融合Gist特征与卷积自编码的闭环检测算法  

Loop Closure Detection Algorithm Based on Convolutional Autoencoder Fused with Gist Feature

作　　者：邱晨力 黄东振 刘华巍[1] 袁晓兵[1] 李宝清[1] Qiu Chenli;Huang Dongzhen;Liu Huawei;Yuan Xiaobing;Li Baoqing(Science and Technology on Microsystem Laboratory, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 201800, China;University of Chinese Academy of Sciences, Beijing 100049, China)

机构地区：[1]中国科学院上海微系统与信息技术研究所微系统技术重点实验室,上海201800 [2]中国科学院大学,北京100049

出　　处：《激光与光电子学进展》2019年第18期190-198,共9页Laser & Optoelectronics Progress

基　　金：微系统技术国防科技重点实验室基金(CXJJ-17S072)

摘　　要：闭环检测算法可消除视觉同时定位与建图(VSLAM)系统的累计误差,并对构建全局一致性地图有重要作用。针对现有传统闭环检测算法在视角与场景外观变化下准确率与稳健性降低,及部分基于深度学习方法特征提取与闭环识别实时性不佳的问题,设计了一种融合Gist特征与卷积自编码的闭环检测算法,将Gist特征作为卷积自编码网络重构目标,可增强模型在外观变化下的场景特征表达能力;同时通过透视变换构造视角变化训练图像对,以提升模型在视角变化下闭环检测的准确率与稳健性。所设计的模型较精简,可实现实时关键帧特征提取与闭环检测。在Gardens Point与Nordland数据集的实验结果表明,相较于传统视觉词袋模型(BoVW)、Gist算法及现有部分深度学习方法,本文算法可以达到更高的准确率和稳健性。Loop closure detection algorithm is essential for the visual simultaneous localization and mapping (VSLAM) systems to reduce accumulative error and build a globally consistent map. When detecting loops under the change of viewpoint and scene appearance, the precision and robustness of traditional loop closure detection algorithms decline and some algorithms based on deep learning are difficult to extract features and perform loop closure detection in real time. To overcome these problems, we propose a novel loop closure detection algorithm based on convolutional autoencoder fused with Gist feature, forcing the encoder to reconstruct the Gist feature to enhance the expressive ability of the model when the scene appearance changes. In the same time, we warp images with randomized projective transformations to make the training pairs to improve the precision and robustness of the model when the viewpoint changes. Our model is relatively lightweight which is capable of extracting keyframe features and detecting loops in real time. The results of experiments on Gardens Point and Nordland datasets show that our model can achieve better precision and robustness compared with traditional methods, like bag of visual word (BoVW), Gist, and some other methods based on deep learning.

关 键 词：机器视觉 同时定位与建图 闭环检测 卷积自编码 深度学习 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]