VC VFWcamera 和 VideoCapH263笔记

内容参考：

http://www.codeproject.com/KB/IP/videonet.aspx

Displaying the Captured Video Frame

There are various methods and APIs for displaying the captured frame. You can use SetDIBitsToDevice() method to directly display the frame. But this is quite slow as it is based on Graphics Device Interface (GDI) functions. The better method is to use DrawDib API to draw the frame. The DrawDib functions provide high performance image-drawing capabilities for device-independent bitmaps (DIBs). DrawDib functions write directly to video memory, hence provide better performance.

MSDN：The capPreview macro enables or disables preview mode. In preview mode, frames are transferred from the capture hardware to system memory and then displayed in the capture window using GDI functions

BOOL capPreview(

  hwnd,  

  f);

Here is the brief view of how to use DrawDib API to display frame:
// Initialize DIB for drawing... 
HDRAWDIB hdib=::DrawDibOpen(); 

// Then call this function will suitable parameters.... 
::DrawDibBegin(hdib,...); 

// Now if you are ready with frame data just 
// invoke this function to display the frame 
::DrawDibDraw(hdib,...); 

// Finally termination... 
::DrawDibEnd(hdib); 
::DrawDibClose(hdib);
 
VFWcamerea工程：
使用capCreateCaptureWindow关联一个窗口进行视频显示。
// 设置预览窗口
	CWnd *pWnd=AfxGetMainWnd()->GetDlgItem(IDC_VIDEO);
	//得到预览窗口指针，为Animate控件
	CRect rect;
	pWnd->GetWindowRect(&rect); // 得到窗口大小
	m_hCapWnd=capCreateCaptureWindow("Capture",
		WS_CHILD|WS_VISIBLE,
		0,0,
		rect.Width(),rect.Width(),
		pWnd->GetSafeHwnd(),0); // 设置预览窗口
	if(!m_hCapWnd)
	{
		MessageBox("Set preview window failed!");
		return FALSE;
	}

	// 连接摄像头
	if(!capDriverConnect(m_hCapWnd,0))
	{
		MessageBox("Connect to Camera failed!");
		return FALSE;
	}
	
	//得到驱动器的属性
	capDriverGetCaps(m_hCapWnd,sizeof(CAPDRIVERCAPS), &m_CapDrvCap);
		
	if(m_CapDrvCap.fCaptureInitialized)		// 查看摄像头是否初始化成功
	{
		capGetStatus(m_hCapWnd, &m_CapStatus,sizeof(m_CapStatus)); 
		// 得到驱动器状态
		capPreviewRate(m_hCapWnd,30); // 设置预览帧频，30 ms
	}
	else{
		AfxMessageBox("Camera initilize failed!");
		return FALSE;
	}
 
使用capPreview来预览、显示视频
void CVFWcameraDlg::OnOpen() 
{
	// TODO: Add your control notification handler code here
	capPreview(m_hCapWnd,TRUE);		// 开始预览
}
 
//**********************************************************************************//*********************************************************************************
VideoCap工程：
使用VideoCapture.h and VideoCapture.c实现，其定义了一个VideoCapture类。
初始化VideoCapture类，在vidcap->Initialize()中设置了一个回调函数：
capSetCallbackOnVideoStream(m_capwnd,OnCaptureVideo);
capSetCallbackOnVideoStream
The capSetCallbackOnVideoStream macro sets a callback function in the application. AVICap calls this procedure during streaming capture when a video buffer is filled.
在视频流获取的过程中，当一个视频缓冲区慢之后，AVICap会调用这个回调函数。
// Initialize Video Capture class
	vidcap=new VideoCapture();
	vidcap->SetDialog(this);

	if(vidcap->Initialize()) // Success
	{
		// Set Dialog's BMP info as Video Capture's
		this->m_bmpinfo_dlg = &vidcap->m_bmpinfo;
	}
	else
	{
		MessageBox("Unable to initialize the camera!");
	}
 
初始化DIB
// Initialize DIB for drawing...
	hDrawDIB=::DrawDibOpen();
	if(hDrawDIB!=NULL)
	{
		::DrawDibBegin(hDrawDIB,
					   NULL,
					   -1,				// don't stretch
					   -1,				// don't stretch
					   &m_bmpinfo_dlg->bmiHeader,
					   IMAGE_WIDTH,         // width of image
					   IMAGE_HEIGHT,        // height of image
					   0
					   );
	
	}

点击Start按钮，会开始视频获取
void CVideoCapDlg::OnStart() 
{
	// TODO: Add your control notification handler code here
	if(isCaptureStarted == TRUE) // Already opened
	{
		vidcap->Destroy();
		isCaptureStarted = FALSE;
		SetDlgItemText(IDC_START, "Start");
	}
	else
	{
		if(vidcap->StartCapture() == FALSE)
		{
			MessageBox("Unable to start capture!");
			isCaptureStarted = FALSE;
		}
		else	// Successfully started
		{
			isCaptureStarted = TRUE;
			SetDlgItemText(IDC_START, "Stop");
		}
	}
}
  
然后，在视频流获取的过程，当视频缓冲区满之后，会调用回调函数OnCaptureVideo
LRESULT CALLBACK OnCaptureVideo(HWND mwnd,LPVIDEOHDR lphdr)
{
	VideoCapture *vidcap=(VideoCapture *)capGetUserData(mwnd);

	// 在这里显示视频不好弄，因为要用到hDrawDIB，而这个在Dlg中定义，
	// 而且，显示的区域也在Dlg里，所以在这里调用Dlg的一个函数SendVideo()。
	if(vidcap!=NULL )
		((CVideoCapDlg*) (vidcap->dlg))->DisplayVideo(lphdr->lpData,lphdr->dwBytesUsed);

return TRUE;
}
 
在回调函数里又会调用Dialog类中的DisplayVideo函数
void CVideoCapDlg::DisplayVideo(BYTE *data, int size)
{
	// 设置预览窗口

