Wyniki wyszukiwania - BazTech

1

3D maps integration based on overlapping regions matching

Drwięga Michał

Journal of Automation Mobile Robotics and Intelligent Systems

|

2021

|

Vol. 15, No. 3

70--80

EN

This paper presents a developed method of 3D maps integration based on overlapping regions detection and matching that works without an initial guess about transformation between maps. The presented solution is based on a classic pipeline approach from computer vision that has been applied to the 3D maps integration with multiple improvements related to model extraction and the descriptors matching. The process of finding transformation between maps consists of three steps. The first one is the extraction of the model from one of the maps. Then the initial transformation is estimated between extracted model and another map based on feature extraction, description, and matching. The assumption is that the maps have an overlapping area that can be used during the feature‐based alignment. In the last step, the initial so‐ lution is corrected using local alignment approaches, for example, ICP or NDT. The maps are stored in the octree based representation (octomaps) but during transformation estimation, a point cloud representation is used as well. In addition, the presented method was verified in various experiments: in a simulation, with wheeled robots, and with publicly available datasets. Eventually, the solution can be applied to many robotic applications related to the exploration of unknown environments. Nevertheless, so far it was validated with a group of wheeled robots. Furthermore, the developed method has been implemented and released as a part of the open‐source ROS package 3d_map_server.

2

Depth-based Descriptor for Matching Keypoints in 3D Scenes

Matusiak K., Skulimowski P., Strumillo P.

International Journal of Electronics and Telecommunications

|

2018

|

Vol. 64, No. 3

299--306

EN

Keypoint detection is a basic step in many computer vision algorithms aimed at recognition of objects, automatic navigation and analysis of biomedical images. Successful implementation of higher level image analysis tasks, however, is conditioned by reliable detection of characteristic image local regions termed keypoints. A large number of keypoint detection algorithms has been proposed and verified. In this paper we discuss the most important keypoint detection algorithms. The main part of this work is devoted to description of a keypoint detection algorithm we propose that incorporates depth information computed from stereovision cameras or other depth sensing devices. It is shown that filtering out keypoints that are context dependent, e.g. located at boundaries of objects can improve the matching performance of the keypoints which is the basis for object recognition tasks. This improvement is shown quantitatively by comparing the proposed algorithm to the widely accepted SIFT keypoint detector algorithm. Our study is motivated by a development of a system aimed at aiding the visually impaired in space perception and object identification.

3

Mathematical Model of Errors of Odometry and Georeferencing Channels in Visual Correlation Extreme Navigation

Mukhina M. P., Kharchenko V. P.

Logistics and Transport

|

2017

|

Vol. 33, No. 1

31--36

EN

The mathematic model of errors in correlation with the extreme navigation system (CENS) is developed basing on odometry and geo-referencing channels. The realization of the model is done in Simulink, and based on regular and random components of additive noise. The results of simulations prove accumulation of errors for odometry errors and its mitigation in case of geo-referencing in periods of correction.

4

Simultaneous localization and mapping: A feature-based probabilistic approach

Skrzypczyński P.

International Journal of Applied Mathematics and Computer Science

|

2009

|

Vol. 19, no 4

575-588

EN

This article provides an introduction to Simultaneous Localization And Mapping (SLAM), with the focus on probabilistic SLAM utilizing a feature-based description of the environment. A probabilistic formulation of the SLAM problem is introduced, and a solution based on the Extended Kalman Filter (EKF-SLAM) is shown. Important issues of convergence, consistency, observability, data association and scaling in EKF-SLAM are discussed from both theoretical and practical points of view. Major extensions to the basic EKF-SLAM method and some recent advances in SLAM are also presented.