Navigation image matching filtering method based on K-means clustering algorithm

Abstract

The invention provides a navigation image matching filtering method based on K-means clustering algorithm, which comprises the following steps: 1. Obtaining an image of the motion process through the camera by the mobile terminal; 2. Extracting feature points for the image acquired by the camera; 3. Obtaining the matching point pairs by matching the extracted feature points of the image; 4. Using the K-means clustering algorithm to eliminate the mismatching point pairs to correct the measurement error of the mobile terminal to calibrate the coordinate position. Using the method, highly accurate matching point pairs can be obtained in a stable and real-time way, the accuracy of image matching and navigation route can be improved, and the accuracy of navigation can be finally improved. Compared with the existing navigation image matching filtering method, the proposed method has the characteristics of simplicity, high efficiency, timeliness and strong stability, which can improve the calculation speed of SLAM system and reduce the burden of SLAM system.

Description

technical field [0001] The invention relates to autonomous positioning and navigation technology, in particular to a navigation image matching filtering method based on K-means clustering algorithm. Background technique [0002] Mobile terminals refer to positioning and navigation devices with image recognition functions, such as mobile robots, cars, drones, etc. During their movement, they need to complete a series of operations such as obstacle avoidance, positioning, map construction, and path planning. When the mobile terminal is moving in an uncertain and unknown environment, it needs to gradually build a map of the surrounding environment, and refer to its position and attitude, and use this map for autonomous positioning and navigation. [0003] In recent years, with the development of computer vision technology and the enhancement of computer computing power, Simultaneous Localization And Mapping (SLAM) based on visual sensors has received extensive attention. SLAM ...

AI Technical Summary

Problems solved by technology

[0006] However, the above-mentioned methods have the following disadvantages: (1) The accuracy of mismatch detection in the method based on function fitting completely depends on the fitted model obtained by the method

If there are outliers with large errors, the error of the fitted model will be large

(2) The graph-based method has the disadvantages that if there are mismatching points with the same graph structure, it cannot be correctly distinguished, and images with relatively large deformations are prone to misjudgment. lead to a slower algorithm

(3) In the statistical model-based method, the distance ratio method cannot eliminate all mismatching points for elastic image matching, and if there are many matching points, the calculation speed will be slow; if there are multiple models, the RANSAC algorithm usually cannot be used

Images