How to display OpenCV Mat on MFC View - c++

I thought displaying OpenCV2 Mat on MFC View is simple but is not. This is only relevant material I found on google. Excuse me for my ignorance but I can't find any other materials showing how to use SetDIBitsToDevice with one dimensional array "data" member returns. More specifically, I need to know how to specify BITMAPINFO for the function. Do I go back to Old C-style OpenCV to work with MFC?
UPDATE:
I found an example of SetDIBitsToDevice which is actually for old C-style OpenCV. But it was straightforward to convert it for OpenCV2. There are things I need to mention to make it work:
Bpp method does not work well as Mat's depth returns 0. I just changed like this:
static int Bpp(cv::Mat img) { return 8 * img.channels(); }
Mat does not have origin member. But simply putting 0 is fine for origin argument of FillBitmapInfo method.
Other than that, following code works great. Hope this helps other devs too.
void COpenCVTestView::OnDraw(CDC* pDC)
{
COpenCVTestDoc* pDoc = GetDocument();
ASSERT_VALID(pDoc);
if (!pDoc)
return;
if(pDoc->m_cvImage.empty()) return;
// TODO: add draw code for native data here
int height=pDoc->m_cvImage.rows;
int width=pDoc->m_cvImage.cols;
uchar buffer[sizeof( BITMAPINFOHEADER ) + 1024];
BITMAPINFO* bmi = (BITMAPINFO* )buffer;
FillBitmapInfo(bmi,width,height,Bpp(pDoc->m_cvImage),0);
SetDIBitsToDevice(pDC->GetSafeHdc(), 0, 0, width,
height, 0, 0, 0, height, pDoc->m_cvImage.data, bmi,
DIB_RGB_COLORS);
}
void COpenCVTestView::FillBitmapInfo(BITMAPINFO* bmi, int width, int height, int bpp, int origin)
{
assert(bmi && width >= 0 && height >= 0 && (bpp == 8 || bpp == 24 || bpp == 32));
BITMAPINFOHEADER* bmih = &(bmi->bmiHeader);
memset(bmih, 0, sizeof(*bmih));
bmih->biSize = sizeof(BITMAPINFOHEADER);
bmih->biWidth = width;
bmih->biHeight = origin ? abs(height) : -abs(height);
bmih->biPlanes = 1;
bmih->biBitCount = (unsigned short)bpp;
bmih->biCompression = BI_RGB;
if (bpp == 8)
{
RGBQUAD* palette = bmi->bmiColors;
for (int i = 0; i < 256; i++)
{
palette[i].rgbBlue = palette[i].rgbGreen = palette[i].rgbRed = (BYTE)i;
palette[i].rgbReserved = 0;
}
}
}

Here is another possible way of displaying OpenCV data in MFC which I use and works great:
IplImage* image// <-- this contains the image you want to display
CvvImage tempdefault;
RECT myrect; // <-- specifiy where on the screen you want it to be displayed
myrect.top = 0;
myrect.bottom = _pictureh;
myrect.left = _picturex;
myrect.right = _picturew+_picturex;
tempdefault.Create(_pictureh,_picturew,32);
tempdefault.CopyOf(image);
tempdefault.DrawToHDC(pDC->GetSafeHdc(),&myrect);

From MSDN:
lpvBits [in]
A pointer to the color data stored as an array of bytes. For more information, see the following Remarks section.
This is the pointer you must init with the data returned from Mat::data.

CvvImage is not available in new versions of OpenCV. Using the following code you can convert Mat to CImage and then display CImage everywhere you want:
int Mat2CImage(Mat *mat, CImage &img){
if(!mat || mat->empty())
return -1;
int nBPP = mat->channels()*8;
img.Create(mat->cols, mat->rows, nBPP);
if(nBPP == 8)
{
static RGBQUAD pRGB[256];
for (int i = 0; i < 256; i++)
pRGB[i].rgbBlue = pRGB[i].rgbGreen = pRGB[i].rgbRed = i;
img.SetColorTable(0, 256, pRGB);
}
uchar* psrc = mat->data;
uchar* pdst = (uchar*) img.GetBits();
int imgPitch = img.GetPitch();
for(int y = 0; y < mat->rows; y++)
{
memcpy(pdst, psrc, mat->cols*mat->channels());//mat->step is incorrect for those images created by roi (sub-images!)
psrc += mat->step;
pdst += imgPitch;
}
return 0;
}

