Michael J. Freeman

Mid-infrared supercontinuum generation to 4.5 μm in ZBLAN fluoride fibers by nanosecond diode pumping

A mid-infrared supercontinuum (SC) is generated in ZBLAN (ZrF4-BaF2-LaF3-AlF3-NaF...) fluoride fibers from amplified nanosecond laser diode pulses with a continuous spectrum from approximately 0.8 microm to beyond 4.5 microm. The SC has an average power of approximately 23 mW, a pump-to-SC power conversion efficiency exceeding 50%, and a spectral power density of approximately -20 dBm/nm over a large fraction of the spectrum. The SC generation is initiated by the breakup of nanosecond laser diode pulses into femtosecond pulses through modulation instability, and the spectrum is then broadened primarily through fiber nonlinearities in approximately 2-7 m lengths of ZBLAN fiber. The SC long-wavelength edge is consistent with the intrinsic ZBLAN material absorption.

Supercontinuum generation from ~1.9 to 4.5 μmin ZBLAN fiber with high average power generation beyond 3.8 μm using a thulium-doped fiber amplifier

A mid-IR supercontinuum (SC) fiber laser based on a thulium-doped fiber amplifier (TDFA) is demonstrated. A continuous spectrum extending from ∼1.9 to 4.5 μm is generated with ∼0.7 W time-average power in wavelengths beyond 3.8 μm. The laser outputs a total average power of up to ∼2.6 W from ∼8.5 m length of ZrF4─BaF2─LaF3─AlF3─NaF (ZBLAN) fiber, with an optical conversion efficiency of ∼9% from the TDFA pump to the mid-IR SC. Optimal efficiency in generating wavelengths beyond 3.8 μm is achieved by reducing the losses in the TDFA stage and optimizing the ZBLAN fiber length. We demonstrate a novel (to our knowledge) approach of generating modulation instability-initiated SC starting from 1.55 μm by splitting the spectral shifting process into two steps. In the first step, amplified approximately nanosecond-long 1.55 μm laser diode pulses with ∼2.5 kW peak power generate a SC extending beyond 2.1 μm in ∼25 m length of standard single-mode fiber (SMF). The ∼2 μm wavelength components at the standard SMF output are amplified in a TDFA and coupled into ZBLAN fiber leading to mid-IR SC generation. Up to ∼270 nm SC long wavelength edge extension and ∼2.5× higher optical conversion efficiency to wavelengths beyond 3.8 μm are achieved by switching an Er:Yb-based power amplifier stage with a TDFA. The laser also demonstrates scalability in the average output power with respect to the pulse repetition rate and the amplifier pump power. Numerical simulations are performed by solving the generalized nonlinear Schrodinger equation, which show the long wavelength edge of the SC to be limited by the loss in ZBLAN.

10.5 W Time-Averaged Power Mid-IR Supercontinuum Generation Extending Beyond 4

A novel, all-fiber-integrated supercontinuum (SC) laser is demonstrated and provides up to 10.5 W time-averaged power with a continuous spectrum from ~0.8 to 4 mum. The SC is generated in a combination of standard single-mode fibers and ZrF4-BaF2-LaF3-AlF3-NaF (ZBLAN) fluoride fibers pumped by a laser-diode-based cladding-pumped fiber amplifier system. The output SC pulse pattern can be modulated by directly modulating the seed laser diode. Near-diffraction-limited beam qualities are maintained over the entire SC spectrum. The SC average power is also linearly scalable by varying the input pump power and pulse repetition rate. We further investigate the theoretical limitations on the achievable average power handling and spectral width for the SC generation in ZBLAN fibers. Based on the thermal modeling, the standard ZBLAN fiber can handle a time-averaged power up to ~15 W, which can be further scaled up to ~40 W with a proper thermal coating applied onto the ZBLAN fiber. The SC long-wavelength edge is limited by the nonlinear wavelength generation processes, fiber bend-induced loss, and glass material loss. By using a ZBLAN fiber with a 0.3 numerical aperture, the SC spectrum could extend out to ~4.5 mum, which is then limited by the material loss.

