Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Improved ORB-SLAM algorithm for outdoor offline navigation system

An offline navigation and outdoor technology, applied in computing, computer components, instruments, etc., can solve problems such as low accuracy and low efficiency

Inactive Publication Date: 2019-10-08
电子科技大学成都学院
View PDF4 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to: provide an improved ORB-SLAM algorithm for outdoor offline navigation systems, solve the problem that the feature point extraction process in the current outdoor ORB-SLAM mainly uses a fixed threshold T selection algorithm and multi-scale Harris corner point detection, There are problems such as low efficiency and low precision

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Improved ORB-SLAM algorithm for outdoor offline navigation system
  • Improved ORB-SLAM algorithm for outdoor offline navigation system
  • Improved ORB-SLAM algorithm for outdoor offline navigation system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0088] Comparison experiment of the number of feature points extracted:

[0089] The present invention compares different images of the same scene, by changing the brightness of the image, the SIFT algorithm such as Figure 4 As shown, the OFAST algorithm before optimization is as Figure 5 As shown, and the optimized OFAST algorithm such as Figure 6 As shown, the comparison chart of the feature point extraction effect is compared

[0090] As shown in Table 1, in the experimental results, the number of feature points extracted by the SIFT algorithm is 206, the number of feature points extracted by the original OFAST algorithm is 179, and the number of feature points extracted by the improved OFAST algorithm is 116

[0091] Compared with the original OFAT algorithm, the improved OFAST algorithm reduces the number of feature points by 35.2%. It can be seen that the number of feature points extracted by the algorithm in the present invention is far less than that of the origi...

Embodiment 2

[0095] AVERAGE TIME COMPARISON EXPERIMENT

[0096] The extraction time represents the speed of the algorithm in extracting feature points. When navigating in the wild environment, time is a major focus, and the faster the time, the stronger the algorithm performance;

[0097] Table 2 Average Time

[0098]

[0099] As shown in Table 2, the SIFT algorithm extracts feature points in 375.3s, the original OFAST algorithm extracts feature points in 9.3s, and the improved OFAST algorithm extracts feature points in 1.3s. The timeliness of feature points is increased by 80.6%, and the improved OFAST is 99.6% faster than SIFT algorithm in extracting feature points. It can be seen that the improved OFAST algorithm saves a lot of time, and its work efficiency is much higher than the original OFAST algorithm and SIFT algorithm.

Embodiment 3

[0101] Matching accuracy comparison experiment

[0102] The matching accuracy of extracted feature points plays an important role in the later mapping. The higher the matching accuracy, the better the effect of later mapping, and the clearer the route planning.

[0103] Table 3 matching accuracy

[0104]

[0105] As shown in Table 3, the correct rate of the SIFT algorithm is 15.2%, the correct rate of the original OFAST algorithm is 37.4%, and the correct rate of the improved OFAST algorithm is 51.8%. Compared with the original OFAST algorithm, the correct rate of the improved OFAST algorithm is 14.4%. The correct rate is 51.8%. The correct rate of the improved OFAST algorithm is 36.6% higher than that of the SIFT algorithm. It can be seen that the improved OFAST algorithm can match feature points more accurately, and the accuracy is much greater than the original OFAST algorithm and SIFT algorithm.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an improved ORB-SLAM algorithm for an outdoor offline navigation system. In the FAST algorithm, a threshold value T represents the minimum contrast ratio of the extracted feature points with surrounding neighborhood points, and the minimum contrast ratio of the extracted feature points with surrounding neighborhood points is the maximum limit of noise elimination. The valueselected by the threshold T directly affects the accuracy of feature point extraction. Wherein the larger the threshold T is, the fewer the extracted feature points are, and the smaller the thresholdT is, the more the selected feature points are. Although the fixed threshold value T method adopted in the original FAST-12 algorithm reduces the calculation amount to a certain extent, as the threshold value T is fixed, after people take pictures in the field, objective factors such as shadows, poor illumination and sudden noise possibly exist in the images, and selection of the threshold valueT cannot be well changed along with changes of the gray level and the noise of the global images.

Description

technical field [0001] The invention belongs to the field of ORB-SLAM algorithms, and relates to an improved ORB-SLAM algorithm for an outdoor off-line navigation system. Background technique [0002] With unmanned driving and unmanned exploration, the application of AR / VR is increasing day by day. In outdoor real-time mapping applications, for the problem of image mismatch, it is necessary to improve the flexibility of the fixed threshold T, and for the problem of long mapping time, it is necessary to optimize the Laplacian response value by improving the efficiency of rough extraction Computing; how to optimize ORB-SLAM and apply ORB-SLAM more effectively has become a problem to be solved. The ORB algorithm in ORB-SLAM includes two steps of feature point extraction and feature description, the most important of which is feature point extraction. The traditional OFAST algorithm uses a fixed threshold T selection algorithm to quickly exclude some non-feature points when ext...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/46G06K9/62
CPCG06V10/757G06V10/44G06V10/462
Inventor 邹倩颖关杰文肖航符鑫珺
Owner 电子科技大学成都学院
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com