Related

MFC casting Handle into pointer and DIB to DDB conversion

I am trying to create a bitmap by hardcoding an array of pixel values, converting this array of pixels into a DIB, and then turn this DIB into a DDB. I found two functions to convert CreateBitmapFromPixels and DIBToDDB on the internet. My problem is that the program would crash at line 244. I found that, at line 243, lpbi does not retrieve information from hDIB. Then I added the code at lines 229 and 230 to see if doing the same thing in the function that created the BITMAPINFO structure would help. Still, nothing was gotten from the HBITMAP. I am wondering if there is anything wrong with casting a handle into a pointer, what does it do, and are there other ways to get the HBITMAPINFOHEADER from a handle to a DIB so I can fix the problem.
HBITMAP ColorChange2Dlg::CreateBitmapFromPixels( HDC hDC,
UINT uWidth, UINT uHeight, UINT uBitsPerPixel, LPVOID pBits)
{
if(uBitsPerPixel < 8) // NOT IMPLEMENTED YET
return NULL;
if(uBitsPerPixel == 8)
return Create8bppBitmap(hDC, uWidth, uHeight, pBits);
HBITMAP hBitmap = 0;
if ( !uWidth || !uHeight || !uBitsPerPixel )
return hBitmap;
LONG lBmpSize = uWidth * uHeight * (uBitsPerPixel/8) ;
BITMAPINFO bmpInfo = { 0 };
bmpInfo.bmiHeader.biBitCount = uBitsPerPixel;
bmpInfo.bmiHeader.biHeight = uHeight;
bmpInfo.bmiHeader.biWidth = uWidth;
bmpInfo.bmiHeader.biPlanes = 1;
bmpInfo.bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
if(bmpInfo.bmiHeader.biBitCount==32) {
bmpInfo.bmiHeader.biCompression=BI_RGB;
//bmpInfo.bmiColors=NULL;
}
// Pointer to access the pixels of bitmap
UINT * pPixels = 0;
hBitmap = CreateDIBSection( hDC, (BITMAPINFO *)&
bmpInfo, DIB_RGB_COLORS, (void **)&
pPixels , NULL, 0);
if ( !hBitmap )
return hBitmap; // return if invalid bitmaps
//SetBitmapBits( hBitmap, lBmpSize, pBits);
// Directly Write
memcpy(pPixels, pBits, lBmpSize );
LPBITMAPINFOHEADER lpbi; //Line 229
lpbi = (LPBITMAPINFOHEADER)hBitmap; //Line 230
return hBitmap;
}
HBITMAP ColorChange2Dlg::DIBToDDB( HANDLE hDIB, CDC& dc )
{
LPBITMAPINFOHEADER lpbi;
HBITMAP hbm;
CPalette pal;
CPalette* pOldPal;
//CClientDC dc(NULL);
if (hDIB == NULL)
return NULL;
lpbi = (LPBITMAPINFOHEADER)hDIB; //Line 243
int nColors = lpbi->biClrUsed ? lpbi->biClrUsed : 1 << lpbi->biBitCount; //Line 244
BITMAPINFO &bmInfo = *(LPBITMAPINFO)hDIB ;
LPVOID lpDIBBits;
if( bmInfo.bmiHeader.biBitCount > 8 )
lpDIBBits = (LPVOID)((LPDWORD)(bmInfo.bmiColors +
bmInfo.bmiHeader.biClrUsed) +
((bmInfo.bmiHeader.biCompression == BI_BITFIELDS) ? 3 : 0));
else
lpDIBBits = (LPVOID)(bmInfo.bmiColors + nColors);
// Create and select a logical palette if needed
if( nColors <= 256 && dc.GetDeviceCaps(RASTERCAPS) & RC_PALETTE)
{
UINT nSize = sizeof(LOGPALETTE) + (sizeof(PALETTEENTRY) * nColors);
LOGPALETTE *pLP = (LOGPALETTE *) new BYTE[nSize];
pLP->palVersion = 0x300;
pLP->palNumEntries = nColors;
for( int i=0; i < nColors; i++)
{
