Visual SLAM algorithm based on improved CoHOG

doi:10.13474/j.cnki.11-2246.2022.0355

Abstract

Abstract: The mobile robot has a large amount of calculation, long running time and low accuracy of matching in the closed-loop detection link of SLAM. So this paper improves the CoHOG closed-loop detection algorithm, improves the HOG descriptor in the algorithm to GDF-HOG descriptor, to enhance image characteristics and improve the efficiency of image feature extraction. Before the matching process, GDF-HOG global rough matching is added to reduce the number of visual templates and improve the computational efficiency of the algorithm. After the matching process, region of interest (ROI) position matching is added for testing to reduce false positives of closed-loop detection and improve accuracy.Finally, the closed-loop detection algorithm proposed in this paper is combined with RatSLAM, and tested in open data sets and real environments. The test results show that the proposed algorithm has advantages in the accuracy of closed-loop detection and adaptability to the environment.

Key words: image recognition and apparatus, visual place recognition, CoHOG algorithm, closed-loop detection, RatSLAM

CLC Number:

P237

YU Yao, SUN Xinzhu, GUO Junyang, CHEN Mengyuan. Visual SLAM algorithm based on improved CoHOG[J]. Bulletin of Surveying and Mapping, 2022, 0(12): 42-50.

References

[1] 李国竣, 徐延海, 段杰文, 等.利用局部自适应阈值方法提取ORB-SLAM特征点[J].测绘通报, 2021(9):32-36.
[2] 盛超, 潘树国, 赵涛, 等.基于图像语义分割的动态场景下的单目SLAM算法[J].测绘通报, 2020(1):40-44.
[3] NICHOLSON L, MILFORD M, SVNDERHAUF N.QuadricSLAM:dual quadrics from object detections as landmarks in object-oriented SLAM[J].IEEE Robotics and Automation Letters, 2019, 4(1):1-8.
[4] MOLLOY T L, FISCHER T, MILFORD M, et al.Intelligent reference curation for visual place recognition via Bayesian selective fusion[J].IEEE Robotics and Automation Letters, 2021, 6(2):588-595.
[5] CHEN Zetao, LIU Lingqiao, SA I, et al.Learning context flexible attention model for long-term visual place recognition[J].IEEE Robotics and Automation Letters, 2018, 3(4):4015-4022.
[6] FISCHER T, MILFORD M.Event-based visual place recognition with ensembles of temporal windows[J].IEEE Robotics and Automation Letters, 2020, 5(4):6924-6931.
[7] 刘强, 段富海, 桑勇, 等.复杂环境下视觉SLAM闭环检测方法综述[J].机器人, 2019, 41(1):112-123.
[8] NG P C, HENIKOFF S.SIFT:predicting amino acid changes that affect protein function[J].Nucleic Acids Research, 2003, 31(13):3812-3814.
[9] BAY H, ESS A, TUYTELAARS T, et al.Speeded-up robust features (SURF)[J].Computer Vision and Image Understanding, 2008, 110(3):346-359.
[10] 张良桥, 陈国良, 许晓东, 等.一种用于图像特征提取的改进ORB-SLAM算法[J].测绘通报, 2019(3):16-20.
[11] 刘国忠, 胡钊政.基于SURF和ORB全局特征的快速闭环检测[J].机器人, 2017, 39(1):36-45.
[12] 张云洲, 胡航, 秦操, 等.基于栈式卷积自编码的视觉SLAM闭环检测[J].控制与决策, 2019, 34(5):981-988.
[13] 龚希, 吴亮, 谢忠, 等.融合全局和局部深度特征的高分辨率遥感影像场景分类方法[J].光学学报, 2019, 39(3):19-29.
[14] 牛杰, 卜雄洙, 钱堃, 等.一种融合全局及显著性区域特征的室内场景识别方法[J].机器人, 2015, 37(1):122-128.
[15] CHEN Z, LAM O, JACOBSON A, et al.Convolutional neural network-based place recognition[C]//Proceedings of 2014 Conference on Robotics and Automati-on.Melbourne:[s.n.], 2014.
[16] KHALIQ A, EHSAN S, CHEN Zetao, et al.A holistic visual place recognition approach using lightweight CNNs for significant ViewPoint and appearance changes[J].IEEE Transactions on Robotics, 2020, 36(2):561-569.
[17] MCMANUS C, UPCROFT B, NEWMANN P.Scene signatures:localised and point-less features for localisation[C]//Proceedings of Conference on Robotics:Science and Systems.[S.l.]:Science and Systems Foundation, 2014.
[18] CHEN Zetao, JACOBSON A, SVNDERHAUF N, et al.Deep learning features at scale for visual place recognition[C]//Proceedings of 2017 IEEE International Conference on Robotics and Automation.Singapore:IEEE, 2017:3223-3230.
[19] ARANDJELOVIĆ R, GRONAT P, TORII A, et al.NetVLAD:CNN architecture for weakly supervised place recognition[C]//Proceedings of 2016 IEEE Transactions on Pattern Analysis and Machine Intelligence.[S.l.]:IEEE, 2016:1437-1451.
[20] CHEN Zetao, MAFFRA F, SA I, et al.Only look once, mining distinctive landmarks from ConvNet for visual place recognition[C]//Proceedings of 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).Vancouver, BC, Canada:IEEE, 2017:9-16.
[21] SHEKHAR R, JAWAHAR C V.Word image retrieval using bag of visual words[C]//Proceedings of the 10th IAPR International Workshop on Document Analysis Systems.Gold Coast, QLD, Australia:IEEE, 2012:297-301.
[22] ZAFFAR M, EHSAN S, MILFORD M, et al.CoHOG:a light-weight, compute-efficient, and training-free visual place recognition technique for changing environments[J].IEEE Robotics and Automation Letters, 2020, 5(2):1835-1842.
[23] HANBAY K, ALPASLAN N, TALU M F, et al.Continuous rotation invariant features for gradient-based texture classification[J].Computer Vision and Image Understanding, 2015, 132:87-101.
[24] GUCLU O, CAN A B.Histogram based visual place recognition for improving SLAM performance[C]//Proceedings of 2016 International Conference on Autonomous Robot Systems and Competitions (ICARSC).Bragança, Portugal:IEEE, 2016:174-180.
[25] PAZ L M, PINIÉS P, TARDÓS J D, et al.Large-scale 6-DOF SLAM with stereo-in-hand[J].IEEE Transactions on Robotics, 2008, 24(5):946-957.
[26] SCARAMUZZA D, FRAUNDORFER F.Visual odometry tutorial[J].IEEE Robotics & Automation Magazine, 2011, 18(4):80-92.