Algorithm 2: Adaptive Blurred Image Classification

Input: Image sequence $g_1,g_2,\cdots ,g_n$ 1  Initialization: scalar factor γ, window size S, bias parameters β, $K_0=0$ 2  for $i=1,\cdots ,n$ do 3  if $i>S$ then 4    $K_i=\gamma K_{i-1}+(1-\gamma )(\frac{1}{S}{\sum }_{s=1}^S{b}_{i-s}+\beta )$ 5  else 6    if $i==S$ then 7      $K_i=K_i/S$ 8    else 9      $K_i=K_{i-1}+b_i$ 10    end 11  end 12  if $b_i>K_i$ then 13    $g_i$ is blurred 14  else 15    $g_i$ is clear 16  end 17 end