pLP->palPalEntry[i].peRed = bmInfo.bmiColors[i].rgbRed;
pLP->palPalEntry[i].peGreen = bmInfo.bmiColors[i].rgbGreen;
pLP->palPalEntry[i].peBlue = bmInfo.bmiColors[i].rgbBlue;
pLP->palPalEntry[i].peFlags = 0;
}
pal.CreatePalette( pLP );
delete[] pLP;
// Select and realize the palette
pOldPal = dc.SelectPalette( &pal, FALSE );
dc.RealizePalette();
}
hbm = CreateDIBitmap(dc.GetSafeHdc(), // handle to device context
(LPBITMAPINFOHEADER)lpbi, // pointer to bitmap info header
(LONG)CBM_INIT, // initialization flag
lpDIBBits, // pointer to initialization data
(LPBITMAPINFO)lpbi, // pointer to bitmap info
DIB_RGB_COLORS ); // color-data usage
if (pal.GetSafeHandle())
dc.SelectPalette(pOldPal,FALSE);
return hbm;
}
void ColorChange2Dlg::OnBnClickedButton1()
{
// TODO: Add your control notification handler code here
CClientDC dc(this);
COLORREF *pix = (COLORREF *)malloc(255*255*sizeof(COLORREF));
//int x = 1;
if(pix!=NULL){
for(int i=0;i<255;i++)
{
for(int j=0;j<255;j++)
{
pix[i*255+j] = RGB(i,j,0);
}
}
}
CDC tempDC;
tempDC.CreateCompatibleDC(&dc);
HBITMAP dib = CreateBitmapFromPixels(tempDC.m_hDC,255,255,8*sizeof(COLORREF),(BYTE*)pix);
HBITMAP finalMap = DIBToDDB(dib,tempDC);
HBITMAP oldMap = (HBITMAP)tempDC.SelectObject(finalMap);
dc.BitBlt(201,50,255,255,&tempDC,0,0,SRCCOPY);
tempDC.SelectObject(oldMap);
tempDC.DeleteDC();
}
To write compatible code, it's better not to access bits directly at all. You can use Gradient functions and GDI or GDI+ draw functions to do anything you want.
The code you have in mind pix[i*255+j] = RGB(i,j,0); is of a 32-bit image. Each pixel points to a color. It's not a palette image where each pixel points to an entry in the color table.
If display is 32 bit (most modern computers are, but check to make sure), you can do this with the following code
CBitmap m_bitmap;
void CMyWnd::make_bitmap()
{
if (m_bitmap.GetSafeHandle()) return;
int w = 255;
int h = 255;
int *pix = new int[w*h];
for (int i = 0; i < w; i++)
for (int j = 0; j < h; j++)
pix[i + j*w] = RGB(i, j, 0);
m_bitmap.CreateBitmap(w, h, 1, 32, pix);
delete[]pix;
}
And to draw the bitmap:
void CMyWnd::paint_bitmap(CDC &dc)
{
if (!m_bitmap.GetSafeHandle()) return;
CDC memdc;
memdc.CreateCompatibleDC(&dc);
HBITMAP oldbitmap = (HBITMAP)memdc.SelectObject(m_bitmap);
BITMAP bm;
m_bitmap.GetBitmap(&bm);
dc.BitBlt(0, 0, bm.bmWidth, bm.bmHeight, &memdc, 0, 0, SRCCOPY);
memdc.SelectObject(oldbitmap);
}
void CMyWnd::OnPaint()
{
__super::OnPaint();
CClientDC dc(this);
paint_bitmap(dc);
}
Edit: For historical reasons the RGB value are saved backward as BGR. Use this function instead:
void CMyWnd::make_bitmap()
{
if (m_bitmap.GetSafeHandle()) return;
int w = 256;
int h = 256;
BYTE *pix = new BYTE[4*w*h];
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
int p = (i + j*w) * 4;
pix[p + 0] = 0;//blue
pix[p + 1] = i;//green
pix[p + 2] = j;//red
pix[p + 3] = 0;//not used in GDI functions
}
}
m_bitmap.CreateBitmap(w, h, 1, 32, pix);
delete[]pix;
}

