An Improved RANSAC Algorithm Based on Adaptive Threshold for Indoor Positioning

Algorithm 3

Indoor positioning method based on adaptive threshold RANSAC.

INPUT: imageL, imageR, F, K1, K2
(1)	Matrix_E = K1’FK2;
(2)	Vector_t, Matrix_R = recover (Matrix_E);
(3)	Vector_Keypoints, Mathes = feature_extract_match (imageL, imageR)
(4)	keypoints1 = detectore (imageL);
(5)	keypoints2 = detectore (imageR);
(6)	points1 = Pointchange (keypoints1);
(7)	points2 = Pointchange (keypoints2);
(8)	space_points = triangulation (keypoints1, keypoints2, Matches, Matrix_R, Vector_t)
(9)	for i = 0 to matches.size do
(10)	points1_cam = pixel2cam (keypoints1);
(11)	points2_cam = pixel2cam (keypoints2);
(12)	points2_trans = Matrix_R (points[i].x, points[i].y, points[i].z) + Vector_t;
(13)	end for
OUTPUT: location