	// Adjust display windows
	CWnd *wnd;
	CRect rect;

	// For local video display window
	wnd=this->GetDlgItem(IDC_VIDEO);
   	wnd->GetWindowRect(rect);
	ScreenToClient(rect);

	xDisplaywnd=rect.TopLeft().x+(rect.Width()-IMAGE_WIDTH)/2;
	yDisplaywnd=rect.TopLeft().y+(rect.Height()-IMAGE_HEIGHT)/2;
	
	// Centre the local video window
	wnd->SetWindowPos(&wndTop,xDisplaywnd-4,yDisplaywnd-4,
						IMAGE_WIDTH+9,IMAGE_HEIGHT+9,SWP_SHOWWINDOW | SWP_DRAWFRAME);

	// Get Dialog DC
	m_hdc=this->GetDC()->m_hDC;

	// hDrawDIB，前面仅仅是设置了一个这个句柄，而并没有和任何一个DC关联，
	// 所以，这里必须要和一个DC关联。
	::DrawDibDraw(hDrawDIB,
				  m_hdc,
				  xDisplaywnd,			// dest : left pos
				  yDisplaywnd,			// dest : top pos
				  -1,					// don't zoom x
				  -1,					// don't zoom y
				  &m_bmpinfo_dlg->bmiHeader,// bmp header info
				  data,					 // bmp data
				  0,					 // src :left
				  0,					 // src :top
				  IMAGE_WIDTH,				 // src : width
				  IMAGE_HEIGHT,				 // src : height
				  DDF_SAME_DRAW			 // use prev params....
				  );
}
得到IDC_VIDEO所在的矩形区域：
wnd=this->GetDlgItem(IDC_VIDEO);
   	wnd->GetWindowRect(rect);
	ScreenToClient(rect);
  
将坐标原点放在矩形区域的中心：
xDisplaywnd=rect.TopLeft().x+(rect.Width()-IMAGE_WIDTH)/2;
	yDisplaywnd=rect.TopLeft().y+(rect.Height()-IMAGE_HEIGHT)/2;
设置窗口的z轴显示，放在z轴的最上层，-4和+9可以使视频边框不闪，具体什么原因还没弄清楚。
// Centre the local video window
	wnd->SetWindowPos(&wndTop,xDisplaywnd-4,yDisplaywnd-4,
						IMAGE_WIDTH+9,IMAGE_HEIGHT+9,SWP_SHOWWINDOW | SWP_DRAWFRAME);
得到Dialog的DC，显示内容：
// Get Dialog DC
	m_hdc=this->GetDC()->m_hDC;
将前面初始化好的DIB与这里的DC关联，同时也与视频的数据关联：
// hDrawDIB，前面仅仅是设置了一个这个句柄，而并没有和任何一个DC关联，
	// 所以，这里必须要和一个DC关联。
	::DrawDibDraw(hDrawDIB,
				  m_hdc,
				  xDisplaywnd,			// dest : left pos
				  yDisplaywnd,			// dest : top pos
				  -1,					// don't zoom x
				  -1,					// don't zoom y
				  &m_bmpinfo_dlg->bmiHeader,// bmp header info
				  data,					 // bmp data
				  0,					 // src :left
				  0,					 // src :top
				  IMAGE_WIDTH,				 // src : width
				  IMAGE_HEIGHT,			// src : height
				  DDF_SAME_DRAW		// use prev params....
				  );
    
 
Encoder and Decoder Library
Encoder
I have used fast h.263 encoder library for the encoding. This library was the modified version of Tmndecoder to make it more faster for real time encoding. I have converted this library from C to C++ so that it can be integrated into any Windows application easily. I have removed some of the unnecessary codes/files from the fast h263 library. Also moved definitions and declarations in their proper .h and .cpp files.
Brief view of usage of H263 Encoder library
// Initialize the compressor 
CParam cparams; 
cparams.format = CPARAM_QCIF; 
InitH263Encoder(&cparams); 


//If you need conversion from RGB24 to YUV420 then call this 
InitLookupTable(); 


// Set up the callback function 
// OwnWriteFunction is the global function called 
// during encoding to return the encoded data... 
WriteByteFunction = OwnWriteFunction; 


// For compression data must be in YUV420 format... 
// Hence before compression invoke this method 
ConvertRGB2YUV(IMAGE_WIDTH,IMAGE_HEIGHT,data,yuv); 

// Compress the frame..... 
cparams.format=CPARAM_QCIF; 
cparams.inter = CPARAM_INTRA; 
cparams.Q_intra = 8; 
cparams.data=yuv; // Data in YUV format... 
CompressFrame(&cparams, &bits); 

// You can get the compressed data from callback function 
// that you have registerd at the begining... 

// Finally terminate the encoder 
// ExitH263Encoder();
Decoder
This is the modified version of tmndecoder (H.263 decoder). It was in ANSI C. I have converted it into C++ so that it can be integrated into any Windows application. I have removed some of the files which had display and file storing functions. I have removed the unnecessary code and also added some new files.
Original library dealt with files. It was not suitable to use for real time decoding. I have done some major changes so that it can be easily integrated into the application for real time decoding process. Now one can use this library for decoding H263 frames. This library is quite fast and gives better performance.
Usage of Decoder .....
//Initialize the decoder 
InitH263Decoder(); 

// Decompress the frame.... 
// > rgbdata must be large enough to hold the output data... 
// > decoder produces the image data in YUV420 format.After 
// decoding it is converted into RGB24 format... 
DecompressFrame(data,size,rgbdata,buffersize); 

// Finaly terminate the decoder 
ExitH263Decoder();