raw bitmap data to jpeg or png C++

I have bytearray where every three bytes describes 1 pixel (RGB). The task is to convert it to jpeg or png.
Actually, I am using Zint (open source lib for generating barcodes) that uses libpng to generate image file and save it to file system, but in Zintthe function png_plot() except generating image also save it on disk which is undesirable.
As result I think there two ways:
1. from bitmap bytearray to bmp -> jpeg / png (using some other lib)
2. writing hook or some similar to png_plot()
Can you give me some advices?
Thank you.
Upd: for #peacemaker
FILE *f;
zint_symbol *my_symbol;
my_symbol = ZBarcode_Create();
ZBarcode_Encode_and_Buffer(my_symbol, (unsigned char *)argv[1], 0, 0);
f = fopen("bitmap.bmp", "w");
fwrite(my_symbol->bitmap, sizeof(*(my_symbol->bitmap)), my_symbol->bitmap_height * my_symbol->bitmap_width, f);
ZBarcode_Delete(my_symbol);
fclose(f);
In order to convert between image formats, the easiest way would be using the class CImage shared by MFC and ATL and defined in the header file atlimage.h.
CImage image;
HRESULT res = image.Load("in.bmp");
image.Save("out.jpg");
image.Save("out.gif");
image.Save("out.png");
image.Save("out.tif");
If you have a RGB buffer and want to create a bitmap: just create and save a bitmap header into a file and add the RGB buffer to it.
To create the header you can use the BITMAPFILEHEADER, BITMAPINFOHEADER and RGBQUAD structures from GDI defined in the header WinGDI.h
Here is an example on how to fill the header data:
BITMAPINFOHEADER bmpInfoHdr;
bmpInfoHdr.biSize = sizeof(BITMAPINFOHEADER);
bmpInfoHdr.biHeight = nHeight;
bmpInfoHdr.biWidth = nWidthPadded;
bmpInfoHdr.biPlanes = 1;
bmpInfoHdr.biBitCount = bitsPerPixel;
bmpInfoHdr.biSizeImage = nHeight * nWidthPadded * nSPP;
bmpInfoHdr.biCompression = BI_RGB;
bmpInfoHdr.biClrImportant = 0;
bmpInfoHdr.biClrUsed = 0;
bmpInfoHdr.biXPelsPerMeter = 0;
bmpInfoHdr.biYPelsPerMeter = 0;
bmpFileHdr.bfType = BITMAP_FORMAT_BMP;
bmpFileHdr.bfSize = (DWORD) (sizeof(BITMAPFILEHEADER) + bmpInfoHdr.biSize +
sizeof(RGBQUAD)*numColors + bmpInfoHdr.biSizeImage);
bmpFileHdr.bfReserved1 = 0;
bmpFileHdr.bfReserved2 = 0;
bmpFileHdr.bfOffBits = (DWORD) (sizeof(BITMAPFILEHEADER) + bmpInfoHdr.biSize +
sizeof(RGBQUAD)*numColors);
Keep into account that the bitmaps are stored upside-down and that the width of the image must be aligned on a DWORD except for RLE-compressed bitmaps.(they must be multiple of 4 bytes, add a padding if necessary).
if ((nWidth%4) != 0)
nPadding = ((nWidth/4) + 1) * 4;
When saving your buffer, add the needed padding to each row...
Summarizing, these are the needed steps to create a bitmap file from a rgb buffer:
//1. create bmp header
//2. save header to file:
write(file, &bmpFileHdr, sizeof(BITMAPFILEHEADER));
write(file, &bmpInfoHdr, sizeof(BITMAPINFOHEADER));
write(file, &colorTable, numColors * sizeof(RGBQUAD));
//3. add rgb buffer to file:
for(int h=0; h<nHeight; h++) {
for(int w=0; w<nWidth; w++) {
//3.a) add row to file
//3.b) add padding for this row to file
}
}
I used the CImage Class from ATL.
int width=0, height=0;
char * val = "9788994774480";
zint_symbol *my_symbol;
my_symbol = ZBarcode_Create();
//ZBarcode_Encode_and_Buffer(my_symbol,(unsigned char *) val, 0, 0);
ZBarcode_Encode(my_symbol, (unsigned char *) val, 0);
ZBarcode_Buffer(my_symbol, 0);
height = my_symbol->bitmap_height;
width = my_symbol->bitmap_width;
char * imgBits = my_symbol->bitmap;
CImage img;
img.Create(width, height, 24 /* bpp */, 0 /* No alpha channel */);
int nPixel = 0;
for(int row = 0; row < height; row++)
{
for(int col = 0; col < width; col++)
{
BYTE r = (BYTE)imgBits[nPixel];
BYTE g = (BYTE)imgBits[nPixel+1];
BYTE b = (BYTE)imgBits[nPixel+2];
img.SetPixel(col, row , RGB(r, g, b));
nPixel += 3;
}
}
img.Save("CImage.bmp", Gdiplus::ImageFormatBMP);
ZBarcode_Delete(my_symbol);
is there anyway to do this other than using SetPixel? I am experiencing major performance issues with SetPixel and need an alternative method... I have tried using CreateDIBSection to no avail. The barcode displays slanted and is unusable. here is my code for that:
void *bits = (unsigned char*)(my_symbol->bitmap);
HBITMAP hBitmap = CreateDIBSection(pDC->GetSafeHdc(), &info, DIB_RGB_COLORS, (void **)&pDestData, NULL, 0);
memcpy(pDestData, my_symbol->bitmap, info.bmiHeader.biSizeImage);
img.Attach(hBitmap);
Another option that produces the same result is this:
BITMAPINFO info;
BITMAPINFOHEADER BitmapInfoHeader;
BitmapInfoHeader.biSize = sizeof(BITMAPINFOHEADER);
BitmapInfoHeader.biWidth = my_symbol->bitmap_width;
BitmapInfoHeader.biHeight = -(my_symbol->bitmap_height);
BitmapInfoHeader.biPlanes = 1;
BitmapInfoHeader.biBitCount = 24;
BitmapInfoHeader.biCompression = BI_RGB;
BitmapInfoHeader.biSizeImage = 0;
BitmapInfoHeader.biXPelsPerMeter = 0;
BitmapInfoHeader.biYPelsPerMeter = 0;
BitmapInfoHeader.biClrUsed = 0;
BitmapInfoHeader.biClrImportant = 0;
info.bmiHeader = BitmapInfoHeader;
HBITMAP hbmp = CreateDIBitmap(dc, &BitmapInfoHeader, CBM_INIT, (LPVOID *)my_symbol->bitmap, (LPBITMAPINFO)&info, DIB_RGB_COLORS);
img.Attach(hbmp);

