C ++：读取位图图像时出现问题

Question

I'm currently writing a function that will manipulate a 512x512 .BMP image and I'm struggling with reading in the pixel data. I'm including fstream in this to deal with reading the image data.

My function should work explicitly with 8-bit Grayscale .BMP images, so I really shouldn't need to worry about the file Headers. You can see in my code I just read the entire header into an array for later usage.

Here is my Constructor for my Image Class:

// Constructor
Image::Image(const char* inFileName){
    // First we will initialize our Image Data Arrays:
    ImageData = new int* [SIZE];
    complexImageData = new complex<double>* [SIZE];

    for(int i=0; i<SIZE; ++i){
        ImageData[i] = new int [SIZE];
        complexImageData[i] = new complex<double> [SIZE];
    }

    // Now we will initialize our I/O Files.
    outFile = new ofstream;

    inFile = new ifstream;

    // And now we will fill our data arrays with the input file data.

    inFile->open(inFileName,ios::in | ios::binary); // Reads file as binary
    inFile->seekg(0,ios::beg); // sets filter position to beginning
    inFile->read(header,HEADERSIZE); // Reads header

    // Now to read in the image data:
    for(int i=0; i<SIZE; ++i){
        inFile->read(reinterpret_cast<char*>(ImageData[i]),SIZE);
        for(int j=0; j<SIZE; ++j){
            cout << ImageData[i][j] << ' ';
        }
        cout << endl;
    }

    inFile->close();
}

As you can see, it prints out the data as it is read in. The output of this constructor is strange to me. Here's the first 512 entries read into the ImageData array:

236854814 237117986 505286158 236854814 505290270 236854814 572662306 237112866 571346462 505290254 505288222 371072534 236328478 504763918 370546198 152442390 503911958 236330518 370546198 152439062 369690902 504763926 236852758 571347470 572662306 571351570 235802142 504237582 505290254 236330526 504241694 235802126 505290262 370546206 370548246 504765974 371070494 505288214 371072534 369694230 235806222 236854814 237117986 235802146 370548254 369690902 504763926 371072534 370546198 152442390 504760598 504763926 237117986 304226850 235802146 235802126 235807246 504241678 504761870 504237590 371072542 370548246 504763926 236854806 235802142 235802126 235802126 504242702 235802126 303178258 505286178 571351566 572662286 572658210 236065314 303178254 571609618 571351586 571613730 571351586 237112866 571351566 303702562 572658210 237112846 304222754 303700506 304226850 168434194 471079956 169088010 169088020 437914122 437918234 169085466 436865546 169482778 336204810 168430100 336857610 336862228 336204820 471077898 471866380 471079964 169090076 203168778 169088012 471077914 336862236 168430090 168434186 168434202 437918234 169482778 168430106 168430106 169482762 336206346 169085450 336206356 135796768 404228120 270014488 303698458 303174162 303174162 404752922 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 

As you can see, approximately the second half of this data is all zeros. Another line is even stranger:

236852766 571346446 236850702 371072542 505291278 371070494 504763934 236330518 370548238 503911966 236852758 371072542 504763913 504237582 571350542 235802126 504242722 503911958 152966678 370542870 369694230 369690905 236330505 504237590 370548246 504763926 370546206 370546198 370546198 235806230 371073550 504763926 370548254 572658206 235807246 571346446 504237582 236854798 505553422 370548254 152442390 236852758 572662306 504765986 370546198 369694217 504763926 370548246 504241686 371072526 235804190 504765974 370542870 151587081 152635673 504763913 369690902 370546206 152443145 505290262 236852766 371072534 505284886 370546206 572397078 371072526 504237598 370548254 371070486 371073566 152439049 370542870 152442377 504763926 571350558 505286174 504765982 236854814 235806238 235802126 235807246 572399134 571351566 571613730 571611666 303698466 437916186 169478682 168430090 169089546 336862236 336860180 336204810 336857620 336202260 168432660 168430090 168430106 471075338 203431964 470551564 336202250 336860180 336860180 336862228 336204820 168430100 436865562 337384458 436868106 168430090 168430090 336204820 436865546 168430090 168432650 269487114 135796760 202905624 437918226 437918234 403705868 404232208 269490200 337252884 571613722 304222738 269228562 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 181497 53543152 0 53543152 0 57016416 0 57016416 0 57135104 0 925696 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

You can see that it also has the same zeros, but amongst those zeros are some unexpected non-zero entries. This is certainly not expected from the input image .

Any help would be greatly appreciated! I'm sorry if this is a duplicate but I tried so very hard to find an answer to my problem before posting this!

EDIT:

I noticed immediately that the numbers being printed (236854814, 237117986 etc) were not in the 8-bit range. I find this so confusing since I'm reading the data reinterpreted as a char, which is 8-bit on my system. Would it not be the case that the data being read into the integer array should be 8-bit? I even tried making my ImageData array of type __int8**, so that it could only store 8 bits per entry, but It still didn't work.

Answer 1

Your image is 8bpp grayscale but you read it into block int [SIZE] (supplying wrong buffer size in the process) potentially storing 4 pixel values in each int . If image is 8bpp you could've just used uint8_t buffer. You should also perform header validation (because right now your may read at wrong position) and check for actual amount of bytes read. Depending on image type / size you may also need to check padding for each line.