Name
Abstract | ||
|---|---|---|
With the development of remote sensing technology, the simultaneous acquisition of 3D point cloud and color information has become the constant goal for scientific research and commercial applications in this field. However, since radar echo data in practice refer to the value of the spectral channel and its corresponding energy, it is still impossible to obtain accurate tristimulus values of the point through color integral calculation after traditional normalization and multispectral correction. Furthermore, the reflectance of the target, the laser transmission power and other factors lead to the problems of no echo energy or weak echo energy in some bands of the visible spectrum, which further leads to large chromatic difference compared to the color calculated from the spectral reflectance of standard color card. In response to these problems, the hyperbolic tangent spectrum correction model with parameters is proposed for the spectrum correction of the acquired hyperspectral LiDAR in the 470-700 nm band. In addition, the improved gradient boosting decision tree sequence prediction algorithm is proposed for the reconstruction of missing spectrum in the 400-470 nm band where the echo energy is weak and missing. Experimental results show that there is relatively small chromatic difference between the obtained spectral information after correction and reconstruction and the spectrum of standard color card, achieving the purpose of true color reconstruction. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.3390/rs13152854 | REMOTE SENSING |
Keywords | DocType | Volume |
true-color reconstruction, multispectral correction, spectrum reconstruction, hyperspectral LiDAR | Journal | 13 |
Issue | Citations | PageRank |
15 | 0 | 0.34 |
References | Authors | |
0 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Tengfeng Wang | 1 | 0 | 0.34 |
| Xiaoxia Wan | 2 | 0 | 1.01 |
| Bowen Chen | 3 | 33 | 8.73 |
| Shuo Shi | 4 | 15 | 6.21 |