Display a image in a MFC/C++ application using OpenCV

I would like to display in a MFC application, frames that I capture from an avi file with OpenCV (cvCaptureFromAVI function).
I'm new to MFC but feel like I'm close to making it work. But instead of the frames being displayed in the picture box they are displayed in a new window.
cvGetWindowName returns always a null value.
There is my code:
CWnd* hPic = 0;
hPic = GetDlgItem(IDC_STATICPIC1);
const char* szWindName = cvGetWindowName(hPic->GetSafeHwnd());
cvShowImage(szWindName, frame_copy);
So I found something to make it work after long researches.
The solution is to create the window and then insert it inside the picture box. I'm not sure it's good practice but I haven't found anything better for now.
cvNamedWindow("IDC_STATIC_OUTPUT", 0);
cvResizeWindow("IDC_STATIC_OUTPUT", 420, 240);
HWND hWnd = (HWND) cvGetWindowHandle("IDC_STATIC_OUTPUT");
HWND hParent = ::GetParent(hWnd);
::SetParent(hWnd, GetDlgItem(IDC_PIC1)->m_hWnd);
::ShowWindow(hParent, SW_HIDE);
cvShowImage("IDC_STATIC_OUTPUT", frame_copy);
In this case the picture box is called IDC_PIC1 and frame_copy is a OpenCV IplImage.
Hope this helps somebody.
Using the following code you can convert Mat to CImage and then display CImage everywhere you want:
int Mat2CImage(Mat *mat, CImage &img){
if(!mat || mat->empty())
return -1;
int nBPP = mat->channels()*8;
img.Create(mat->cols, mat->rows, nBPP);
if(nBPP == 8)
{
static RGBQUAD pRGB[256];
for (int i = 0; i < 256; i++)
pRGB[i].rgbBlue = pRGB[i].rgbGreen = pRGB[i].rgbRed = i;
img.SetColorTable(0, 256, pRGB);
}
uchar* psrc = mat->data;
uchar* pdst = (uchar*) img.GetBits();
int imgPitch = img.GetPitch();
for(int y = 0; y < mat->rows; y++)
{
memcpy(pdst, psrc, mat->cols*mat->channels());//mat->step is incorrect for those images created by roi (sub-images!)
psrc += mat->step;
pdst += imgPitch;
}
return 0;
}
NOTE: If you use the StretchDIBits() method with the BITMAPINFO approach, you MUST be aware that StretchDIBits() expects the raw OpenCV Mat::data pointer to have row lengths in even multiples of 4 bytes! If not, you'll get freaky shearing when you try to copy the data to the DC via StretchDIBits() - where the image is not only sheered along an angle, but the colors are all be trashed as well.
Here is my completely working edition of the code, which also supports maintaining image aspect ratio in the target control's rectangle. It can probably be made a bit faster, but it works for now:
void AdjustAspectImageSize( const Size& imageSize,
const Size& destSize,
Size& newSize )
{
double destAspectRatio = float( destSize.width ) / float( destSize.height );
double imageAspectRatio = float( imageSize.width ) / float( imageSize.height );
if ( imageAspectRatio > destAspectRatio )
{
// Margins on top/bottom
newSize.width = destSize.width;
newSize.height = int( imageSize.height *
( double( destSize.width ) / double( imageSize.width ) ) );
}
else
{
// Margins on left/right
newSize.height = destSize.height;
newSize.width = int( imageSize.width *
( double( destSize.height ) / double( imageSize.height ) ) );
}
}
void DrawPicToHDC( Mat cvImg,
UINT nDlgID,
bool bMaintainAspectRatio /* =true*/ )
{
// Get the HDC handle information from the ID passed
CDC* pDC = GetDlgItem(nDlgID)->GetDC();
