Chenghao Liu, Zhi Li, Can Xu, Qichen Tian. BRDF Model for Commonly Used Materials of Space Targets Based on Deep Neural Network[J]. Acta Optica Sinica, 2017, 37(11): 1129001
- Acta Optica Sinica
- Vol. 37, Issue 11, 1129001 (2017)
Fig. 1. Angles schematic in BRDF measurement
Fig. 2. Activation effect of ReLU. (a) Before activation; (b) after activation
Fig. 3. Dropout diagram
Fig. 4. Modeling process of DNN
Fig. 5. Fitting effect of anode plate coated by gold film. (a) θi=2°; (b) θi=15°; (c) θi=60°; (d) θi=75°
Fig. 6. Loss function value decline curve of anode plate coated by gold film
Fig. 7. Fitting effect of epoxy paint. (a) θi=2°; (b) θi=30°; (c) θi=75°; (d) θi=80°
Fig. 8. Loss function value decline curve of epoxy paint
Fig. 9. Comparison of fitting effects of three models for epoxy paint. DNN model: (a) θi=5°, (b) θi=30°; five-parameter model: (c) θi=5°, (d) θi=30°; Phong model: (e) θi=5°, (f) θi=30°
Fig. 10. Comparison of fitting effects of three models for polyimide film. DNN model: (a) θi=5°, (b) θi=30°; five-parameter model: (c) θi=5°, (d) θi=30°; Phong model: (e) θi=5°, (f) θi=30°
|
Table 1. Modeling errors of 2A12 white paint
|
Table 2. Modeling errors comparison of training samples with different scales
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