A Single Objective GA and PSO for the Multimodal Palmprint Recognition System

Algorithm 2

Proposed multimodal system using PSO.

	Step 1: we initialize the number of chromosomes K as 50 and the number of iterations as 100. We initialize ω = 0.73, C1 = 1.496, and C2 = 1.496
	Step 2: we generate 50 random particle positions and initialize 50 random velocities.
	Step 3: modified bilobe and trilobe ordinal filters are used to extract the features of left and right palmprint images for both training and testing sets.
	Step 4: the subset features for training and testing sets are selected by using particles’ positions present in population k.
	Step 5: here, the recognition rate (Pbest) is the fitness function. It is calculated by the NN classifier. We consider two feature vectors A and B and . The Euclidean distance measure is used to calculate the distance between two vectors.
	Step 6: if Pbest_current > Pbest_previous
	Yes-update Pbest_current
	No-Pbest_previous
	Step 7: if gbest_current > gbest_previous
	Yes-update gbest_current
	No-gbest_previous and update velocity, the position of the random particle. We continue until the last epochs are reached.
	end