Skip to main content

Table 1 Algorithm for weight-adaptive model

From: Human Visual Attention Mechanism-Inspired Point-and-Line Stereo Visual Odometry for Environments with Uneven Distributed Features

Input

Pixel coordinates of xp, xs, and xe; feature number nf

Output

Adaptive coefficients of weights ω

1:

for i←1 to nf do

2:

Cj ← The pixel coordinates(ORB,LSD)

3:

if (CjSi) then si++, Si ‹- Cj

4:

end

5:

for i←1 to n2 do

6:

if si ≠ 0 then SkSi, k++

7:

end

8:

for m←1 to k do

9:

if \(({s_{i \cdot }}_{near(S)} = 0)\& \& ({s_m} > {s_i}_{ \cdot near(near(S))})\)

10:

then Sm Sm + near(S)

11:

end

12:

\(\omega { = }q \cdot (1/s_{m} )\)

13:

return ω