\mu

Mid-infrared supercontinuum (SC) extending to ~4.0 mum is generated with 1.3 W time-averaged power, the highest power to our knowledge, in ZBLAN (ZrF(4)-BaF(2)-LaF(3)-AlF(3)-NaF...) fluoride fiber by using cladding-pumped fiber amplifiers and modulated laser diode pulses. We demonstrate the scalability of the SC average power by varying the pump pulse repetition rate while maintaining the similar peak power. Simulation results obtained by solving the generalized nonlinear Schrödinger equation show that the long wavelength edge of the SC is primarily determined by the peak pump power in the ZBLAN fiber.

m With Direct Pulse Pattern Modulation

We measure the diffuse reflection spectrum of solid samples such as explosives (TNT, RDX, PETN), fertilizers (ammonium nitrate, urea), and paints (automotive and military grade) at a stand-off distance of 5 m using a mid-infrared supercontinuum light source with 3.9 W average output power. The output spectrum extends from 750-4300 nm, and it is generated by nonlinear spectral broadening in a 9 m long fluoride fiber pumped by high peak power pulses from a dual-stage erbium-ytterbium fiber amplifier operating at 1543 nm. The samples are distinguished using unique spectral signatures that are attributed to the molecular vibrations of the constituents. Signal-to-noise ratio (SNR) calculations demonstrate the feasibility of increasing the stand-off distance from 5 to ~150 m, with a corresponding drop in SNR from 28 to 10 dB.

Power scalable mid-infrared supercontinuum generation in ZBLAN fluoride fibers with up to 1.3 watts time-averaged power

Third order cascaded Raman shifting is used to generate light to 1867 nm in sulfide fibers, and the nonlinearity is measured to be ~5.7 times 10-12 (m/W). Damage at ~1 GW/cm2 limits the wavelength shift range.

Stand-off detection of solid targets with diffuse reflection spectroscopy using a high-power mid-infrared supercontinuum source

Supercontinuum (SC) with a continuous spectrum from ~0.8-3 mum is generated in a standard single-mode fiber followed by high-nonlinearity fiber. The SC is pumped by 2-ns laser diode (LD) pulses amplified in a multistage fiber amplifier, and the two octave spanning continuum is achieved by optimizing a two-stage process that separates pulse breakup and soliton formation from spectral broadening. We also demonstrate scalability of the average power in the continuum from 27 mW to 5.3 W by increasing the pulse repetition rate from 5 kHz to 1 MHz, while maintaining comparable peak power. We attribute the generated SC spectrum to the ensemble average of multiple solitons and the superposition of their corresponding spectra. The hypothesis is confirmed through simulation results obtained by solving the generalized nonlinear Schrodinger equation (NLSE). Similar SC spectra can also be obtained by using both femtosecond and nanosecond pump pulses. Furthermore, by tailoring the input pulse shape, we propose and simulate the generation of the entire SC spectrum in one single soliton under quasi-continuous-wave (CW) pulse pumping scheme.

Third order cascaded Raman wavelength shifting in chalcogenide fibers and determination of Raman gain coefficient

The Support Vector Machine (SVM) provides a robust, accurate and effective technique for pattern recognition and classification. Although the SVM is essentially a binary classifier, it can be adopted to handle multi-class classification tasks. The conventional way to extent the SVM to multi-class scenarios is to decompose an m-class problem into a series of two-class problems, for which either the one-vs-one (OVO) or one-vs-all (OVA) approaches are used. In this paper, a practical and systematic approach using a kernelised SVM is proposed and developed such that it can be implemented in embedded hardware within a road-side camera. The foreground segmentation of the vehicle is obtained using a Gaussian mixture model background subtraction algorithm. The feature vector describing the foreground (vehicle) silhouette encodes size, aspect ratio, width, solidity in order to classify vehicle type (car, van, HGV), In addition 3D colour histograms are used to generate a feature vector encoding vehicle color. The good recognition rates achieved in the our experiments indicate that our approach is well suited for pragmatic embedded vehicle classification applications.

Supercontinuum Generation in Silica Fibers by Amplified Nanosecond Laser Diode Pulses

We have observed amplitude noise levels as much as 1.8 dB below the Standard quantum limit in the light generated by an external-cavity-stabilized index-guided quantum-well laser operating at room temperature. The measured amplitude noise remains more than 1.5 dB below the Standard quantum limit as the lasing wavelength of this system is tuned from 831 to 841 nm.

Road vehicle classification using Support Vector Machines

In recent years, automatic human action recognition has been widely researched within the computer vision and image processing communities. Here we propose a real-time, embedded vision solution for human action recognition, implemented on an FPGA-based ubiquitous device. There are three main contributions in this paper. Firstly, we have developed a fast human action recognition system with simple motion features and a linear support vector machine classifier. The method has been tested on a large, public human action dataset and achieved competitive performance for the temporal template class of approaches, which include “Motion History Image” based techniques. Secondly, we have developed a reconfigurable, FPGA based video processing architecture. One advantage of this architecture is that the system processing performance can be reconfigured for a particular application, with the addition of new or replicated processing cores. Finally, we have successfully implemented a human action recognition system on this reconfigurable architecture. With a small number of human actions (hand gestures), this stand-alone system is operating reliably at 12 frames/s, with an 80% average recognition rate using limited training data. This type of system has applications in security systems, man–machine communications and intelligent environments.