One of the techniques that we tried for image contrasting was using a "brightness differential" which is a parameter to check the range of the brightness of input image. It is defined as the difference between intensities of two pixels of image. Our intuition is to change a pixel’s intensity value accordingly to the brightness differential of that pixel with its neighbor.

After summing up all the brightness differential of the pixel with its neighbor, we compare it with a threshold. If the value is greater than the threshold, we add a slight tint or whiteness to the pixel because the pixel is lighter compared to its neighbor. Otherwise, we add a shade or blackness to the pixel to make it darker compared to its neighbor. In a nutshell, our method increases the contrast of the image by adding tint or shade to pixel value based on a pool of neighboring pixel values.

This technique can be represented by the following mathematical formulation:

A potential drawback of this method is that it often outputs noisy images. It is because the algorithm loops over the image pixel by pixel which leads to some noise.