Video for Windows --> Drawing a Line on the Capture Window

Posted on 2004-04-09
Last Modified: 2007-12-19
I'd like to ask if anyone  have already done this trick.
I'd like to draw a line on the capture window for my program.
I'd like to draw a small cross on the center of the capture
If I can use a function can anybody share this information(function),
if not can anybody share the procedure on how to do this trick.

Thank you ing advance.

Question by:colossus_21
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 3
LVL 48

Expert Comment

ID: 10789937
I have the project which does exactly the same trick. This is MFC application, you you don't use MFC, convert this to API. Appropriate code fragments:

BOOL CVideoCaptureDlg::InitCapture()
    m_hCaptureWnd = capCreateCaptureWindow(

        // ... show available drivers in listbox

BOOL CVideoCaptureDlg::Connect()
    int n = m_list_drivers.GetCurSel();  // driver number

    if ( n < 0 )
        MessageBox("Capture driver is not selected");
        return FALSE;

    if ( ! capDriverConnect(m_hCaptureWnd, n) )
        MessageBox("capDriverConnect failed");
        return FALSE;

    CAPSTATUS CapStatus;

    if ( ! capGetStatus(m_hCaptureWnd, &CapStatus, sizeof (CAPSTATUS)) )
        MessageBox("capGetStatus failed");
        return FALSE;

    ::MoveWindow(m_hCaptureWnd, 0, 0,
                 CapStatus.uiImageWidth, CapStatus.uiImageHeight, FALSE);
    ::ShowWindow(m_hCaptureWnd, SW_SHOW);

    if ( ! capDriverGetCaps(m_hCaptureWnd, &m_DriverCaps, sizeof(CAPDRIVERCAPS)) )
        MessageBox("capDriverGetCaps failed");
        return FALSE;

    if ( ! capSetCallbackOnError(m_hCaptureWnd, ErrorCallback) )
        MessageBox("capSetCallbackOnError failed");
        return FALSE;

    // !!! set frame callback - this is what you need
    if ( ! capSetCallbackOnFrame(m_hCaptureWnd, FrameCallback) )
        MessageBox("capSetCallbackOnFrame failed");
        return FALSE;

    return TRUE;

// This function is used to set RGB video format programmatically -
// I need this format to draw in the video frame
BOOL CVideoCaptureDlg::SetVideoFormatRGB()
    DWORD size = capGetVideoFormatSize(m_hCaptureWnd);

    if ( size <= 0 )
        AddString("capGetVideoFormatSize failed");
        return FALSE;

    void* pBuffer = new BYTE[size];
    BITMAPINFO* pBitmapInfo = (BITMAPINFO*) pBuffer;

    if ( capGetVideoFormat(m_hCaptureWnd, pBuffer, size) <= 0 )
        delete[] pBuffer;
        AddString("capGetVideoFormat failed");
        return FALSE;

    pBitmapInfo->bmiHeader.biBitCount = 24;
    pBitmapInfo->bmiHeader.biCompression = 0;
    pBitmapInfo->bmiHeader.biSizeImage = pBitmapInfo->bmiHeader.biWidth * pBitmapInfo->bmiHeader.biHeight * 3;

    if ( ! capSetVideoFormat(m_hCaptureWnd, pBitmapInfo, size) )
        delete[] pBuffer;
        AddString("capSetVideoFormat failed");
        return FALSE;

    // keep results in global variables to be available in callback functions
    g_nImageWidth = pBitmapInfo->bmiHeader.biWidth;
    g_nImageHeight = pBitmapInfo->bmiHeader.biHeight;
    g_nBitCount = pBitmapInfo->bmiHeader.biBitCount;
    g_nBitsPerPixel = pBitmapInfo->bmiHeader.biBitCount;

    delete[] pBuffer;
    return TRUE;

This is general stuff, I suppose you have something like tiis in your project. Important thing is capSetCallbackOnFrame and SetVideoFormatRGB which sets well-known RGB video format.

#define LINE_LEN 10

    if ( g_nBitCount != 24 )   // only RGB format !
        return 0;

    if ( ! g_hFrameBitmap )

    if ( ! g_hFrameBitmap )   // failed
        return 0;

    HPEN hOldPen = (HPEN)SelectObject(g_hFrameDC, g_hWhitePen);       // global pen handle created in initialization code
    HBRUSH hOldBrush = (HBRUSH)SelectObject(g_hFrameDC, g_hBrush);// global brush handle created in initialization code

    // copy video frame to bitmap bits
    SetBitmapBits(g_hFrameBitmap, g_nImageWidth*g_nImageHeight*3, lpVHdr->lpData);

    // draw something
    MoveToEx(g_hFrameDC, g_nImageWidth/2 - LINE_LEN, g_nImageHeight/2, NULL);
    LineTo(g_hFrameDC, g_nImageWidth/2 + LINE_LEN + 1, g_nImageHeight/2);
    MoveToEx(g_hFrameDC, g_nImageWidth/2, g_nImageHeight/2 - LINE_LEN, NULL);
    LineTo(g_hFrameDC, g_nImageWidth/2, g_nImageHeight/2 + LINE_LEN + 1);

