Li Fu, Jinhao Fan, Zhaoyi Zhang, Lei Zhang. Wall emissivity solved by bidirectional reflection distribution function combined with Bi-LSTM network[J]. Infrared and Laser Engineering, 2023, 52(2): 20220355
- Infrared and Laser Engineering
- Vol. 52, Issue 2, 20220355 (2023)
Fig. 1. Geometric definition of bidirectional reflection distribution function
Fig. 2. Space angle definition of observation direction
Fig. 3. Reflectance envelope of light source perpendicular to wall surface
Fig. 4. Block diagram of wall reflection brightness measurement system
Fig. 5. Beam shaping
Fig. 6. Schematic diagram of wall reflection brightness light path
Fig. 7. RNN architecture
Fig. 8. LSTM node model
Fig. 9. Bi-LSTM network model
Fig. 10. Regression band and atmospheric window
Fig. 11. Luminance regression network structure
Fig. 12. Loss of RMSE during training
Fig. 13. Calibrate blackbody brightness and emissivity. (a) Radiance curve; (b) Emissivity curve
Fig. 14. Wall reflection radiance measurement experimental system. (a) Reflected radiation source; (b) MR170 radiometer
Fig. 15. Brightness curve measured in the direction of observation angle
=35°. (a) 3-5 μm; (b) 8-14 μm
观测角度
=35°方向测量所得亮度曲线。 (a) 3~5 μm；(b) 8~14 μm
Fig. 16. Measurement results of each observation angle (3-5 μm). (a) Luminance regression curve; (b) BRDF curve
Fig. 17. Measurement results of each observation angle (8-14 μm). (a) Luminance regression curve; (b) BRDF curve
| |||||||||||||||||||||
Table 1. Raw training set data
| ||||||||||||||||||||||||||
Table 2. Results of standard emissivity training input set
|
Table 3. Projection area of reflection area
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20