Trouble with GetDIBits()

Hello guys, I'm new to C++ and I couldn't figure out the proper way to store pixel data of a device context into an array using GetDIBits(). Currently I am trying to get the RGB values of a single pixel of a bitmap file:

Code:

---

#include <windows.h>
#include <iostream>
using namespace std;
int main() {HDC MemDC=CreateCompatibleDC(NULL);
        SelectObject(MemDC,(HBITMAP)LoadImage(NULL,(LPCTSTR)"F:\\gBit.bmp",IMAGE_BITMAP,1366,768,LR_LOADFROMFILE));
        HBITMAP hBit=(HBITMAP)LoadImage(NULL,(LPCTSTR)"F:\\gBit.bmp",IMAGE_BITMAP,1366,768,LR_LOADFROMFILE);
        BITMAPINFO bmi;
        BYTE p[3];


        bmi.bmiHeader.biSize=sizeof(BITMAPINFOHEADER);
        bmi.bmiHeader.biWidth=1;
        bmi.bmiHeader.biHeight=-1;
        bmi.bmiHeader.biPlanes=1;
        bmi.bmiHeader.biBitCount=24;
        bmi.bmiHeader.biCompression=BI_RGB;
        bmi.bmiHeader.biSizeImage=0;
        bmi.bmiHeader.biXPelsPerMeter=0;
        bmi.bmiHeader.biYPelsPerMeter=0;
        bmi.bmiHeader.biClrUsed=0;
        bmi.bmiHeader.biClrImportant=0;


        GetDIBits(MemDC,hBit,0,0,p,&bmi,DIB_RGB_COLORS);
        cout<<p[0]<<endl<<p[1]<<endl<<p[2];
        ReleaseDC(NULL,MemDC); DeleteDC(MemDC);
        DeleteObject(hBit); while (1) {}
}

---

gBit.bmp, the bitmap file selected into MemDC, is an entirely white 24-bit bitmap image, so I thought cout should display 255, but for some reason it displays some weird characters. Most probably p is not initialized since the same values are returned when I remove the line containing GetDIBits(). Am I using this function properly? What could I possibly be doing wrong here?

What does the GetDIBits return? And did you red the GetDIBits documentation?

Well, in my case the return value is 1, so I guess the pixel is successfully scanned. But whether or not I call GetDIBits(), the same characters are stored in the array. I did read the documentation of the function but I didn't find anything that could cause the error. What do you expect me to know from the documentation?

BYTE is a char.
cout << (BYTE)32;
outputs a character space (ascii value 32), not the value 32.

to display char's (BYTE's) as numeric values, typecast to int.

First, in your other thread COLORREF is 4-bytes, as it stores 32-bit values RGBA (RGB and Alpha for transparency). Both your examples declare 24-bit colour values, so you would need only 3-bytes per pixel. It's ok to use 4-bytes like COLORREF, but you would ignore the 4th byte each time as it would be 0. Though, that is a waste of memory.

In this example you are loading an image the size 1366x768. So you would need (1366 * 768 * 3) bytes to store it 3147264. Fortunately the width (768) is a multiple of 4, otherwise you would need to add some padding.

Code:

---

BYTE p[3];

---

Should be more like

Code:

---

BYTE p[1366 * 768 * 3]  // Can't actually be done

---

This would need to be done with new in the heap otherwise it will crash.

Code:

---

bmi.bmiHeader.biSizeImage=0;

---

This also needs to be the value of 1366 * 768 * 3.

Also, the fourth variable needs to be the image height (scanlines). So instead of:

Code:

---

GetDIBits(MemDC,hBit,0,0,p,&bmi,DIB_RGB_COLORS); // Incorrect

---

It should be more like:

Code:

---

GetDIBits(MemDC,hBit,0,768,p,&bmi,DIB_RGB_COLORS);

---