Paper Info
Title
Image Processing And Splicing Method For 3d Optical Scanning Surface Reconstruction Of Wood Grain
Abstract
Based on environment compensation, scanning image processing technology was employed to investigate point cloud data and space matching method for wood grain. Collision avoidance recognition algorithm was used to collocate mark points, when remarkably reduced the error matching of distance coincidence mark points. The proposed method used marking of flag sample points based on weight to compensate for the marking points ambiguity of distinguishing information in scanning environment, and select the optimal path for the weighted results. The same splicing points in different images was identified, solving the problem of fuzzy splicing by distance matching. Experimental results and three-dimensional (3D) printing wood cross-section model reconstructed by surface fitting were compared. Results showed that the 3D scanning image mosaic of wood growth texture at the cross-section had no obvious stereo characteristics. The proposed method has improved the accuracy of surface mosaic in reverse scanning imaging for wood grain. This method can be applied to support the application needs of reverse surface reconstruction.
Year
DOI
Venue
2020
10.1142/S021800142054021X
INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE
Keywords
DocType
Volume
Wood texture 3D scanning, environmental compensation, avoid overlapping marked points, image processing, curved surface reverse reconstruction
Journal
34
Issue
ISSN
Citations 
8
0218-0014
0
PageRank 
References 
Authors
0.34
0
2
Name
Order
Citations
PageRank
Shufeng Jiang100.34
Keqi Wang200.68