aligoglos's profile - activity

how to calculate scale with matchTemplate I have problem that I hope someone give me a idea. In general I want to use ma

I want to convert some codes from python to c++ . for example we have 3 dimensional array like (25 * 40 * 10) and we want to separate array, in python we have numpy arrays and simply do it with:

Probs = Y[...,0]
Affines = Y[...,2:] 
rx,ry = Y.shape[:2]
ywh = Y.shape[1::-1]
iwh = np.array(I.shape[1::-1],dtype=float).reshape((2,1))

but how can I implement this line in c++ with Mat?

Sub Mat in mulidimensional Mat I want to convert some codes from python to c++ . for example we have 3 dimensional array

Sub Mat in mulidimensional Mat I want convert some codes from python to c++ . for example we have 3 dimensional array li

i train cascade classifier in 20k positive and negative samples but after 1 day because power outage, turn off my pc. know there is way to recover my training from stage 3 or must began from first ?

but whats mean 2(r21,r11) ? mean of ',' its product or other mean ?

i use landmark to extract face points and use solvePnP to estimate head pose but now i don't know how calculate head angles from X, Y and Z axis with solvePnP output parameters ? i try to convert rvet values to degree but seem some times values bigger than 180 !!!

i want stream IP camera that compression is H264 i download open.h264 from openh and add system path var OPENH264_LIBRARY to dll files and even copy dll to my exe path but when i get image from camera this errors occur :

[h264 @ 000001ef9fd88c40] missing picture in access unit with size 120
[h264 @ 000001ef9fd88c40] No start code is found.
[h264 @ 000001ef9fd88c40] Error splitting the input into NAL units.

my code :

video_path = "http://admin:[email protected]:80/video.cgi?&.H264";
mCamera = new cv::VideoCapture();
mCamera->set(CV_CAP_PROP_FOURCC, CV_FOURCC('X', '2', '6', '4'));
isOpenedFile = mCamera->open(video_path);

thanks works. but i think must write in tracking README file that tracking depend on dnn.

I'm sorry I was wrong. i use im.ptr<float>(i)[j] but my Mat type is CV_8UC. i convert my Mat to CV_32F and use im.at<float>(i,j), fixed issue.

but i don't use With_opencl when i compiled.

in my function get Mat image : imtemp2 = cv::abs(im);

i want use abs function but every time i use it a type of exception occur !! once this exception :

Exception thrown: read access violation.

on below code in ocl.cpp line 4 :

bool useOpenCL()
{
    CoreTLSData* data = getCoreTlsData().get();
    if( data->useOpenCL < 0 )
    {
        try
        {
            data->useOpenCL = (int)haveOpenCL() && Device::getDefault().ptr() && Device::getDefault().available();
        }
        catch (...)
        {
            data->useOpenCL = 0;
        }
    }
    return data->useOpenCL > 0;
}

other this :

A heap has been corrupted .

on below code in alloc.cpp line 3:

void* fastMalloc( size_t size )
{
    uchar* udata = (uchar*)malloc(size + sizeof(void*) + CV_MALLOC_ALIGN);
    if(!udata)
        return OutOfMemoryError(size);
    uchar** adata = alignPtr((uchar**)udata + 1, CV_MALLOC_ALIGN);
    adata[-1] = udata;
    return adata;
}

another exception :

A heap has been corrupted  (ntdll.dll).

on opencl_core.cpp line 110 : handle = LoadLibraryA(path);

I am totally confused please help.

i have vector<vector<float> > of features in size 1800*160 now i need train svm on it, i try use OPENCV SVM but in debug mode svm->train return false and in release mode this exeption raised :

Exception thrown at 0x00007FFF587AC387 (vcruntime140.dll):Access violation reading location 0x00000048B7FED000.

my code:

void Classifier::trainSVM(vector<vector<float> > data,cv::Mat Lable)
{
    // Train the SVM
    cv::Ptr<cv::ml::SVM> svm = cv::ml::SVM::create();
    svm->setType(cv::ml::SVM::C_SVC);
    svm->setKernel(cv::ml::SVM::LINEAR);
    svm->setTermCriteria(cv::TermCriteria(cv::TermCriteria::MAX_ITER, 50000, 1e-6));
    cv::Mat trainingData = cv::Mat(data.size(), 160, CV_32FC1, data.data());
    std::cout << "\nBegan Training Svm in vector faces.";
    bool trained = svm->train(trainingData, cv::ml::ROW_SAMPLE, Lable);
    if (trained)
        svm->save("svm_data.xml");
    std::cout << "\nEnd Training Svm in vector faces.";

}

my opencv version is: 3.2

first i want know how much face angles change, haarcascade can detect faces?(because that seems haarcascade is very sensitive to angle) i want real time detect faces to 20 degree maximum and if no faces found report this. i use camshift tracking to improve face detection but track face in every angle and degree and i don't know how can detect in tracking step angle is more than 20 degree? thanks of your helping.

i want use ShapeContextDistanceExtractor according to shape_example.cpp source code in real time application and my images and mask is 85*55 pixel and simpleContour size is 300 but every ShapeContextDistanceExtractor.computeDistance takes about 18 to 20 seconds. why computeDistance is slow? Does the implementation in gpu? thanks.

i train ann to detect eye state with out problem but when i want to load pre trained file and predict samples this exception occurred : my training image size is 24*24;

Exception thrown at 0x00007FF8BB6980DC (opencv_world310d.dll) in SvmTrainingEye.exe: 0xC0000005: Access violation reading location 0x0000001F8768B000.

this is my code:

