|
I have a program that analyses bitmap and at the moment i am using getPIxel and setPixel, as you will know if you have used them they are very slow. Does any one know of an alternative that will increase the speed?
|
|
|
|
|
Use a DIB and access the bits directly. This will be especially fast when you need to get a lot of pixels since you won't need to recalculate the pixel's offset and convert to a DWORD each time.
|
|
|
|
|
Im not too sure what you mean by "recalculate the pixel's offset and convert to a DWORD each time" because i dont think im doing that now. This is my code it compares each pixel in the 2 dc's :
<br />
for (int w = 0; w < width; w++)<br />
{<br />
for (int h = 0; h < height; h++)<br />
{<br />
if (GetPixel(hdcOld, w, h) != GetPixel(hdcNew, w, h))<br />
{<br />
SetPixel(dcNew, w, h, RGB(255, 255, 255));<br />
pixelCount++;<br />
}<br />
else<br />
{<br />
SetPixel(dcNew, w, h, RGB(0, 0, 0));<br />
}<br />
}<br />
}<br />
|
|
|
|
|
Maybe you are not doing it yourself, but what do you thing GetPixel() is doing internally? Each time you call it, it must first read the header, get the bits, calculate the size of a pixel, calculate the the size of a row, calculate which row, calculate how far into that row, get the pixel, convert it to a COLORREF then give it back to you. ALL FOR ONE PIXEL! This is for every single call.
If you use a dib, you can do all that calculating yourself and do it only once.
|
|
|
|
|
So how would i do it using dib, im pretty new to c++?
|
|
|
|
|
have a search of the GDI and bitmap articles, there are quite a few DIB wrappers available.
|
|
|
|
|
Ok so i have put my bitmaps in DIB's and placed them in seperate dc's but what function do i use now instaed of getPixel()?
|
|
|
|
|
There is no specific function. If you have downloaded and integrated one of the DIB wrapper classes, derive a new class from it and add a new member.
Lets say you are scanning a web-cam image and you want to get every 10th pixel. In your new function, get the images-bits from the base class which is just one big array of pixels. Calculate the bitmaps paramaters, then in a loop calculate which pixels you want ( y * bitmapwidth_inbytes + x ) and read them from the array.
Just remember, with a DIB, the pixels are not always a 32 value like you get back from GetPixel. They could be:
1 Bits ( 0 == black, 1 == white )
4 Bits ( 1 byte with two indexes into the color table )
8 Bits ( an array of bytes each representing an index into the color table )
16 Bits ( an array of WORD each representing an index into the color table )
24 Bits ( an array of DWORD look on MSDN for RGBQUAD )
32 Bits ( an array of DWORD look on MSDN for RGBQUAD )
64 Bits ( only windows vista supports these I think )
By far, the easiest to deal with are 24 and 32 bit images, all of the others use a color table. If you need the actual COLORREF value, then you must calculate it from whatever value you get from the array.
|
|
|
|
|
Hi !!
I have one main method and now i want to call this main method in another method..
for example...
void NORMAL()
{
}
void MAJOR ()
{
Here i want to call void NORMAL() method
}
In void MAJOR () method, I want to call void NORMAL ()..
how can i call it ??
Both are in same .cpp file..n i have no class.
Thanking you..
Shah
|
|
|
|
|
what about :
void MAJOR()
{
NORMAL();
}
This is rrrreeeaaallllyyyy basic stuff, or I did not understood the question.
|
|
|
|
|
oppsss..Really really sorry.. i have Problem in my code..
|
|
|
|
|
and what might be the problem ?
|
|
|
|
|
void MAJOR ()
{
NORMAL();
}
Come on, this is really beginner stuff! Aren't you calling your MAJOR() function from somewhere else? Where's it getting called from?
|
|
|
|
|
oppsss..Really sorry..Problem in my code..
|
|
|
|
|
Is this a specifc question or simple question? 
|
|
|
|
|
Heee heee..This is not Specific nor Simple.. but its stupid question..Actually i tried same code but it was not working.. problem was in my code but in method.
|
|
|
|
|
Not problem,one thing is important that you could find your problem Its good;)
|
|
|
|
|
I'm getting a bad value when doing a drag-right-click ( doing a selection with the right mouse button) in a CTreeCtrl.
