Multimode Fiber Imaging Based on Temporal-Spatial Information Extraction

Runze Zhu; Fei Xu

doi:10.3788/LOP230726

Journals >Laser & Optoelectronics Progress >Volume 60 >Issue 11 >Page 1106011 > Article

Laser & Optoelectronics Progress
Vol. 60, Issue 11, 1106011 (2023)

Multimode Fiber Imaging Based on Temporal-Spatial Information Extraction

Runze Zhu^1、† and Fei Xu^†、*

Author Affiliations

College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093, Jiangsu, China

show less

DOI: 10.3788/LOP230726 Cite this Article Set citation alerts

Runze Zhu, Fei Xu. Multimode Fiber Imaging Based on Temporal-Spatial Information Extraction[J]. Laser & Optoelectronics Progress, 2023, 60(11): 1106011 Copy Citation Text

show less

Fig. 1. Research framework of multimode fiber imaging

Download full size

Fig. 2. Research framework of MMF imaging based on TM measurement

Download full size

Fig. 3. MMF imaging based on spatial-domain TM measurement. (a) Principle of spatial-domain TM; (b) experimental diagram of spatial-domain TM measurement; (c) experimental diagram of MMF imaging based on spatial-domain TM measurement

Download full size

Fig. 4. MMF imaging based on frequency-domain TM measurement. (a) Principle of frequency -domain TM; (b) experimental diagram of spatial-domain TM measurement; (c) experimental diagram of MMF imaging based on frequency -domain TM measurement

Download full size

Fig. 5. Imaging system of deep brain of living mouse based on multimode fiber^[31]

Download full size

Fig. 6. Endoscopic LIDAR^[37]. (a) Schematic of experimental setup; (b) snapshot of true scene; (c) typical depth resolved images

Download full size

Fig. 7. Research framework of MMF imaging based on phase conjugation and phase optimization

Download full size

Fig. 8. MMF imaging based on phase conjugation and phase optimization. (a) Experimental scheme of MMF imaging based on phase conjugation; (b) experimental scheme of MMF imaging based on phase optimization

Download full size

Fig. 9. Research framework of MMF compressive imaging based on structure illumination

Download full size

Fig. 10. MMF compressive imaging based on speckle illumination. (a) Principle; (b) experimental scheme

Download full size

Fig. 11. MMF-based super-resolution and super-speed endo-microscopy^[51]. (a) Characterization of imaging resolution and speed using 0.22NA MMF; (b) characterization of imaging resolution and speed using 0.1NA MMF

Download full size

Fig. 12. Evolution of ultrashort pulses in MMF

Download full size

Fig. 13. High-speed all-fiber imaging based on temporal information extraction^[59]. (a) Schematic of the experimental setup; (b) flow of the reconstruction process; (c)–(e) detailed imaging devices

Download full size

Fig. 14. Research framework of machine learning-assisted MMF imaging

Download full size

Fig. 15. Process of machine learning-assisted multimode fiber imaging

Download full size

Fig. 16. High-speed all-fiber micro-imaging with large depth of field^[71]

Download full size

Fig. 17. Research framework of MMF imaging under dynamic perturbance

Download full size

Fig. 18. Anti-interference imaging based on proximal wavefront measurement. (a) Based on the virtual beacon source^［73］; (b) based on the partial reflector^[74]; (c) based on the metasurface reflector stacks^[75]; (b) based on the guide star^[76]

Download full size

Fig. 19. Image transmission through a dynamically perturbed multimode fiber by deep learning^[82]

Download full size

Method	Imaging system	Calibration process	Imaging process	Focused light spot generation
TM measurement	Reference light path/no reference	TM calibration	Focused spot raster- scanning/inverse matrix calculation	Loading holograms calculated from TM calculation
Phase conjugation	Reference light path	Calculating phase conjugation	Focused spot raster- scanning	Loading holograms calculated from phase conjugation
Phase optimization	No reference	Select target light field and phase optimization	Focused spot raster- scanning	Loading holograms calculated by phase iterative optimization
Structure illumination	No reference	Generation and recording of illumination patterns	Compressive imaging	—

