Blurring a .BMP with box BLUR in C - c

Question

To resume. the code I ve written below doesn t get me a blur image. But a black IMAGE instead ( except for the pixel edges).
For testing i ve just tried the interior PIXELS, without edges RGBTRIPLE pixels because I wanted to see if I was making well so far.
Could anybody tell me whats wrong with my code, please?
is muy BUFFER RGB newImage correct done?? if image[height][width].rgbtRed it means that newImage[height][width].rgbtRed will exist?? thanks in advance
// Blur image
void blur(int height, int width, RGBTRIPLE image[height][width])
{
//create a buffer to allocate new pixels.
RGBTRIPLE newImage[height][width];
// iterate through each row
for (int i = 0 ; i < height ; i++)
{
// iterate through each column passing through every RGBTRIPLE
for ( int j = 0; j < width ; j++)
{
int redSum = 0;
int greenSum = 0;
int blueSum = 0;
//counter for pixels being add in every avg.
//int count = 0;
// take the average amount of RGB of every pixel in a range of 3x3 pixels
// avg for left top corner
/*if ( r == 0 && l == 0) {
}
// avg left bottom corner
else if ( r == height - 1 && l == 0 ){
}
// avg right top corner
else if ( r == 0 && l == width - 1){
}
// avg left bottom corner
else if ( r == height - 1 && l == width - 1){
}
// avg top side
else if ( r == 0 && l > 0 && l != width - 1) {
}
// avg right side
else if ( l == width - 1 && r > 0 && r != height - 1) {
}
// avg bottom side
else if ( r == height - 1 && l > 0 && l != width - 1) {
}
// avg left side
else if ( l == 0 && r > 0 && r != height - 1){
}*/
for ( int iy = -1 ; iy >= 1 ; iy++)
{
for ( int jx = -1 ; jx >= 1 ; jx++)
{
redSum = redSum + image[i + iy][j + jx].rgbtRed;
greenSum = greenSum + image[i + iy][j + jx].rgbtGreen;
blueSum = blueSum + image[i + iy][j + jx].rgbtBlue;
//count++;
}
}
int avgRed = redSum / 9;
int avgGreen = greenSum / 9;
int avgBlue = blueSum / 9;
newImage[i][j].rgbtRed = avgRed;
newImage[i][j].rgbtGreen = avgGreen;
newImage[i][j].rgbtBlue = avgBlue;
}
}
for ( int y = 1 ; y < height - 1; y++)
{
for ( int x = 1; x < width - 1; x++)
{
image[y][x].rgbtRed = newImage[y][x].rgbtRed;
image[y][x].rgbtGreen = newImage[y][x].rgbtGreen;
image[y][x].rgbtBlue = newImage[y][x].rgbtBlue;
}
}
return;
}

Answer 1

A major mistake not already mentioned in a comment is that redSum, greenSum and blueSum are initialized to 0 only once, while they must be set to zero before each pixel is processed, so better move the definitions right before the for ( int py = 0 ; py >= 2 ; py++) loop.

Answer 2

It's hardly believable how the many eyes (including mine) who looked at the code could overlook the main, simple error - the inner loops (in their original as well as in the edited version) are never executed:
for ( int iy = -1 ; iy >= 1 ; iy++)
{
for ( int jx = -1 ; jx >= 1 ; jx++)
Of course, the iy >= 1 condition is false from the start; you surely meant iy <= 1 and jx <= 1.