基于Vision Transformer多模型融合的视觉闭环检测算法  

Vision loop closure detection algorithm based on Vision Transformer multi-model fusion

作　　者：胡正南 胡立坤[1,2] HU Zhengnan;HU Likun(School of Electrical Engineering,Guangxi University,Nanning 530004,China;Advanced Measurement&Control&Intelligent Power Research Center,Guangxi University,Nanning 530004,China)

机构地区：[1]广西大学电气工程学院,南宁530004 [2]广西大学先进测控与智能电力研究中心,南宁530004

出　　处：《激光杂志》2024年第6期75-81,共7页Laser Journal

基　　金：国家自然基金资助项目(No.61863002);广西重点研发计划基金资助项目(No.AB21220039)。

摘　　要：针对闭环检测在图像特征表示方面存在信息丢失的问题,提出一种基于Vision Transformer (Vi T)与卷积神经网络进行多模型融合的特征提取算法。首先,将输入图像进行特征提取,然后将高维的图像特征向量进行核主成分分析(KPCA)降维,构建成新的图像特征表示;同时,提出了一种新的范围匹配算法,通过相应的范围框架去限制并选择范围进行特征匹配。实验结果表明:所提算法相比于其他的算法,有着更高的准确率和匹配速率,达到了更好的鲁棒性与实时性的要求,证明了该算法在闭环检测上的有效性。Aiming at the problem of information loss in image feature representation of loop closure detection,a feature extraction algorithm based on Vision Transformer(ViT)with convolutional neural network for multi-model fusion was proposed.Firstly,feature extraction was carried out on the input image,and then the high-dimensional image feature vector was reduced by kernel principal component analysis(KPCA)to construct a new image feature representation.At the same time,a new range-matching algorithm was proposed,which limited and selected the range for feature matching through the corresponding range framework.The experimental results show that the proposed algorithm compared with other algorithms has higher accuracy and matching rate,and achieves better robustness and real-time requirements,which proves the effectiveness of the proposed algorithm in loop closure detection.

关 键 词：闭环检测 VIT 多模型融合 KPCA 范围匹配 

分 类 号：TN249[电子电信—物理电子学]