Vector Image Enlarge

I'm trying to get better at C++ so I have been doing some of the challenges at topcoder.

This was the problem

Given an image as a vector <string>, scale it up by factor. That is, if the original image is M x N, the scaled image should be M*factor x N*factor. Each character in the input vector <string> should be represented by factor x factor of the same character in the output (see examples).

example

{"abc",
"def",
"ghi"}

3
Returns:
{ "aaabbbccc",
"aaabbbccc",
"aaabbbccc",
"dddeeefff",
"dddeeefff",
"dddeeefff",
"ggghhhiii",
"ggghhhiii",
"ggghhhiii" }

Here is my solution but 4 loops seems too much. Can anybody show me what I could do better? Thanks

Code:

#include <vector>
#include <iostream>
using namespace std;

class ImageEnlarger{
public:
vector <string> enlarge(vector <string> image, int factor){
  string tempImage="", tempReturn="";  
  vector <string> returnImage;  
 
  for(int i=0;i<image.size();i++){  
    tempImage=image[i];   
    for(int j=0;j<factor;j++){ 
      for(int k=0;k<tempImage.length();k++){    
        for(int z=0;z<factor;z++){      
          tempReturn+=tempImage[k];
        }//end for
      }//end for       
   
     returnImage.push_back(tempReturn);  
     tempReturn="";
     }//end for
  }//end for
    
  return returnImage;
}//end enlarge
};//end class

Here's my take on it, although I didn't bother to make it a function or put it in a class or anything. I'm sure the scaling algorithm could be better still...and I think I miss-used the size_t type, but I'm still in the process of learning myself

EDIT: I decided to put it in function form like you did, so I'm replacing the old code with the new code. I'm just not sure it's a good idea to be passing and returning a vector this way...I believe I read that it's more efficient to pass/return an iterator instead.

Code:

#include <iostream>
#include <string>
#include <vector>

using namespace std;

vector<string> Enlarge(const vector<string> &inImage, int scaleFactor)
{
	vector<string> outImage;

	size_t inWidth = inImage[0].length();  // first string determines width of image
	size_t inHeight = inImage.size();
	size_t outWidth = inWidth * scaleFactor;
	size_t outHeight = inHeight * scaleFactor;

	double step = 1.0 / scaleFactor;

	for (size_t y = 0; y != outHeight; ++y)
	{
		string temp;
		size_t yIndex = static_cast<size_t>(step * y);
		for (size_t x = 0; x != outWidth; ++x)
		{
			size_t xIndex = static_cast<size_t>(step * x);
			temp += inImage[yIndex][xIndex];
		}
		outImage.push_back(temp);
	}
	
	return outImage;
}


int main()
{
	char *data[] = {"abc", "def", "ghi"};

	size_t dataSize = sizeof(data) / sizeof(char *);

	vector<string> image(data, data + dataSize);
	vector<string> scaledImage;

	scaledImage = Enlarge(image, 3);

	// Output both image and scaledImage vectors
	copy(image.begin(), image.end(), ostream_iterator<string>(cout, "\n"));
	cout << endl;
	copy(scaledImage.begin(), scaledImage.end(), ostream_iterator<string>(cout, "\n"));

	return 0;
}