    SelectObject(g_hFrameDC, g_hBlackPen);

    MoveToEx(g_hFrameDC, g_nImageWidth/2 - LINE_LEN, g_nImageHeight/2 - 1, NULL);
    LineTo(g_hFrameDC, g_nImageWidth/2 + LINE_LEN + 1, g_nImageHeight/2 - 1);
    MoveToEx(g_hFrameDC, g_nImageWidth/2 - 1, g_nImageHeight/2 - LINE_LEN, NULL);
    LineTo(g_hFrameDC, g_nImageWidth/2 - 1, g_nImageHeight/2 + LINE_LEN + 1);

    SelectObject(g_hFrameDC, hOldBrush);
    SelectObject(g_hFrameDC, hOldPen);

    // copy bitmap bits back to video frame    
    GetBitmapBits(g_hFrameBitmap, g_nImageWidth*g_nImageHeight*3, lpVHdr->lpData);

    return 0;

#define WIDTHBYTES(bits) ((((bits) + 31) / 32) * 4)

void CreateGraphicObjects(HWND hWnd)
    HDC hDC = ::GetDC(hWnd);
    g_hFrameDC = CreateCompatibleDC(hDC);
    ::ReleaseDC(hWnd, hDC);


    lpbi = (LPBITMAPINFO) new BYTE[sizeof(BITMAPINFOHEADER) + (256 * sizeof(RGBQUAD))];
    lpbi->bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
    lpbi->bmiHeader.biWidth = g_nImageWidth;
    lpbi->bmiHeader.biHeight = g_nImageHeight;
    lpbi->bmiHeader.biPlanes = 1;
    lpbi->bmiHeader.biBitCount = 24;
    lpbi->bmiHeader.biCompression = BI_RGB;
    lpbi->bmiHeader.biSizeImage = WIDTHBYTES((DWORD)g_nImageWidth * 8) * g_nImageHeight;
    lpbi->bmiHeader.biXPelsPerMeter = 0;
    lpbi->bmiHeader.biYPelsPerMeter = 0;
    lpbi->bmiHeader.biClrUsed = 0;
    lpbi->bmiHeader.biClrImportant = 0;

    BYTE* pBits;

    g_hFrameBitmap = CreateDIBSection(
        NULL,                       // DC - we don't need it here
        lpbi,                       // BITMAPINFOHEADER pointer
        DIB_RGB_COLORS,             // palette is in BITMAPINFO, contains RGBQUAD entries (doesn't matter for RGB)
        (void **)&pBits,             // output pointer to bitmap bits
        0 );

    delete[] lpbi;

    SelectObject(g_hFrameDC, (HBITMAP)g_hFrameBitmap);

g_hFrameBitmap is 24 bpp DIB selected in memory DC g_hFrameDC. Each frame is copied to this bitmap bits, after this lines (and some other stuff in my project) are drawn in this DC, and bitmap bits are copied back to video frame.

Expert Comment

ID: 10789955

if you can get the window handles of your programs windows, but not others......

You can call "GetDC" function to get the "display device context" handle for the client area of that window.
And can draw anything...

check this link..

To draw line you can use "Line Functions"

--> MoveToEx
--> LineTo

check this link....

If you are using win32 i.e WinMain()
Then in the WM_PAINT Message of "Window Procedure" you have to call line functions.

LVL 48

Accepted Solution

AlexFM earned 250 total points
ID: 10790673
By the way, you can draw line manually setting pixel values in lpVHdr->lpData array. Drawing horizontal line may be done using memset, drawing of vertical line may be done in loop calculating every pixel address.

    // change pixels here directly setting them using lpVHdr->lpData array.
    // Each pixel gets 3 bytes (assuming video format is RGB).
    // If you know image size, you have enough information to do this.

    // Example:
    memset(lpVHdr->lpData, 0, 300);   // first 100 pixels in the video capture imahe will be black

Which approach to use - it depends on complexity of graphics overlay you want to draw. Simple line may be drawn by direct pixel access. Complicated overlay is better to draw using device context.
LVL 48

Expert Comment

ID: 10992840
I think my answer is OK.

Featured Post

Want Experts Exchange at your fingertips?

With Experts Exchange’s latest app release, you can now experience our most recent features, updates, and the same community interface while on-the-go. Download our latest app release at the Android or Apple stores today!

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

This article shows you how to optimize memory allocations in C++ using placement new. Applicable especially to usecases dealing with creation of large number of objects. A brief on problem: Lets take example problem for simplicity: - I have a G…
Basic understanding on "OO- Object Orientation" is needed for designing a logical solution to solve a problem. Basic OOAD is a prerequisite for a coder to ensure that they follow the basic design of OO. This would help developers to understand the b…
The viewer will learn how to pass data into a function in C++. This is one step further in using functions. Instead of only printing text onto the console, the function will be able to perform calculations with argumentents given by the user.
The viewer will learn how to user default arguments when defining functions. This method of defining functions will be contrasted with the non-default-argument of defining functions.

623 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question