HDC hDC = pDC->GetSafeHdc();
CRect rect;
GetDlgItem(nDlgID)->GetClientRect(rect);
Size winSize( rect.right, rect.bottom );
// Calculate the size of the image that
// will fit in the control rectangle.
Size origImageSize( cvImg.cols, cvImg.rows );
Size imageSize;
int offsetX;
int offsetY;
if ( ! bMaintainAspectRatio )
{
// Image should be the same size as the control's rectangle
imageSize = winSize;
}
else
{
Size newSize;
_AdjustAspectImageSize( origImageSize,
winSize,
imageSize );
}
offsetX = ( winSize.width - imageSize.width ) / 2;
offsetY = ( winSize.height - imageSize.height ) / 2;
// Resize the source to the size of the destination image if necessary
Mat cvImgTmp;
resize( cvImg,
cvImgTmp,
imageSize,
0,
0,
INTER_AREA );
// To handle our Mat object of this width, the source rows must
// be even multiples of a DWORD in length to be compatible with
// SetDIBits(). Calculate what the correct byte width of the
// row should be to be compatible with SetDIBits() below.
int stride = ( ( ( ( imageSize.width * 24 ) + 31 ) & ~31 ) >> 3 );
// Allocate a buffer for our DIB bits
uchar* pcDibBits = (uchar*) malloc( imageSize.height * stride );
if ( pcDibBits != NULL )
{
// Copy the raw pixel data over to our dibBits buffer.
// NOTE: Can setup cvImgTmp to add the padding to skip this.
for ( int row = 0; row < cvImgTmp.rows; ++row )
{
// Get pointers to the beginning of the row on both buffers
uchar* pcSrcPixel = cvImgTmp.ptr<uchar>(row);
uchar* pcDstPixel = pcDibBits + ( row * stride );
// We can just use memcpy
memcpy( pcDstPixel,
pcSrcPixel,
stride );
}
// Initialize the BITMAPINFO structure
BITMAPINFO bitInfo;
bitInfo.bmiHeader.biBitCount = 24;
bitInfo.bmiHeader.biWidth = cvImgTmp.cols;
bitInfo.bmiHeader.biHeight = -cvImgTmp.rows;
bitInfo.bmiHeader.biPlanes = 1;
bitInfo.bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
bitInfo.bmiHeader.biCompression = BI_RGB;
bitInfo.bmiHeader.biClrImportant = 0;
bitInfo.bmiHeader.biClrUsed = 0;
bitInfo.bmiHeader.biSizeImage = 0; //winSize.height * winSize.width * * 3;
bitInfo.bmiHeader.biXPelsPerMeter = 0;
bitInfo.bmiHeader.biYPelsPerMeter = 0;
// Add header and OPENCV image's data to the HDC
StretchDIBits( hDC,
offsetX,
offsetY,
cvImgTmp.cols,
cvImgTmp.rows,
0,
0,
cvImgTmp.cols,
cvImgTmp.rows,
pcDibBits,
& bitInfo,
DIB_RGB_COLORS,
SRCCOPY );
free(pcDibBits);
}
ReleaseDC(pDC);
}
int DrawImageToHDC(IplImage* img, HDC hdc, int xDest, int yDest, UINT iUsage, DWORD rop)
char m_chBmpBuf[2048];
BITMAPINFO *m_pBmpInfo = 0;
m_pBmpInfo = (BITMAPINFO*)m_chBmpBuf;
m_pBmpInfo->bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
m_pBmpInfo->bmiHeader.biWidth = img->width;
m_pBmpInfo->bmiHeader.biHeight = -img->height;
m_pBmpInfo->bmiHeader.biBitCount = 24;
m_pBmpInfo->bmiHeader.biPlanes = 1;
m_pBmpInfo->bmiHeader.biCompression = BI_RGB;
m_pBmpInfo->bmiHeader.biSizeImage = 0;
m_pBmpInfo->bmiHeader.biXPelsPerMeter = 0;
m_pBmpInfo->bmiHeader.biYPelsPerMeter = 0;
m_pBmpInfo->bmiHeader.biClrUsed = 0;
m_pBmpInfo->bmiHeader.biClrImportant = 0;
return StretchDIBits(hdc, xDest, yDest, img->width, img->height, 0, 0,
img->width, img->height, img->imageData, m_pBmpInfo, DIB_RGB_COLORS, SRCCOPY);
Usage: DrawImageToHDC(img, pDC->m_hDC, Area.left, Area.top, DIB_RGB_COLORS, SRCCOPY);

