David_Lavy's profile - activity

Thank you for your reply. I think the wall example is a bit extreme, and I should've been a bit clearer. In my specific

How to obtain an 'approximate' true color from an image in different lightning conditions? I'm working on an app that ca

Thank you! That solved the problem.

Hi,

I'm trying to run the tutorial provided here. The training model works just fine. However, when I do the testing I always get a segmentation fault. And this is due to using the learning based model.

I went and created a simple test using the .yml file created from the training test here. Uploaded any picture and I get the same segmentation error. Debugging the code it seems the error is in this line:

wb->balanceWhite(img, img_out);

But I have no idea why this error shows up. Any help is appreciated.

Thanks!

Here it is: https://drive.google.com/file/d/1_5PUUb_VsSC18nMzsy_DYzWNnO926UI9/view?usp=sharing

Something that I'm realizing is that I think the WhiteBalance pointer is not using the yml I'm referencing. If I switch

The same "Segmentation fault (core dumped)"

Problem using xphoto::LearningBasedWB Hi, I'm trying to run the tutorial provided here. The training model works just

Thank you, I modified the last part of the main function and I indicated it in my post as an update, my problem now is, when I run the program it doesn't stream the video since the beginning, it only shows the first frame until I click on the screen, and then streams the video, how can I modify so it can stream the video right after I run my program?

Hi, I'm trying to select a rectangle in a video, but when I run my code nothing shows up; this is the code I'm using:

cv::Mat frame;
const char* src_window = "Select ROI";

int drag = 0, select_flag = 0;

cv::Point point1, point2;    
bool callback = false;
void mouseHandler(int event, int x, int y, int flags, void* param);

int main(int argc, char** argv) {

cv::VideoCapture cap;

if( argc == 1 || (argc == 2 && strlen(argv[1]) == 1 && isdigit(argv[1][0])))
        cap.open(argc == 2 ? argv[1][0] - '0' : 0);
else if( argc == 2 )
    cap.open(argv[1]);

if( !cap.isOpened() )
{
    std::cout << "Could not initialize capturing...\n";
    return 0;
}

cap >> frame;

cv::namedWindow(src_window,CV_WINDOW_AUTOSIZE);
//cv::imshow(src_window,frame);
//cv::setMouseCallback(src_window,mouseHandler,0);

for (;;) {
    cap >> frame;
    if( frame.empty() )
            break;

    if (callback == false) {
        cv::setMouseCallback(src_window,mouseHandler,0);
    }
    else {
    cv::imshow(src_window,frame);
    }
}
    cv::waitKey(0);
    return 0;
}

void mouseHandler(int event, int x, int y, int flags, void* param)
{
if (event == CV_EVENT_LBUTTONDOWN && !drag && !select_flag)
{
    /* left button clicked. ROI selection begins */
    point1 = cv::Point(x, y);
    drag = 1;
}

if (event == CV_EVENT_MOUSEMOVE && drag && !select_flag)
{
    /* mouse dragged. ROI being selected */
    cv::Mat img1 = frame.clone();
    point2 = cv::Point(x, y);
    cv::rectangle(img1, point1, point2, CV_RGB(255, 0, 0), 3, 8, 0);
    cv::imshow(src_window, img1);
}

if (event == CV_EVENT_LBUTTONUP && drag && !select_flag)
{
    cv::Mat img2 = frame.clone();
    point2 = cv::Point(x, y);
    drag = 0;
    select_flag = 1;
    cv::imshow(src_window, img2);
    callback = true;
}
}

I don't know what can I be doing wrong. Thanks for your help.

UPDATE: Thanks to Vladislav Vinogradov now I can stream the video, but when I run the program it shows the first frame and stays like that until I click on the image, then starts the streaming, how can I modify this part so the streaming occurs since the beginning?

cap >> frame;

cv::namedWindow(src_window,CV_WINDOW_AUTOSIZE);
cv::imshow(src_window,frame);

for (;;) {
    if (callback) {
        cap >> frame;
        if( frame.empty() )
                break;
        cv::imshow(src_window,frame);
    }
    else {
        cap >> frame;
        cv::setMouseCallback(src_window,mouseHandler,0);
    }
    cv::waitKey(5);
}

Hi,

So I have a matrix cv::Mat M (NxN) and a vector of Points std::vector< cv::Point2f > Vec (Nx2).

How can I perform a multiplication between these two elements: M x Vec = P, and have a final vector std::vector< cv::Point2f > P (Nx2)??

Thank you!

Hi,

Let's say I have two vectors std::vector< float > pX,pY. Is there any way I can fusion those 2 vectors into one single std::vector< cv::Point2f > points, where the vector pX is the X elements and pY the Y elements of the vector points? I know I can use a for loop, but I wanted to know a more matricial, or simpler function can achieve this.

Thanks!

Hi, I have a problem doing the first part of the tutorial XML/YML Persistence (C API))

It fails with the error:

OpenCV Error: Unspecified error (Incorrect element name {:) in operator<<, file /tmp/buildd/ros-fuerte-opencv2-2.4.2-0oneiric-20120908-1631/modules/core/src/persistence.cpp, line 5239 terminate called after throwing an instance of 'cv::Exception' what(): /tmp/buildd/ros-fuerte-opencv2-2.4.2-0oneiric-20120908-1631/modules/core/src/persistence.cpp:5239: error: (-2) Incorrect element name {: in function operator<<in function="" operator&lt;&lt;<="" p="">

The same issue was explained here but it only says the issue is fixed now, although doesn't say how was it fixed. I need help with this.

Thank you!

P.S. I'm using OpenCV 2.4.2 from ROS and Ubuntu 11.10

Hi, so I've installed ROS Fuerte in my PC, and now I want to use the OpenCV libraries from it in a Qt project. I don't know how to do it ... Before I had ROS Electric and by only adding:

unix {    
CONFIG += link_pkgconfig
PKGCONFIG += opencv
}

To my .pro file was enough to start using OpenCV, but this doesn't work now. Do you have any idea how to do it? Thanks.

Additional Info:

• Ubuntu version 11.10
• OpenCV version 2.4.2
• ROS Fuerte
• Qt 4.7.3