Abstract:For the problem where dynamic objects in complex environments have a large impact on simultaneous localization and mapping (SLAM) pose estimation and mapping accuracy, you only look once (YOLO) detection method combining optical flow and geometric constraints is proposed to identify and eliminate dynamic feature points. Firstly, the latent dynamic feature points in the environment are detected based on YOLO, and the static and dynamic regions are divided according to motion consistency. Secondly, motion detection and tracking are carried out based on the optical flow method to preliminarily estimate the motion state, and then, re-judgment is performed in combination with geometric constraints to reduce tracking loss and precision reduction caused by false elimination of feature points. Finally, on the basis of eliminating dynamic feature points, static feature points are used for pose estimation to avoid the interference of dynamic feature points on pose estimation. Finally, based on the Technical University Munich (TUM) dataset and SLAM accuracy evaluation indexes, such as absolute trajectory error (ATE), and so on, the proposed method of this paper is experimentally tested and evaluated, and the ATE index of this paper's algorithm is improved by 8.1% and 96.35% compared with ORB-SLAM2 under TUM's low-dynamic sequences and high dynamic sequences, respectively, and the results show that this paper's algorithm has a better accuracy under the dynamic environment.

Local Binary Pattern Based Ghost Elimination in Objects Detection

An Efficient Method For Detecting Ghosts And Left Objects In Intelligent Video Surveillance

Vision-Based Moving Objects Detection with Background Modeling

Background extraction model and shadow detection algorithm in traffic flow video detection

Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals

Effective abandoned object detection method in complex scene

Moving Object Detection in Video Sequence Images Based on an Improved Visual Background Extraction Algorithm

A Fast Moving Object Detection Method Via Local Neighborhood Similarity

Dynamic Background Subtraction Based on Spatial Extended Center-Symmetric Local Binary Pattern

Moving Object Detection Algorithm based on Improved Visual Background Extractor

Moving Objects Extraction Method in Complex Traffic Scene

Triple Background Modeling Based Detection of Violative Behaviors Against City Appearance

Shadow Detection Method in Video Sequences Based on LBP

High Efficient Moving Object Extraction and Classification in Traffic Video Surveillance

A Moving Detection Algorithm Based on Space-Time Background Difference

Dynamic background subtraction based on appearance and motion pattern

Real-Time Human Detection Based on the Improved Local Binary Pattern

Detection and Elimination of Dynamic Feature Points Based on YOLO and Geometric Constraints

Research on Background Modeling Method Based on Center-Symmetric Local Binary Patterns

An Efficient Approach of Moving Objects Detection in Complex Background

Dynamic Background Subtraction Using Spatial-Color Binary Patterns