Junction Point Detection Algorithm for SAR Image

Algorithm 1

Junction point classification algorithm.

Input: M={𝑥₁,𝑥₂,…,𝑥_𝑚}, the dip angle set {𝜑₁,𝜑₂,…,𝜑_𝑚} corresponding to M, and the threshold angle 𝜏

Output: The classification result set E of branch edges’ points

Initialize: Set the set E = NULL

Main steps of algorithm:

(1) Define a 𝑚×𝑚 matrix H and initialize H = null; and then compute all the elements of the matrix

  according to the equation

          H=𝜙𝑖,𝑗cos_𝑖−𝜙_𝑗,ifcos𝜙_𝑖−𝜙_𝑗≥cos(𝜏),null,otherwise,1≤𝑖<𝑗≤𝑚

(2) Calculate the position of the maximal value of H; assuming the maximum is at (𝑖,𝑗), set H(𝑖,𝑗) = null,

   𝜐₁=sin(𝜑_𝑖)+sin(𝜑_𝑗) and 𝜐₂=cos(𝜑_𝑖)+cos(𝜑_𝑗), and then, compute the classification center 𝜑

               𝜙=tan⁻¹𝜐₁𝜐₂,𝜐₁>0,𝜐₂>0tan⁻¹𝜐₁𝜐₂+360^∘,𝜐₁>0,𝜐₂<0tan⁻¹𝜐₁𝜐₂+180^∘,𝜐₂<090^∘,𝜐₁>0,𝜐₂=0270^∘,𝜐₁<0,𝜐₂=0

(3) Update 𝜐₁, 𝜐₂ and 𝜑 by all angles that satisfy the condition cos(𝜑_𝑘−𝜑)≥cos(𝜏)(𝑘=1,2,…,𝑚) searched

  from the set {𝜑₁,𝜑₂,…,𝜑_𝑚};

(4) Repeat step (3) until the angle set that satisfies the condition cos(𝜑_𝑘−𝜑)≥cos(𝜏) never change and place

  the junction points corresponding to the angle set into E;

(5) Repeat steps (2), (3) and (4) until all the elements of H are null, and then return E.