Visual SLAM method based on instance segmentation

Abstract

The invention provides a visual SLAM algorithm based on instance segmentation, and the algorithm comprises the steps: firstly extracting feature points of an input image, and carrying out the instancesegmentation of the image through employing a convolutional neural network; secondly, using instance segmentation information for assisting in positioning, removing feature points prone to causing mismatching, and reducing a feature matching area; and finally, constructing a semantic map by using the semantic information segmented by the instance, thereby realizing reuse and man-machine interaction of the established map by the robot. According to the method, experimental verification is carried out on image instance segmentation, visual positioning and semantic map construction by using a TUM data set. Experimental results show that the robustness of image feature matching can be improved by combining image instance segmentation with visual SLAM, the feature matching speed is increased,and the positioning accuracy of the mobile robot is improved; and the algorithm can generate an accurate semantic map, so that the requirement of the robot for executing advanced tasks is met.

Description

technical field [0001] The present invention relates to the field of visual image technology, in particular to a visual SLAM method based on instance segmentation. Background technique [0002] The SLAM algorithm is that the mobile robot starts from a certain place in the unknown environment, repeatedly reads the sensor observation data during the movement, analyzes and obtains the environmental characteristics and its own position and posture, and builds an incremental map of the surrounding environment in real time. Among them, visual sensors can obtain richer image information than other sensors, and visual sensors are light, cheap, and easy to install. Therefore, SLAM research based on visual sensors has become a current research hotspot. The realization of visual SLAM mainly includes the feature point method and the direct method. Among them, the direct method completely relies on the search image pixel gradient to estimate the robot pose, which requires the robot to no...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems of the current visual simultaneous localization and map construction (Visual Simultaneous Localization and Mapping, VSLAM) algorithm, the robustness of pose estimation is poor, the positioning accuracy is low, and the semantic map suitable for autonomous navigation cannot be generated, etc., the present invention proposes a method based on instance segmentation. Visual SLAM algorithm

Images