Take a $2$x$2$ window $($starting from the upper$-$left corner of the image$)$ and sum all the red, green, and blue values separately for all the pixels in that window. In a $2-$by$-2$ window, there will be a total of $4$ pixels for which to sum the three separate channels.

Take the sum found for each channel of all the pixels in the window and divide them by $4.$

Set the red, green, and blue values of the top$-$left$-$most pixel of the $2-$by$-2$ window to be the value calculated from #$2.$

Perform this on the entire image moving the $2$x$2$ window by one pixel each time from left$-$to$-$right and top$-$to$-$bottom where top$-$most will be row $0$ and left$-$most is column $0.$