How to compress image on linux?

I am writing function to compress image using GDI+ on windows, and it's working well,
void ImageProcessorImpl::compressImpl(const std::string& path, int size, UInt8 quality)
{
HBITMAP hbmReturn = NULL;
Bitmap* bmPhoto = NULL;
std::wstring upath;
UnicodeConverter::toUTF16(path, upath);
// make source file close automatically, Bitmap detructor will be called
{
Bitmap image(upath.c_str());
int srcWidth = image.GetWidth();
int srcHeight = image.GetHeight();
float percent = 0;
int destX = 0, destY = 0;
if (srcWidth > srcHeight)
{
percent = ((float)size/(float)srcWidth);
destX = (int)((size - (srcWidth * percent))/2);
}
else
{
percent = ((float)size/(float)srcHeight);
destY = (int)((size - (srcHeight * percent))/2);
}
if (percent >= 1.0f)
return; // skip compress
int destWidth = (int)(srcWidth * percent);
int destHeight = (int)(srcHeight * percent);
bmPhoto = new Bitmap(destWidth, destHeight, PixelFormat24bppRGB);
bmPhoto->SetResolution(image.GetHorizontalResolution(), image.GetVerticalResolution());
Graphics *grPhoto = Graphics::FromImage(bmPhoto);
Color colorW(255, 255, 255, 255);
grPhoto->Clear(colorW);
grPhoto->SetInterpolationMode(InterpolationModeHighQualityBicubic);
grPhoto->DrawImage(&image, Rect(destX, destY, destWidth, destHeight));
bmPhoto->GetHBITMAP(colorW, &hbmReturn);
delete grPhoto;
} // end source image file, Bitmap image(upath.c_str());
// find appropriate encoder, jpeg
CLSID encoderClsid;
getEncoderClsid(L"image/jpeg", &encoderClsid);
// set output quality for jpeg alone
EncoderParameters encoderParameters;
setEncoderQuality(&encoderParameters, &quality);
// output to image file with desired quality
bmPhoto->Save(upath.c_str(), &encoderClsid, &encoderParameters);
// release resources
delete bmPhoto;
DeleteObject(hbmReturn);
}
int ImageProcessorImpl::getEncoderClsid(const WCHAR* format, void* clsid)
{
UINT num = 0; // number of image encoders
UINT size = 0; // size of the image encoder array in bytes
ImageCodecInfo* pImageCodecInfo = NULL;
GetImageEncodersSize(&num, &size);
if(size == 0)
return -1; // Failure
pImageCodecInfo = (ImageCodecInfo*)(malloc(size));
if(pImageCodecInfo == NULL)
return -1; // Failure
GetImageEncoders(num, size, pImageCodecInfo);
for (UINT j = 0; j < num; ++j)
{
if (wcscmp(pImageCodecInfo[j].MimeType, format) == 0)
{
*(CLSID*)clsid = pImageCodecInfo[j].Clsid;
free(pImageCodecInfo);
return j; //Success
}
}
free(pImageCodecInfo);
return -1; // Failure
}
void ImageProcessorImpl::setEncoderQuality(void* params, UInt8* quality)
{
EncoderParameters* encoderParams = (EncoderParameters*)params;
encoderParams->Count = 1;
encoderParams->Parameter[0].Guid = EncoderQuality;
encoderParams->Parameter[0].Type = EncoderParameterValueTypeLong;
encoderParams->Parameter[0].NumberOfValues = 1;
encoderParams->Parameter[0].Value = quality;
}
But, I want to have this function on linux, I don't know what lib I can use to implement such function on linux, who can help me?
Thnx
You can use the ImageMagick library or netpbm library. Netpbm also has command line tools to manipulate images.
OpenCV provides an easy to use interface for all types of image processing as well as simple things such as image compression. It may be a bit of an overkill for image compression only, but I suppose it may helpful if you plan on doing other things with these images.

