This will count as one of your downloads.
You will have access to both the presentation and article (if available).
We demonstrate a deep-learning-based fiber imaging system that can transfer real-time artifact-free cell images through a meter-long Anderson localizing optical fiber. The cell samples are illuminated by an incoherent LED light source. A deep convolutional neural network is applied to the image reconstruction process. The network training uses data generated by a setup with straight fiber at room temperature (∼20 ° C) but can be utilized directly for high-fidelity reconstruction of cell images that are transported through fiber with a few degrees bend or fiber with segments heated up to 50°C. In addition, cell images located several millimeters away from the bare fiber end can be transported and recovered successfully without the assistance of distal optics. We provide evidence that the trained neural network is able to transfer its learning to recover images of cells featuring very different morphologies and classes that are never “seen” during the training process.
The experiments were enabled by an ultrafast thulium-doped fiber laser delivering 110 fs pulses at high repetition rates, and an argon gas-filled antiresonant hollow-core fiber (ARHCF) with excellent transmission and weak anomalous dispersion, leading to the self-compression of the pulses. We have shown that ARHCFs are well-suited for nonlinear pulse compression around 2 μm wavelength and that this concept features excellent power handling capabilities. Based on this result, we discuss the next steps for energy and average power scaling including upscaling the fiber dimensions in order to fully exploit the capabilities of our laser system, which can deliver several GW of peak power. This way, a 100 W-class laser source with mJ-level few-cycle pulses at 2 μm wavelength is feasible in the near future.
This work explores a very short fiber length high average power multi-mode Raman laser system. The custom 200um graded index fiber is pumped by 30ns pulses with average powers up to 750W and pulse energies up to 7.5mJ at 1030nm, by a solid state commercial laser system. Pump-only and seeded configurations are examined. In the seeded case, higher order mode activation is demonstrated by detuning the single mode seed to preferentially feed energy to the less confined modes.
5 orders of Stokes are demonstrated, ranging from 1078nm to 1350 nm. Beam enhancement is observed by qualitative measurement of minimum beam waist, and average powers up to 70W are achieved at an energy of 1.4mJ.
The cause of these variations in delay is the change in the glass properties of the optical fiber with temperature. Both the refractive index of the glass and the length of the fiber are dependent on the ambient temperature. Traditional optical fiber suffers from a delay sensitivity of 39 ps/km/K. We are reducing the temperature sensitivity of the fiber delay through the application of a novel design of optical fiber, Anti-Resonant Hollow Core Fiber. The major improvement in the thermal sensitivity of this fiber comes from the fact that the light is guided in an air core, with very little overlap into the glass structure. This drastically reduces the impact that the thermally sensitive glass properties have on the propagation time of the optical signal. Additionally, hollow core fiber is inherently radiation insensitive, due to the light guidance in air, making it suitable for space applications.
View contact details
No SPIE Account? Create one
