImg, *greyImg = 0;
IplImage *imageProcessed;
CvSize size;
detectImg = (IplImage*)inputImg;
if (inputImg->nChannels > 1) {
size = cvSize(inputImg->width, inputImg->height);
greyImg = cvCreateImage(size, IPL_DEPTH_8U, 1 );
cvCvtColor( inputImg, greyImg, CV_BGR2GRAY );
detectImg = greyImg; // Use the greyscale image.
}
imageProcessed = cvCreateImage(cvSize(inputImg->width, inputImg->height), IPL_DEPTH_8U, 1);
cvResize(detectImg, imageProcessed, CV_INTER_LINEAR);
cvEqualizeHist(imageProcessed, imageProcessed);
return imageProcessed;
}
IplImage* resizeImage(const IplImage *origImg, int newWidth,
int newHeight, bool keepAspectRatio)
{
IplImage *outImg = 0;
int origWidth;
int origHeight;
if (origImg) {
origWidth = origImg->width;
origHeight = origImg->height;
}
if (newWidth <= 0 || newHeight <= 0 || origImg == 0
|| origWidth <= 0 || origHeight <= 0) {
//cerr << "ERROR: Bad desired image size of " << newWidth
// << "x" << newHeight << " in resizeImage().\n";
exit(1);
}
if (keepAspectRatio) {
// Resize the image without changing its aspect ratio,
// by cropping off the edges and enlarging the middle section.
CvRect r;
// input aspect ratio
float origAspect = (origWidth / (float)origHeight);
// output aspect ratio
float newAspect = (newWidth / (float)newHeight);
// crop width to be origHeight * newAspect
if (origAspect > newAspect) {
int tw = (origHeight * newWidth) / newHeight;
r = cvRect((origWidth - tw)/2, 0, tw, origHeight);
}
else { // crop height to be origWidth / newAspect
int th = (origWidth * newHeight) / newWidth;
r = cvRect(0, (origHeight - th)/2, origWidth, th);
}
IplImage *croppedImg = cropImage(origImg, r);
// Call this function again, with the new aspect ratio image.
// Will do a scaled image resize with the correct aspect ratio.
outImg = resizeImage(croppedImg, newWidth, newHeight, false);
cvReleaseImage( &croppedImg );
}
else {
// Scale the image to the new dimensions,
// even if the aspect ratio will be changed.
outImg = cvCreateImage(cvSize(newWidth, newHeight),
origImg->depth, origImg->nChannels);
if (newWidth > origImg->width && newHeight > origImg->height) {
// Make the image larger
cvResetImageROI((IplImage*)origImg);
// CV_INTER_LINEAR: good at enlarging.
// CV_INTER_CUBIC: good at enlarging.
cvResize(origImg, outImg, CV_INTER_LINEAR);
}
else {
// Make the image smaller
cvResetImageROI((IplImage*)origImg);
// CV_INTER_AREA: good at shrinking (decimation) only.
cvResize(origImg, outImg, CV_INTER_AREA);
}
}
return outImg;
}
void learn()
{
int i, offset;
// load training data
nTrainFaces = loadFaceImgArray("C:/Users/HP/Desktop/OpenCV/50_images_of_15_people.txt");
if( nTrainFaces < 2 )
{
fprintf(stderr,
"Need 2 or more training faces\n"
"Input file contains only %d\n", nTrainFaces);
return;
}
// do PCA on the training faces
doPCA();
// project the training images onto the PCA subspace
projectedTrainFaceMat = cvCreateMat( nTrainFaces, nEigens, CV_32FC1 );
offset = projectedTrainFaceMat->step / sizeof(float);
for(i=0; i<nTrainFaces; i++)
{
//int offset = i * nEigens;
cvEigenDecomposite(
faceImgArr[i],
nEigens,
eigenVectArr,
0, 0,
pAvgTrainImg,
//projectedTrainFaceMat->data.fl + i*nEigens);
projectedTrainFaceMat->data.fl + i*offset);
}
// store the recognition data as an xml file
storeTrainingData();
}
void recognize()
{
int i, nTestFaces = 0; // the number of test images
CvMat * trainPersonNumMat = 0; // the person numbers during training
float * projectedTestFace = 0;
// load test images and ground truth for person number
nTestFaces = loadFaceImgArray("C:/Users/HP/Desktop/OpenCV/test.txt");
printf("%d test faces loaded\n", nTestFaces);
// load the saved training data
if( !loadTrainingData( &trainPersonNumMat ) ) return;
