China
Africa
Explore chapters and articles related to this topic
Epipolar geometry
Published in Aditi Majumder, M. Gopi, Introduction to Visual Computing, 2018
Epipolar geometry defines geometric constraints across multiple cameras capturing the same scene. This enables simplification of common problems (like finding correspondences) when dealing with important vision tasks like motion estimation or 3D depth reconstruction. It is fascinating to see how even relatively simple constraints can make such hard problems tractable. In this chapter we will cover the fundamental concepts of epipolar geometry. We start this chapter by defining the notations we will be using.
A survey on vision guided robotic systems with intelligent control strategies for autonomous tasks
Published in Cogent Engineering, 2022
Abhilasha Singh, V. Kalaichelvi, R. Karthikeyan
This technique can be useful when two monocular cameras or stereo camera is used to capture the same scene then we use epipolar geometry to calculate the relation between the two views. When the optical centers of each camera are projected onto the other camera<apos;>s image plane epipolar points are formed (Prasad et al, 2018). Becerra et al (2011) proposed epipolar geometry for dynamic pose estimation of differential drive mobile robot. The pose was further predicted using discrete Extended Kalman Filter (EKF) and it showed better performance than the Essential matrix-based approach. Chesi et al. (2006) used epipolar geometry to estimate the position of apparent contours in unstructured 3D space so that using VS robot can navigate efficiently. Mutib et al (2014) used ANN-based VS to learn the mapping between mobile drifts and visual kinematics and used multi-view epipolar geometry to compensate for the rotational errors. The main advantage of epipolar geometry is that they are easy to implement and provide the relationship between image planes without the knowledge of 3D structure. The only drawback is that they are more susceptible to noise.
Simultaneous kinematic calibration, localization, and mapping (SKCLAM) for industrial robot manipulators†
Published in Advanced Robotics, 2019
Jinghui Li, Akitoshi Ito, Hiroyuki Yaguchi, Yusuke Maeda
Therefore, the following procedure for implementing SKCLAM method is proposed: Employ an RGB-D camera mounted on the manipulator end effector to capture RGB images and corresponding depth images. Meanwhile, record the variables of each manipulator's joint at that moment.Detect and describe the feature points from the captured RGB images, and match these feature points.Employ the joint variables from step 1 to calculate the homogeneous transformation matrix from the base coordinate system of the robot manipulator to that of the RGB-D camera. Meanwhile, the fundamental matrix of the RGB-D camera also needs to be clarified.Employ the feature matches from step 2 and the fundamental matrix of the camera from step 3 to draw epipolar lines. Two images of the same scene are related by epipolar geometry.Define an error function based on the epipolar geometry and minimize the error function for the kinematic calibration.Reconstruct a map of the environment by employing the calibrated kinematic parameters and the point cloud data from the depth images.
Robust user authentication model for securing electronic healthcare system using fingerprint biometrics
Published in International Journal of Computers and Applications, 2019
Sharmin Jahan, Mozammel Chowdhury, Rafiqul Islam
The stereo algorithm compares two fingerprint images (test and gallery database) and computes the degree of similarity between the test image and the gallery image and identifies the user’s fingerprint that produces the best matching score. Prior to stereo matching, we need to rectify the fingerprint images for their alignment. The test and gallery images are rectified and the similarity score is computed by computing the stereo matching cost (similarity score) of every row of the rectified images. The rectification allows the use of epipolar geometry environment where the epipolar lines are horizontal i.e. parallel to the lines of the image sequences. In epipolar geometry, any point lying on an epipolar line in the reference image (i.e. test image) corresponds to a point lying on the same epipolar line in the target image (i.e. gallery image). After rectification of the two fingerprint images, the matched points have necessarily the same coordinate in the both images. Therefore, in case of searching for corresponding points in two fingerprint images, it is only necessary to search in the same epipolar line, reducing a 2D search space to 1D. In order to achieve rectification, we adopt the algorithm proposed by Fusiello et al. [36].