M=[exp(-(x-3)^2-(y-3)^2) for x in 1:5, y in 1:5].+rand(5,5)*0.3 #fake image with a maximum in the middle affected by random noise

f1=KernelOp(M, (1,1)) do A,Is,I   # (1,1) is the size of the kernelop
    avg=zero(eltype(A))
    @inbounds for i in Is
        avg+=A[i]
    end
    avg/length(Is)
end

f2=KernelOp(f1, (1,1)) do A,Is,I   # (1,1) is the size of the kernelop
    m=A[I]
    @inbounds for i in Is
        m=max(m,A[i])
    end
    m
end

show(stdout,"text/plain",f1.==f2)

5×5 BitArray{2}:
 0  0  0  0  0
 0  0  0  0  0
 0  0  1  0  0
 0  0  0  0  0
 0  0  0  0  0

using KernelOps
b=Int8.([
    0 0 0 1 1 0 0 0;
    0 0 1 0 0 1 0 0;
    0 1 0 0 0 0 1 0;
    1 0 0 0 0 0 0 1;
    1 0 0 0 0 0 0 1;
    0 1 0 0 0 0 1 0;
    0 0 1 0 0 1 0 0;
    0 0 0 1 1 0 0 0;
])
cb=copy(b);

function evolve(cells)
    game_of_life=KernelOp(cells,(1,1)) do M,Is,I
        s=sum(M[Is])-M[I]
        if M[I]==1
            s<2 && return Int8(0)
            s<=3 && return Int8(1)
            return Int8(0)
        else
            s==3 && return Int8(1)
        end
        return M[I]
    end
    return game_of_life|>collect #add this collect, unless you want julia to suffer
end

show(stdout,"text/plain",b)
println("\n")
for i in 1:100
    b=evolve(b)
    show(stdout,"text/plain",b)
    println("\n")
    if b==cb
        print("Pattern repeats itself after $i iterations")
        break;
    end
end