A Rapid Road Image Mosaic Method Based on Monocular Camera

As an important infrastructure, road has attracted a lot of attentions in recent years. Most of current road information extraction methods rely on either elaborate algorithms or complex equipment. This manuscript proposes a simple and rapid image mosaic method to extract road surface based on a mon...

Full description

Saved in:
Bibliographic Details
Published inJournal of physics. Conference series Vol. 1437; no. 1; pp. 12008 - 12018
Main Authors Li, Wunan, Cao, Yu
Format Journal Article
LanguageEnglish
Published Bristol IOP Publishing 01.01.2020
Subjects
Algorithms
Cameras
Computer vision
Coordinates
Frames
Image processing
Information retrieval
Mosaics
Physics
Road surface
Stitching
Tensors
Online AccessGet full text

Cover

More Information
Summary:As an important infrastructure, road has attracted a lot of attentions in recent years. Most of current road information extraction methods rely on either elaborate algorithms or complex equipment. This manuscript proposes a simple and rapid image mosaic method to extract road surface based on a monocular camera. First, key frames are extracted to remove redundant information from image sequences. To eliminate perspective effect, extracted key frames are transformed into top view of road images based on inverse perspective mapping algorithm with the help of an attitude tensor. Then, a coarse-to-fine registration strategy is proposed to align all transformed key frames into a unified coordinate system. Finally, considering the specificity of transformed images, a superposing and overlapping image fusion strategy is proposed to alleviate the effect of stitching seam. The experiments are conducted on a road with the size of 128×17 meters, and the experimental results demonstrate the effectiveness and efficiency of proposed method. In conclusion, the proposed method is simple, effective and efficient and can be applied in a wide range of applications such as large scale road surface diagnose.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1437/1/012008