Abstract
Loop closure detection is a significant requirement for simultaneous localization and mapping (SLAM) to recognize revisited place. This paper presents a novel line-based loop closure detection method for vision-based SLAM that allows reliable loop closure detections, especial under structural environment. The performance of coping with perceptual aliasing conditions is more competitive than point based methods. The bag of words model is extended in this work which uses only line features. A variant of TF-IDF (term frequency & inverse document frequency) scoring scheme is proposed by adding a discrimination coefficient to improve the discrimination of image similarity scores, further to reinforce the similarity evaluation of two images. LBD (Line Band Descriptor) and binary LBD features are extracted to build visual vocabularies. Temporal consistency and spatial continuity checks enhance detection reliability. The performance of proposed scoring scheme was compared with original TF-IDF, results show that our proposed scheme has competitive discrimination ability. We also compared the query performance of our vocabularies with ORB-based, MSLD (mean standard-deviation line descriptor)-based, and PL (Point-and-Line)-based vocabularies, results indicate that our vocabularies obtain the highest successful retrieval rate. The performance of the whole loop closure detection algorithm was also evaluated in terms of precision, recall and efficiency, which were compared with ORB, MSLD, PL-based methods, and also with CNN-based method, results demonstrate that our method is superior to others with satisfactory precision and efficiency.
Highlights
Simultaneous localization and mapping (SLAM) has received a lot of attention in the robotic community during past years
The second experiment compared the query performance of vocabularies trained using ORB, MSLD, LBD, binary LBD and PL(Point-and-Line) features, we used the scheme of [23] to compute the similarity score based on PL features
Take NCEPU.TB2 dataset with α = 0.3 as example, Figure 9 shows the execution time expended for each image with (a) binary LBD, (b) ORB, (c) LBD, (d) MSLD and (e) PL features
Summary
Simultaneous localization and mapping (SLAM) has received a lot of attention in the robotic community during past years. This work focuses on loop closure detection for visionbased SLAM, especially for the visual SLAM under structural environment, such as indoors, urbans, outdoors with buildings, transmission lines, etc. The bag of words (BOW) model has been widely used in appearance-based methods [12]–[15] This model builds a ‘‘dictionary’’ beforehand in an offline process through clustering visual feature descriptors extracted from a large number of training images. Line features are abundant in much man-made structural environment, such as indoors, transmission towers, buildings, etc They can convey structural information effectively, because a 3D line spans over a higher level space than a 3D point as [25] described. Few of loop closure detection work uses line features. A whole loop closure detection algorithm that applies only line features is proposed, especially work for man-made environment.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