// project the test images onto the PCA subspace
projectedTestFace = (float *)cvAlloc( nEigens*sizeof(float) );
for(i=0; i<nTestFaces; i++)
{
int iNearest, nearest, truth;
// project the test image onto the PCA subspace
cvEigenDecomposite(
faceImgArr[i],
nEigens,
eigenVectArr,
0, 0,
pAvgTrainImg,
projectedTestFace);
iNearest = findNearestNeighbor(projectedTestFace);
truth = personNumTruthMat->data.i[i];
nearest = trainPersonNumMat->data.i[iNearest];
printf("nearest = %d, Truth = %d\n", nearest, truth);
}
}
int loadTrainingData(CvMat ** pTrainPersonNumMat)
{
CvFileStorage * fileStorage;
int i;
// create a file-storage interface
fileStorage = cvOpenFileStorage( "facedata.xml", 0, CV_STORAGE_READ );
if( !fileStorage )
{
fprintf(stderr, "Can't open facedata.xml\n");
return 0;
}
nEigens = cvReadIntByName(fileStorage, 0, "nEigens", 0);
nTrainFaces = cvReadIntByName(fileStorage, 0, "nTrainFaces", 0);
*pTrainPersonNumMat = (CvMat *)cvReadByName(fileStorage, 0, "trainPersonNumMat", 0);
eigenValMat = (CvMat *)cvReadByName(fileStorage, 0, "eigenValMat", 0);
projectedTrainFaceMat = (CvMat *)cvReadByName(fileStorage, 0, "projectedTrainFaceMat", 0);
pAvgTrainImg = (IplImage *)cvReadByName(fileStorage, 0, "avgTrainImg", 0);
eigenVectArr = (IplImage **)cvAlloc(nTrainFaces*sizeof(IplImage *));
for(i=0; i<nEigens; i++)
{
char varname[200];
sprintf( varname, "eigenVect_%d", i );
eigenVectArr[i] = (IplImage *)cvReadByName(fileStorage, 0, varname, 0);
}
// release the file-storage interface
cvReleaseFileStorage( &fileStorage );
return 1;
}
void storeTrainingData()
{
CvFileStorage * fileStorage;
int i;
// create a file-storage interface
fileStorage = cvOpenFileStorage( "facedata.xml", 0, CV_STORAGE_WRITE );
// store all the data
cvWriteInt( fileStorage, "nEigens", nEigens );
cvWriteInt( fileStorage, "nTrainFaces", nTrainFaces );
cvWrite(fileStorage, "trainPersonNumMat", personNumTruthMat, cvAttrList(0,0));
cvWrite(fileStorage, "eigenValMat", eigenValMat, cvAttrList(0,0));
cvWrite(fileStorage, "projectedTrainFaceMat", projectedTrainFaceMat, cvAttrList(0,0));
cvWrite(fileStorage, "avgTrainImg", pAvgTrainImg, cvAttrList(0,0));
for(i=0; i<nEigens; i++)
{
char varname[200];
sprintf( varname, "eigenVect_%d", i );
cvWrite(fileStorage, varname, eigenVectArr[i], cvAttrList(0,0));
}
// release the file-storage interface
cvReleaseFileStorage( &fileStorage );
}
int findNearestNeighbor(float * projectedTestFace)
{
//double leastDistSq = 1e12;
double leastDistSq = DBL_MAX;
int i, iTrain, iNearest = 0;
for(iTrain=0; iTrain<nTrainFaces; iTrain++)
{
double distSq=0;
for(i=0; i<nEigens; i++)
{
float d_i =
projectedTestFace[i] -
projectedTrainFaceMat->data.fl[iTrain*nEigens + i];
//distSq += d_i*d_i / eigenValMat->data.fl[i]; // Mahalanobis
distSq += d_i*d_i; // Euclidean
}
if(distSq < leastDistSq)
{
leastDistSq = distSq;
iNearest = iTrain;
}
}
return iNearest;
}
void doPCA()
{
int i;
CvTermCriteria calcLimit;
CvSize faceImgSize;
// set the number of eigenvalues to use
nEigens = nTrainFaces-1;
// allocate the eigenvector images
faceImgSize.width = faceImgArr[0]->width;
faceImgSize.height = faceImgArr[0]->height;
eigenVectArr = (IplImage**)cvAlloc(sizeof(IplImage*) * nEigens);
for(i=0; i<nEigens; i++)
eigenVectArr[i] = cvCreateImage(faceImgSize, IPL_DEPTH_32F, 1);
// allocate the eigenvalue array
eigenValMat = cvCreateMat( 1, nEigens, CV_32FC1 );
// allocate the averaged image
pAvgTrainImg = cvCreateImage(faceImgSize, IPL_DEPTH_32F, 1);
// set the PCA termination criterion
calcLimit = cvTermCriteria( CV_TERMCRIT_ITER, nEigens, 1);
// compute average image, eigenvalues, and eigenvectors
cvCalcEigenObjects(
nTrainFaces,
(void*)faceImgArr,
(void*)eigenVectArr,
CV_EIGOBJ_NO_CALLBACK,
0,
0,
&calcLimit,
