This work proposes algorithm based on MobileNetV3 to address the challenges of decreasing localization accuracy and increasing cumulative errors in traditional visual simultaneous localization and mapping (SLAM) loop closure detection algorithms in complex environments such as illumination variations, dynamic scenes, and viewpoint changes. A pretrained MobileNetV3 model is utilized to extract robust image features, followed by principal component analysis (PCA) and whitening to reduce the dimensionality of feature vectors and improve computational efficiency. Cosine similarity is employed to compute a similarity matrix of image features, and loop closures are identified based on predefined thresholds. Experimental results demonstrated that our MobileNetV3-based algorithm outperformed several comparative methods. Compared with the Bag-of-Visual-Words (BOVW) based approach, when using the New College and City Centre datasets our method achieved improvements in detection accuracy of 18.5 %and 19.3%, and enhanced detection speed of 30.6% and 34.4%, respectively. These results meet the accuracy and real-time performance requirements of visual SLAM systems. Furthermore, the algorithm was integrated into ORB-SLAM2 by replacing its original BOVW loop closure module. Evaluations using the EuRoC dataset showed that the enhanced ORB-SLAM2 achieved a 23.8% improvement in localization accuracy, with the estimated trajectories significantly closer to the Ground Truth. These results validate the feasibility and effectiveness of our algorithm for use within SLAM systems.

To meet the requirements of positioning accuracy in the pose recognition process of the SLAM (simultaneous localization and mapping) algorithm based on point-line features, an improved EPLF-VINS (monocular visual-inertial SLAM with efficient point-line flow features) algorithm has been proposed. The influence of gradient threshold parameters on the EDLines (line segment detection by edge drawing) line segment extraction algorithm was first analyzed. Secondly, after forward optical flow tracing of point features, reverse optical flow tracing was used to eliminate the wrong tracking points, thereby improving the accuracy of optical flow tracing. Then, an adaptive adjustment algorithm was fused at the line segment extraction of the EPLF-VINS algorithm, and the gradient threshold parameters were adjusted in real time by calculating the success rate of point feature optical flow tracing after reverse optical flow tracing. This allows dynamic adjustment of line segment extraction based on changes in the environment, and results in a better balance of calculation cost and positioning accuracy. Finally, based on the Robot Operating System (ROS) platform, the trajectory accuracy and efficiency of the improved EPLF-VINS algorithm and the comparison algorithm in the EuRoc and TUM-VI datasets were analyzed. The results show that the trajectory curve generated by the improved EPLF-VINS algorithm more closely matches the real trajectory, and has higher positioning accuracy while maintaining real-time performance.