Loading an image from resource and converting to bitmap in memory

I've searched around using google but I'm completely confused on how to load an image (PNG in my case) from resource and then converting it to a bitmap in memory for use in my splash screen. I've read about GDI+ and libpng but I don't really know how to do what I want. Could anyone help?
GDI+ supports PNG directly. See here and here.
EDIT: The GDI+ documentation offers some advice for how to use GDI+ in a DLL. In your case, the best solution is probably to define initialisation and teardown functions that the client code is required to call.
I ended up using PicoPNG to convert the PNG to a two dimensional vector which I then manually contructed a bitmap from. My final code looked like this:
HBITMAP LoadPNGasBMP(const HMODULE hModule, const LPCTSTR lpPNGName)
{
/* First we need to get an pointer to the PNG */
HRSRC found = FindResource(hModule, lpPNGName, "PNG");
unsigned int size = SizeofResource(hModule, found);
HGLOBAL loaded = LoadResource(hModule, found);
void* resource_data = LockResource(loaded);
/* Now we decode the PNG */
vector<unsigned char> raw;
unsigned long width, height;
int err = decodePNG(raw, width, height, (const unsigned char*)resource_data, size);
if (err != 0)
{
log_debug("Error while decoding png splash: %d", err);
return NULL;
}
/* Create the bitmap */
BITMAPV5HEADER bmpheader = {0};
bmpheader.bV5Size = sizeof(BITMAPV5HEADER);
bmpheader.bV5Width = width;
bmpheader.bV5Height = height;
bmpheader.bV5Planes = 1;
bmpheader.bV5BitCount = 32;
bmpheader.bV5Compression = BI_BITFIELDS;
bmpheader.bV5SizeImage = width*height*4;
bmpheader.bV5RedMask = 0x00FF0000;
bmpheader.bV5GreenMask = 0x0000FF00;
bmpheader.bV5BlueMask = 0x000000FF;
bmpheader.bV5AlphaMask = 0xFF000000;
bmpheader.bV5CSType = LCS_WINDOWS_COLOR_SPACE;
bmpheader.bV5Intent = LCS_GM_BUSINESS;
void* converted = NULL;
HDC screen = GetDC(NULL);
HBITMAP result = CreateDIBSection(screen, reinterpret_cast<BITMAPINFO*>(&bmpheader), DIB_RGB_COLORS, &converted, NULL, 0);
ReleaseDC(NULL, screen);
/* Copy the decoded image into the bitmap in the correct order */
for (unsigned int y1 = height - 1, y2 = 0; y2 < height; y1--, y2++)
for (unsigned int x = 0; x < width; x++)
{
*((char*)converted+0+4*x+4*width*y2) = raw[2+4*x+4*width*y1]; // Blue
*((char*)converted+1+4*x+4*width*y2) = raw[1+4*x+4*width*y1]; // Green
*((char*)converted+2+4*x+4*width*y2) = raw[0+4*x+4*width*y1]; // Red
*((char*)converted+3+4*x+4*width*y2) = raw[3+4*x+4*width*y1]; // Alpha
}
/* Done! */
return result;
}