Loading an image as a matrix

I want to load an image as a matrix, and perform sparse matrix arithmetic.
The relevant libraries I found for this are:

Image operations (load, resize):
GitHub - nothings/stb: stb single-file public domain libraries for C/C++

Sparse matrix arithmetic:
GSL - GNU Scientific Library - GNU Project - Free Software Foundation

Are there other libraries that can save the pain of adding a translation between these two libraries formats (the loaded image, and the matrix type)?

thanks

I ended up doing the below, I would be happy to get any tips for writing it in a more clean\ "right" way, I've been writing python for a few years and the transition is confusing for me.

Code:

---

#define STB_IMAGE_IMPLEMENTATION
#include <stdio.h>
#include <stdint.h>
#include <stddef.h>
#include <stdbool.h>
#include <gsl/gsl_sf.h>
#include "stb_image.h"
#define STB_IMAGE_WRITE_IMPLEMENTATION
#include "stb_image_write.h"

void slice_image(unsigned char *from_image, unsigned char *to_image, int width,
                int height, int original_channels, int target_channel);
void image2matrix(gsl_matrix * matrix, unsigned char *image, int width,
                  int height);

int main()
{
  printf("Starting\n");

  // Load Image
  int width, height, channels;

  const char path[] =
      "/mnt/c/Users/alexf/Desktop/Projects/3D_Line_Dithering/test_images/sunflowers.jpg";
  unsigned char *image = stbi_load(path,
                                  &width,
                                  &height,
                                  &channels,
                                  STBI_rgb);

  if (image == NULL) {
    printf("Error in loading the image\n");
    exit(1);
  } else {
    printf("Loaded Image\n");
  }


  printf("Image Params:\n");
  printf("Height: %d ; Width: %d ; Channels: %d\n", width, height, channels);
  // Slice Image
  unsigned char *image_r = malloc(sizeof(int) * height * width);
  if (image_r == NULL) {
    printf("Error allocating space for the image slice\n");
    exit(1);
  }
  printf("Slice Image:\n");

  slice_image(image, image_r, width, height, channels, 0);

  printf("Allocate Matrix:\n");
  gsl_matrix *matrix = gsl_matrix_alloc(height, width);
  if (matrix == NULL) {
    printf("Error allocating space for the image slice\n");
    exit(1);
  }
  printf("Image to Matrix:\n");
  image2matrix(matrix, image_r, width, height);

  // FREE ALL
  printf("Free All:\n");
  gsl_matrix_free(matrix);
  free(image_r);
  stbi_image_free(image);
  printf("Finished\n");

  return 0;
}

void slice_image(unsigned char *from_image, unsigned char *to_image, int width,
                int height, int original_channels, int target_channel)
{
  unsigned char *p;
  for (int j = 0; j < height; j++) {
    for (int k = 0; k < width; k++) {
      p = &from_image[original_channels * (j * width + k)];
      to_image[(j * width + k)] = p[target_channel];

      printf("%u\n", p[target_channel]);
    }
    printf("\n");
  }
};

int print_matrix(FILE * f, const gsl_matrix * m)
{
  int status, n = 0;

  for (size_t i = 0; i < m->size1; i++) {
    for (size_t j = 0; j < m->size2; j++) {
      if ((status = fprintf(f, "%g ", gsl_matrix_get(m, i, j))) < 0)
        return -1;
      n += status;
    }

    if ((status = fprintf(f, "\n")) < 0)
      return -1;
    n += status;
  }

  return n;

}

---

You didn't post image2matrix.

> printf("Height: %d ; Width: %d ; Channels: %d\n", width, height, channels);
What do you get printed for channels?

3 maybe, one for each of R,G,B

> unsigned char *image_r = malloc(sizeof(int) * height * width);
I would think you would need channels, not sizeof(int) here.

> p = &from_image[original_channels * (j * width + k)];
original_channels is your limit value.
If you're trying to extract say channel x out of a y-channel image, you need a different expression.
It might be fine for your special case of channel 0 - I don't know.

my bad,

Code:

---

void image2matrix(gsl_matrix * matrix, unsigned char *image, int width,
                  int height)
{
  for (int j = 0; j < height; j++) {
    for (int k = 0; k < width; k++) {
      matrix->data[j * matrix->tda + k] = image[j * width + k];
    }
  }

};

---

Please pay attention to posting code.
Don't include all the HTML nonsense from your IDE.
Use copy-as-text and/or paste-as-text.

copy that, I will remove HTML nonsense next time

"
What do you get printed for channels?
3 maybe, one for each of R,G,B
"
Yea I get three

"
> unsigned char *image_r = malloc(sizeof(int) * height * width);
I would think you would need channels, not sizeof(int) here.
"
I want to take a slice, meaning only one channel from the original image.

"
> p = &from_image[original_channels * (j * width + k)];
original_channels is your limit value.
If you're trying to extract say channel x out of a y-channel image, you need a different expression.
It might be fine for your special case of channel 0 - I don't know.
"
The format that the image is loaded is an array, where the 3 first values are the RGB values of the first pixel.
because I'm slicing one channel, I want to jump over the other channel values while iterating