Algorithm 1 Proposed feature selection method using CML-ELM.

Input:${\eta }_i$,${\eta }_i\in \left\{{\eta }_1,{\eta }_2, {\eta }_3,\dots , {\eta }_i\right\}$

Output:$S_w\left(i\right), S_w\left(i\right)\in \left\{S_w\left(1\right), S_w\left(2\right),\dots , S_w\left(i\right)\right\}$

Start

Step 1: Parameters Initialization

$S_w\left(1\right)={\eta }_i, i=0,1,2,3,\dots n$

${\alpha }_{-1}=0$

${\alpha }_0=1$

$LR=L{R}_0$

Step 2: For$i=1$to K

do

$b_i=\frac{{\alpha }_{i-2}-1}{{\alpha }_{i-1}}$

$A^i=S_w\left(i\right)+b_i\left(S_w\left(i\right)-S_w\left(i-1\right)\right)$

Step 3: Update$S_w\left(i+1\right)$

Step 4: Find the minimum value of$LR$among$(L{R}_{\left(i-1\right)}$,

$2L{R}_{\left(i-1\right)}$,$3L{R}_{\left(i-1\right)}$,…)

:-$f\left(S_w\left(i\right)\right)\le g_{\left(L{R}_i, U^i\right)}{S}_w\left(i+1\right)$

Step 5: Passed computed LR values in ELM classifier

Step 6: Find MSER for ELM classifier

Step 7: If MSER$\ge 0.1$

Update$L{R}_{i+1}$

Step 8:${\alpha }_{i+1}=\frac{1+\sqrt{1+4_i^2}}{2}$

End For

End