Algorithm 1. Pseudo code of SLIC-SegNet algorithm.

Preparatory work for SegNet module [30]:   a: Convolution operation, get the feature value x;   b: Batch Normalizing Transform;   c: Training a Batch-Normalized Network; Result: the trained SegNet module.

Initialize: k = {1 … K}, i = {1 … 480}, j = {1 … 360}, n = {1 … N}, m = { 1… M}. Bi(i−1 ,…, n) represents the set of pixels in each sorted area, the RGB components of each pixel are denoted as IBm(x, y); IAn(x, y) stand for the pixels in $A_n^k$, Mode represents the component value of the most frequently occurring RGB components of all pixels in $A_n^k$, recorded as IMode Repeat 1.  Collect kth image: Ik(xi, yj)    For k = 1…K do 2.    Run the SLIC module:        Assign the best matching pixels;        Compute new cluster centers and residual error E;        until E $\le$ threshold;        Enforce connectivity;        Output the each pixel block $A_n^k$ ∈ {A1, A2 … AN|${\displaystyle \sum _{n=1}^N{A}_n^k}=I_k$}; where An = n each pixel IAn(x, y) ∈ $A_n^k$ 3.    Run the trained SegNet module:        Activate feature value;        Deconvolution, get the feature value Xn;        Find the maximum probability of each pixel in all categories.        Output the label set $B_m^k$ ∈ {B1, B2 … BM|${\displaystyle \sum _{m=1}^M{B}_m^k}=B_k$}, IBm(x, y) ∈ $B_m^k$ 4.     Match $A^k and B^k$, define set $C_n^k$ = $A_n^k$         For i = 1 … n           find each xAn, yAn of ICn           $I_n^{Mode}$ = Mode(IB(xAn, yAn));           For each (xAn, yAn) ∈ $A_n^k$             Assignthe Mode of the RGB component to ICn(xAn, yAn):             ICn(xAn, yAn) = $I_n^{Mode}$           End for          End for          Output $C_n^k$ ∈ {C1, C2 … CN|${\displaystyle \sum _{n=1}^N{C}_n^k}=C_k$}, ICn(x, y) ∈ $C_n^k$ 5.  Gait selection and run the Robot;     End for Until the Robot switched off.