Table 1. Comparison of MMF imaging methods based on spatial-domain information extraction

Work	Fiber/probe parameter	Resolution /μm	FOV /（μm×μm）	Imaging speed	Working distance /μm	Method
Choi et al.^［23］	0.48NA，200 μm diameter	1.8	—	1 frame/s for 12300 pixel	~40	TM
Papadopouloset al.^［40］	450 μm-diameter probe	<1	100×110	—	~200	Phase conjugation
Turtaevet al.^［31］	60 µm external diameter	1.18 ± 0.04	50×50	3.5 frame/s for 7000 pixel	—	TM
Caravaca-Aguirre et al.^［25］	Four different commercial MMFs	2	80×80	a few seconds for one image	~100	TM
Vasquez-Lopez et al.^［30］	50 µm diameter，0.22NA	1.35	50×50	2.4 s per 120 pixel× 120 pixel image	0-100	TM
Leite et al.^［35］	0.2 mm× 0.4 mm dimension endoscope	Angular resolution：（3.59 ± 0.07）mrad	Related to working distance	4.4 s for each 10⁵ pixel image	20000-400000	TM
Wen et al.^［36］	50 µm diameter，0.22NA	1.4	—	3.7 s for one volume image	0-102	TM
Stellingaet al.^［37］	Illumination fiber：0.22NA 25 μm radius，collection fiber：500 μm diameter	Angular resolution：16 mrad	Related to working distance	5 frame/s for ~23000 points	0-2.5 m	TM
Lee et al.^［38］	105 µm core diameter0.22NA	Lateral，axis resolution：10 μm，~267 μm（working distance：600 μm）	FOV diameter： ~167 μm（working distance：600 μm）	120 Hz（limited by the camera）	0-1200	TM
Cheng et al.^［46］	15-meter long MMF（0.22NA，105 µm core）	—	—	∼18 s for one projection through 2000 iterations	—	Phase optimization
Mahalatiet al.^［16］	50 μm diameter0.19NA	Twofold reduction in the width of the PSF	40 × 40	36 min for 3000 patterns，12 s for the reconstruction of 75 pixel×75 pixel image	~25	Structure illumination
Amitonovaet al.^［50］	50 μm diameter，0.22NA	（1.4 ± 0.2）μm	—	0.014 s for 150 patterns 20 s for the calculation of 50 pixel× 50 pixel image	<20	Structure illumination
Caravaca-Aguirre et al.^［60］	250 μm × 125 μm probe	3，1.6 μm	—	Up to a minute for photoacoustic imaging	50	Structure illumination
Amitonovaet al.^［51］	50/105 µm diameter，0.22/0.1NA	2 times better than the diffraction limit	~2000，4000	20 times faster than the Nyquist-Shannon limit	—	Structure illumination
Fukui et al.^［53］	Core diameter of 105 µm and NA of 0.22	Number of resolvable features：1007	~100 × 100	3 s for one pattern	0-100	Structure illumination
Abrashitovaet al.^［52］	50 µm diameter，0.22NA	2-fold higher resolution than the diffraction limit	—	5 frame/s	—	Structure illumination
Zhu et al.^［54］	50 µm diameter，0.22NA	Number of resolvable features：1400	3000 × 3000	242 s for 801 speckle patterns	~6000	Structure illumination
Dong et al.^［55］	Illumination fiber：0.22NA 25 μm core radius，collection fiber：500 μm core diameter	Axial resolution：16 μm	100 × 100 × 200	1.7 s for the acquisition of the entire volume，6.3 min for the 3D reconstruction	0-200	Structure illumination
Liu et al.^［59］	Triple-cladding fiber probe + ball lens（580 µm diameter）	<15 μm	>200 × 200	detection frame rate of 15.4×10⁶ frame/s	—	Time- domain analysis

Table 2. Parameter comparison of representative works of MMF imaging based on temporal-spatial information extraction

Runze Zhu, Fei Xu. Multimode Fiber Imaging Based on Temporal-Spatial Information Extraction[J]. Laser & Optoelectronics Progress, 2023, 60(11): 1106011

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information