pAvgTrainImg,
eigenValMat->data.fl);
cvNormalize(eigenValMat, eigenValMat, 1, 0, CV_L1, 0);
}
int loadFaceImgArray(char * filename)
{
FILE * imgListFile = 0;
char imgFilename[512];
int iFace, nFaces=0;
// open the input file
if( !(imgListFile = fopen(filename, "r")) )
{
fprintf(stderr, "Can\'t open file %s\n", filename);
return 0;
}
// count the number of faces
while( fgets(imgFilename, 512, imgListFile) ) ++nFaces;
rewind(imgListFile);
// allocate the face-image array and person number matrix
faceImgArr = (IplImage **)cvAlloc( nFaces*sizeof(IplImage *) );
personNumTruthMat = cvCreateMat( 1, nFaces, CV_32SC1 );
// store the face images in an array
for(iFace=0; iFace<nFaces; iFace++)
{
// read person number and name of image file
fscanf(imgListFile,
"%d %s", personNumTruthMat->data.i+iFace, imgFilename);
// load the face image
faceImgArr[iFace] = cvLoadImage(imgFilename, CV_LOAD_IMAGE_GRAYSCALE);
if( !faceImgArr[iFace] )
{
fprintf(stderr, "Can\'t load image from %s\n", imgFilename);
return 0;
}
}
fclose(imgListFile);
return nFaces;
}
My answer may came late but it might be useful for pals if i answer it.I am working on a similar project and i have faced the same problem.I solved it by writing a function the saves or write the detected,cropped and preprocessed image on to the hard disk of my computer(Using CvWrite).And feeding the parameter of the saved images to the recognition part of the code. It has made my life easier.It has been a bit harder for me to to pass the parameters of the rect of the region of interest. If you or someone else did this it might be great sharing the code with us.
You can use the following code to save the image after resizing it to a constant value using the resizeimage function on you code.
void saveCroppedFaces(CvSeq* tempon,IplImage* DetectedImage)
{
char* name;
int nFaces;
CvRect rect;
nFaces=tempon->total;
name =new char[nFaces];
IplImage* cropped = 0;
IplImage* croppedResized=0;
Mat croped;
for(int k=0;k<nFaces;k++)
{
itoa(k,(name+k),10);
rect = *(CvRect*)cvGetSeqElem( tempon, k );
cropped= cropImage(DetectedImage,rect);
//i can resize the cropped faces in to a fixed size here
//i can write a function to save images and call it so
//that it will save it in to hard drive
//cvNamedWindow((name+k),CV_WINDOW_AUTOSIZE);
//cvShowImage((name+k),cropped);
croppedResized=resizeImage(cropped,60,60);
croped=IplToMatConverter(croppedResized);
saveROI(croped,itoa(k,(name+k),10));
cvReleaseImage(&cropped);
}
name=NULL;
delete[] name;
}
void saveROI(Mat mat,String outputFileName)
{
string store_path("C://Users/sizusuzu/Desktop/Images/FaceDetection2
/"+outputFileName+".jpg");
bool write_success = imwrite(store_path,mat);
}
After this you can change the IplImage* to Mat using
Mat IplToMatConverter(IplImage* imageToMat)
{
Mat mat = cvarrToMat(imageToMat);
return mat;
}
And use the Mat in FaceRecognizer API.Or just do the other/harder way.
Thanks
I just read
int _tmain(int argc, _TCHAR* argv[])
{
.......
}
part of your code. This code is used for detecting the face in the image. Lets say it is Face_x. Now extract features from Face_x, call it as F_x. In your database, you should store features {F_1, F_2,..., F_N} extracted from n different faces {Face_1, Face_2,..Face_N}.
Simple algorithm to recognize Face_x is to calculate Euclidean distances between F_x and n features. The minimum distance (below threshold) gives corresponding face. If the minimum distance is not below threshold then Face_x is a new face. Add feature F_x to database. This way you can increase your database. You can begin your algorithm with no features in database. With each new face, database grows.
I hope the method suggested by me will lead you to the solution