Multi-level-Color-Image-Segmentation-using-Differential-Evolution

Population initialisation ,Evaluation, Mutation, CrossOver and the segmented image

Flowchart

DE Steps

1)Input: -The input image to be segmented -Size of the population NP 2)Population initialization: -Generate a population P with the given NP 3)Evaluation: -Calculate the fitness value 4)Mutation : -Selecting three random numbers (a, b, c) to create a mutant vector 5)Crossover: -Recombine the mutant vector with P 6)Output: -The segmented image

DE Steps(Flowchart)

Implementation

Population generation

Rand () function is used to get random pixel value from the image.

Fitness Calculation

Variance for n- cluster is found. (Inter Cluster Variance)

Differential Evolution

Cluster Generation

Visualization of each cluster

Parameters for DE

Sample Output

Research Work

https://link.springer.com/chapter/10.1007/978-981-13-8196-6_27