A Novel Artificial Bee Colony Algorithm Based on Internal-Feedback Strategy for Image Template Matching

Algorithm 1

(1) Set population size SN, and maximum cycle number MCN. Set inefficient trial

  time counter trial(𝑖)=1 (𝑖=1,2,…,SN/2).

(2) Randomly initialize locations of SN/2 scout bees using (2)

(3) For  iter=1 to MCN, do % Employed bee phase

(4) For item=1 to SN/2, do

(5)   Generate 𝐗_item for the item-th employed bee to search according to (3)

(6)   If better position is found, then % Implementation of greedy selection

(7)    Update 𝐗_item and set trial(item)←1

(8)   Else

(9)     trial(item)←trial(item)+1

(10)  End if

(11) End for

(12)   For   item=1 to SN/2, do % Onlooker bee phase

(13)  Choose item-th employed bee to follow, and then generate local search location using (5) and (6)

(14)  If better position is found, then % Implementation of greedy selection

(15)   Update 𝐗_item and set trial(item)←1

(16)  Else

(17)    trial(𝑗)←trial(𝑗)+1

(18)  End if

(19) End for

(20) Collect all item∈[1,SN/2]∩ℤ satisfying trial(item)>Limit in an index set Ω

(21)  While Ω≠∅, do

(22)  Choose one 𝑘∈Ω

(23)  Re-initialize location of the 𝑘th employed bee using (2)

(24)  Eliminate the index 𝑘 from Ω

(25)  Set trial(𝑘)=1

(26) End while

(27) Memorize best-ever solution in the current cycle of iteration

(28) End for

(29) Output best-ever solution