Mat openEye, closeEye,imgOpen,imgClose;
imgOpen = cv::imread("open.jpg");
imgClose = cv::imread("close.jpg");

cv::resize(imgOpen, openEye, cv::Size(24,24));
cv::resize(imgClose, closeEye, cv::Size(24, 24));

Mat lbpOpen, lbpClose;
if (openEye.channels() > 1)
    cvtColor(openEye, openEye, CV_BGR2GRAY);
lbp::ELBP(openEye, lbpOpen, 1, 16);
if (closeEye.channels() > 1)
    cvtColor(closeEye, closeEye, CV_BGR2GRAY);
lbp::ELBP(closeEye, lbpClose, 1, 16);

lbpOpen = lbpOpen.reshape(0, 1);
lbpClose = lbpClose.reshape(0, 1);

lbpOpen.convertTo(lbpOpen, CV_32FC1);
lbpClose.convertTo(lbpClose, CV_32FC1);

cv::Ptr<cv::ml::ANN_MLP> eyeNetwork;
eyeNetwork = Algorithm::load<ml::ANN_MLP>("Ann_sigmoid_eye.xml");

Mat outputOpen, outputClose;

eyeNetwork->predict(lbpOpen, outputOpen);
eyeNetwork->predict(lbpClose, outputClose);

this fixed by :

Data.convertTo(Data, CV_32FC1); Lables.convertTo(Lables, CV_32FC1);

i comment this line : //Data = Mat(numberOfCloseEye + numberOfOpenEye, imageDimention*imageDimention, CV_32FC1);

but previous exception again occurred.

i apply this changes but this time this exception occurred : OpenCV Error: Assertion failed ((layout == ROW_SAMPLE && responses.rows == nsamples) || (layout == COL_SAMPLE && responses.cols == nsamples)) in cv::ml::TrainDataImpl::setData, file C:\buildslave64\win64_amdocl\master_PackSlave-win64-vc14-shared\opencv\modules\ml\src\data.cpp, line 290

i want train neural network to classify two type of images but when train network this error occurred :

OpenCV Error: Bad argument (output training data should be a floating-point matrix with the number of rows equal to the number of training samples and the number of columns equal to the size of last (output) layer) in cv::ml::ANN_MLPImpl::prepare_to_train, file C:\buildslave64\win64_amdocl\master_PackSlave-win64-vc14-shared\opencv\modules\ml\src\ann_mlp.cpp, line 675

my code :

#include "opencv2\core.hpp"#include "opencv2\imgproc.hpp"#include "opencv2\imgcodecs.hpp"#include "opencv2\highgui.hpp"#include "opencv2\ml.hpp"#include <string>#include "lbp.h"
using namespace cv;
using namespace cv::ml;
void LoadTrainingData();
 
Mat Data;
Mat Lables;
//const int numberOfClass1 = 2384;
//const int numberOfClass2 = 2462;
const int numberOfClass1 = 23;
const int numberOfClass2 = 24;
int Class1 = 1;
int Class2 = -1;
const int imageDimention = 22;
 
std::string NumberToString(size_t Number)
{
    std::stringstream ss;
    ss << Number;
    return ss.str();
}
 
void main() {
    LoadTrainingData();
    Ptr<ann_mlp> annClassifier;
    annClassifier = ANN_MLP::create();
    annClassifier->setActivationFunction(ANN_MLP::ActivationFunctions::SIGMOID_SYM);
    Mat layers(1, 3, CV_32F);
    layers.at<float>(0) = Data.cols;
    layers.at<float>(1) = 100;
    layers.at<float>(2) = 2;
    annClassifier->setLayerSizes(layers);
    annClassifier->setTermCriteria(TermCriteria(TermCriteria::MAX_ITER, 100, 1e-6));
    annClassifier->setTrainMethod(ANN_MLP::TrainingMethods::BACKPROP);
    bool trained = annClassifier->train(Data,ROW_SAMPLE,Lables);
    if (trained)
        annClassifier->save("Ann_sigmoid_eye");
}
 
void LoadTrainingData() {
    Data = Mat(numberOfClass1 + numberOfClass2, imageDimention*imageDimention, CV_32FC1);
    Lables = Mat(numberOfClass1 + numberOfClass2,1 , CV_32SC1);
    // load openEye 
    Mat img;
    Mat lbpImg;
    Mat row;
    std::string path;
    for (size_t i = 1; i <= numberOfClass2; i++)
    {
        path = "class1 (" + NumberToString(i) + ").jpg";
        img = imread(path);
        if (img.channels() > 1)
            cvtColor(img, img, CV_BGR2GRAY);
        lbp::ELBP(img,lbpImg, 1, 16); 

        row = lbpImg.reshape(0, 1);
        row.convertTo(row, CV_32FC1);
        Data.push_back(row);
        Lables.push_back(Class1);
    }
    for (size_t i = 1; i <= numberOfClass1; i++)
    {
        path ="class2 (" + NumberToString(i) + ").jpg";
        img = imread(path);
        if (img.channels() > 1)
            cvtColor(img, img, CV_BGR2GRAY);
        lbp::ELBP(img,lbpImg, 1, 16);
        row = lbpImg.reshape(0, 1);
        row.convertTo(row, CV_32FC1);
        Data.push_back(row);
        Lables.push_back(Class2);
    }
}

i don't why this happen ! please help me,thanks.

i install visual studio 2015 (vs 14) and reconfiguration linker to vs 14 and copy new vs14 dll into my debug folder but didn't different this exception occurred. i thing is bug in opencv library. my new information : window 10 X64 , opencv 3.1 , visual studio 2015 update 1

and add both of opencv_world310.dll , opencv_world310d.dll in debug directory

i add both of : opencv_world310.lib , opencv_world310d.lib into AdditionalDependencies and my My PC is an x64 architecture.