This is the callback for the NM_RCLICK notification.
void CAboutDlg::OnTreeRclick( NMHDR* pNMHDR, LRESULT* pResult )
{
NM_TREEVIEW* pNMTreeView = (NM_TREEVIEW*)pNMHDR;
HTREEITEM hSelItem = pNMTreeView->itemNew.hItem;
if ( hSelItem == NULL )
{
return;
}
}
Sometimes, the hSelItem tree item is not a valid item ( for example : 0x0000008d ) which is not valid.
I alse tried this where ptDrag is supposed to be the "structure that contains the client coordinates of the mouse at the time the event occurred that caused the notification message to be sent. (from MSDN) "
CPoint point( pNMTreeView->ptDrag );
UINT uHF;
HTREEITEM hItem;
hItem = m_Tree.HitTest( point, &uHF );
But the point is mostly invalid.
I also use which works, but seems not to be the right way to do it.
UINT uHF;
HTREEITEM hItem;
CPoint point;
GetCursorPos( &point );
m_Tree.ScreenToClient( &point );
hItem = m_Tree.HitTest( point, &uHF );
Thanks for you help and/or suggestions.
Max.
|
|
|
|
|
Some of the tree notifications don't set the itemNew element in the NM_TREEVIEW structure. As a result, it can be filled with random garbage.
Another thing to remember is that the right click notification happens whenever you right click on the tree control. If you're doing that over an area that doesn't include an item, you'll need to detect that.
IIRC, the last time I did this kind of stuff with a tree control, I ended up with a similar approach.
Sorry I'm not giving you more definite answers here. I'm working from memory, and don't have access to the code where I was dealing with this stuff.
Software Zen: delete this;
|
|
|
|
|
I tried to export a static member in one of my classes.
All other methods and operators of the same class can be successfully exported and then linked by another program excepte all static data member(declared as public static const in .hpp and implemented in .cpp, according to msdn guild on the dllexport part, it should work). I wonder why?
I am not the 1st one who's suffering from this, I did a simple research with google, some other guys also had this kind of problem and posted it on the web but never got a response.
Hopefully some one here could help me out, thanks.
|
|
|
|
|
Btw, by "unable to export" I mean the program can be compiled but get a link error generated.
|
|
|
|
|
Hello,
For a CDBVariant MFC class, how do I assign value to an object of type CDBVariant?
If
CDBVariant V;
V.m_dwtype = DBVT_SHORT;
V.m_iVal = 5;
Is this the right way to do ?
Priya
|
|
|
|
|
Does anyone know what happens when you have a socket bound to INADDR_ANY and then have a second socket (both TCP or both UDP sockets) bound to a specific address?
I am trying to add something to an existing program that has a TCP socket bound to INADDR_ANY. I need to add a TCP socket that listens to 127.5.0.1 (within the same executable). Will messages sent to 127.5.0.1 be handled by my new socket or will everything go through the existing socket bound to INADDR_ANY?
I guess I can check the address on the existing socket and handle the messages based on that but I'd still like to understand how this works.
Thanks,
Dave
|
|
|
|
|
INADDR_ANY indicates that the caller to bind() doesn't care what address the socket is bound
to. Typically this is used on client ends of TCP connections to let the system select an
unused port.
If you want to listen on a specific address/port then you have to bind to that specific
address.
Dave_ wrote: I need to add a TCP socket that listens to
What do you mean by "listens to"? This is very different on TCP and UDP.
|
|
|
|
|
When I use the listen() call, which socket (or will it be both?) will get the attempt to connect from the client? If I try to recv(), will the socket that is bound to the specific address get the data or will there be a race condition (first socket to read gets the data)?
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
