원문정보
초록
영어
3-D reconstruction of industrial flame could be used to research for the completeness of combustion and flame temperature field. As the flame flicks and has no regular evident characteristic, 3-D reconstruction of flame is difficult and lack of study. The paper proposed a stereo matching algorithm for flame based on distance ratio constraint (DRCFMA). Taking flame characteristics into account and using the processed SIFT operates to get the ratio constraint of matched feature points, the algorithm is applied on the follow-up 3-D reconstruction of flame. The results shows the validity and feasibility of DRCFMA. Compared with conventional algorithms, the proposed algorithm has a better accuracy and shorter time-taken for industrial flame. The experiments had shown that the follow-up 3-D reconstruction of flame has clear and continuously changing greyscale layers.
목차
1. Introduction
2. System Setup
3. SIFT Operator
3.1 SIFT Scale Space
3.2 The Distribution of the Direction of Feature Points
4. Normalized Cross-Correlation Matching Algorithm
5. The Stereo Matching Algorithm for Flame based on Distance Ratio Constraint
5.1 The Post-treatment of SIFT Operation
5.2 The Constraint Ratio of Feature Matched Points
5.3 The Core Idea of Stereo Matching Algorithm for Flame based on Distance Ratio Constraint
5.4 The Stereo Matching Algorithm forFlame based on Distance Ratio Constraint
6. The Experiment and Results
6.1The Test Image
6.2 The Validation of the Match
6.3 The Result of 3-D Reconstruction of Flame
7. Conclusion
Acknowledgement
References