Image enhancement assignment for IP subject at college 👌
as requested by professor Moacir Ponti
The program requires the OpenCV library for matrix operations and image generation, numpy for mathematical operations and matplotlib for histogram plotting.
It can be simply run from a properly configured Windows cmd or Linux terminal, just head to the project's folder and type:
python enhaim.py
There are 4 mathematical functions for image enhancement, each of which the program will properly work on the desired input image and display on screen, alongside with a histogram comparison between the original image and the enhanced one:
- Logarithmic Enhancement
T(f) = c * log(1 + |f|), where c = 255 / log(1 + (R)), where R is the maximum intensity value found in f
- Gamma Correction
T(f) = f ^ y, where y is the desired gamma value
- Histogram Equalization
T(f) = (255 / MN) * hc(f), where M and N are the source's height and width and hc(f) is the cummulative histogram for the value in f
- Sharpening Filter
T(f) = a * f + b (convolution - f), where convolution is the resulting matrix of the convolution of f with the kernel w = [[0.05, 0.1, 0.05], [0.1, 0.4, 0.1], [0.05, 0.1, 0.05]]
Values calculated in the enhancements are then assigned into the output matrix as RGB values, with the following format:
out[x, y] = (result)
A RMSD value is also calculated, for analysis purposes. Please notice that each result is stored as float32, whereas the image shown is clipped as uint8.