I think mentioned issue can be discussed in two views :
1- the computational cost of each techniques in theory
2- the computational complexity of each applied techniques in Code (ex. MATLAB).
In term of the first approach, it can be refereed to the corresponding literatures. for instance, some papers mentioned that: " The computational complexity (CC) of the gray level transformation technique is basically determined by the number of comparison operations for finding the gray-level groups and the to number of multiplication and/or division operations for calculating the gray level transformation functions".
By the way , computational complexity of a applied technique (or a code) is generally measured by time duration of technique performance. But, the time duration depends on several factors such as: memory, frequency of CPU, structure of algorithm (consecutive or parallelism).
As a result, one suggestion to estimate CC of a Matlab code is:
1- find variables of the code function
2- classify variables to : (1)dependent , and (2)independent.
3- find relation of dependent variables
4- iterate performing of desired Code function for a range of independent variables and measured time duration of each iteration.
5- regression analysis of obtained time value as a function of independent variable.
for example, if we want to estimated CC of image histogram code in Matlab we can find independent variables of function : number of gray levels, and size of image. Then, we can run histogram function for a range of gray levels and image size. finally, regression analysis can estimated the worst-case scenario computational complexity.
Image processing is one of the branches of computer science. It is interested in performing operations on images in order to improve them according to specific criteria or extract some information from them. The traditional image processing system consists of six consecutive stages, respectively
Image acquisition by optical sensor (eg camera, lys sensor, etc.)
Pre-processing Filter the image from a blur or convert it to a binary image
Segmentation to separate important information (eg any object in the image) from the background
Extraction features or attributes
Categorize, link to the style you are familiar with and identify patterns