Deep learning for position fixing in the micron scale by using convolutional neural networks

Hongye Li; Hu Liang; Qihao Hu; Meng Wang; Zefeng Wang

doi:doi:10.3788/COL202018.050602

Chinese Optics Letters, 2020, 18 (5): 050602, Published Online: Apr. 28, 2020

Deep learning for position fixing in the micron scale by using convolutional neural networks Download： 808次

Hongye Li ¹Hu Liang ⁴Qihao Hu ¹Meng Wang ^1,2,3Zefeng Wang ^1,2,3,*

Author Affiliations

¹ College of Advanced Interdisciplinary Studies, National University of Defense Technology, Changsha 410073, China

² State Key Laboratory of Pulsed Power Laser Technology, Changsha 410073, China

³ Hunan Provincial Key Laboratory of High Energy Laser Technology, Changsha 410073, China

⁴ Tianjin Navigation Instruments Research Institute, Tianjin 300131, China

Figures & Tables

Fig. 1. Schematic of offset splicing.

下载图片查看原文

Fig. 2. Excitation efficiency of each fiber mode versus offset point (offset direction: $x$ axis).

下载图片查看原文

Fig. 3. Auto-correlation function of specklegrams: (a) offset point in the $x$ direction varies from $- 25 μ m$ to 25 μm; (b) offset point in the whole plane.

下载图片查看原文

Fig. 4. Excitation ratio of (a) the third and (b) the seventh lower modes at different offset points.

下载图片查看原文

Fig. 5. Specklegrams of two centrosymmetric offset points: (a) (6 μm, $- 5 μ m$ ) and (b) ( $- 6 μ m$ , 5 μm).

下载图片查看原文

Fig. 6. (a) Data processing flow and (b) CNN architecture.

下载图片查看原文

Fig. 7. Loss versus training epoch (loss, MSE in training set; and val_loss, MSE in validation set).

下载图片查看原文

Fig. 8. Distribution of difference between the prediction label and validation label.

下载图片查看原文

Fig. 9. Labeled offset points with $d$ higher than 2 μm in the validation set.

下载图片查看原文

Hongye Li, Hu Liang, Qihao Hu, Meng Wang, Zefeng Wang. Deep learning for position fixing in the micron scale by using convolutional neural networks[J]. Chinese Optics Letters, 2020, 18(5): 050602.

Deep learning for position fixing in the micron scale by using convolutional neural networks Download： 808次

Fig. 1. Schematic of offset splicing.

Fig. 2. Excitation efficiency of each fiber mode versus offset point (offset direction: $x$ axis).

Fig. 3. Auto-correlation function of specklegrams: (a) offset point in the $x$ direction varies from $- 25 μ m$ to 25 μm; (b) offset point in the whole plane.

Fig. 4. Excitation ratio of (a) the third and (b) the seventh lower modes at different offset points.

Fig. 5. Specklegrams of two centrosymmetric offset points: (a) (6 μm, $- 5 μ m$ ) and (b) ( $- 6 μ m$ , 5 μm).

Fig. 6. (a) Data processing flow and (b) CNN architecture.

Fig. 7. Loss versus training epoch (loss, MSE in training set; and val_loss, MSE in validation set).

Fig. 8. Distribution of difference between the prediction label and validation label.

Fig. 9. Labeled offset points with $d$ higher than 2 μm in the validation set.

关于本站 Cookie 的使用提示

全站搜索

Deep learning for position fixing in the micron scale by using convolutional neural networks Download： 808次

Fig. 1. Schematic of offset splicing.

Fig. 2. Excitation efficiency of each fiber mode versus offset point (offset direction: x axis).

Fig. 3. Auto-correlation function of specklegrams: (a) offset point in the x direction varies from −25 μm to 25 μm; (b) offset point in the whole plane.

Fig. 4. Excitation ratio of (a) the third and (b) the seventh lower modes at different offset points.

Fig. 5. Specklegrams of two centrosymmetric offset points: (a) (6 μm, −5 μm) and (b) (−6 μm, 5 μm).

Fig. 6. (a) Data processing flow and (b) CNN architecture.

Fig. 7. Loss versus training epoch (loss, MSE in training set; and val_loss, MSE in validation set).

Fig. 8. Distribution of difference between the prediction label and validation label.

Fig. 9. Labeled offset points with d higher than 2 μm in the validation set.

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索

Fig. 2. Excitation efficiency of each fiber mode versus offset point (offset direction: $x$ axis).

Fig. 3. Auto-correlation function of specklegrams: (a) offset point in the $x$ direction varies from $- 25 μ m$ to 25 μm; (b) offset point in the whole plane.

Fig. 5. Specklegrams of two centrosymmetric offset points: (a) (6 μm, $- 5 μ m$ ) and (b) ( $- 6 μ m$ , 5 μm).

Fig. 9. Labeled offset points with $d$ higher than 2 μm in the validation set.