Many hardware approaches have been developed for implementing hyperspectral imaging on fluorescence microscope systems each with tradeoffs. We present a novel approach combining multiple illumination wavelengths using solid state, LED emitters in a 2-mirror configuration similar to a Cassegrain reflector assembly. This approach provides spectral discrimination by scanning a range of fluorescence excitation wavelengths. A model of the spectral illuminator was designed using ray tracing software that included an emitter, lens, parabolic mirror, flat mirror, and liquid light guide input. The geometry of the LED and other optical components was optimized with results indicating a 10% optical transmission.

A 5- Dimensional Rapid Hyperspectral Imaging system is undergoing testing that can provide high-speed live cell imaging. The current design utilizes LEDs, a multifaceted mirror array, and lenses to focus light sources into a liquid light guide. Ideal locations of each component were analyzed with the goal to optimize optical transmission. A model of the system was constructed and tested using Monte Carlo optical ray tracing in TracePro software. LED light sources and lenses were simulated by importing their properties using software tools. Initial results show a transmission increase of 20% is possible.

Existing methods using hyperspectral imaging (HSI) to estimate skin chromophore concentrations fail to give an account of scattering properties crucial to many medical applications. To address this limitation, we propose to combine HSI with spatial frequency domain imaging (SFDI). Total acquisition time is around five seconds, making the process suitable for in vivo application, and analysis of the skin is performed by successive optimizations on the absorption and scattering properties. The problem of shadows occurring on complex shapes such as the face is addressed by using a metric that is robust to irradiance drift and by working on normalized data.

Hyperspectral chemical imaging (HCI) refers to the acquisition of both spatial and spectral information in a single acquisition time frame. HCI is a parallel scheme that allows the collection of spatially-resolved chemical/compositional data at much higher rate compared to traditional chemical imaging which is serial in nature. Recent advances in spatial light modulators (SLM) have enabled many novel HCI imaging schemes. We have implemented several HCI systems using reflective and diffractive SLMs. Their design and performance characterization will

The turbid nature of refractive index distribution within living tissues introduces severe aberrations to light propagation thereby severely compromising image reconstruction using currently available non-invasive techniques. Numerous approaches of endoscopy, based mainly on fibre bundles or GRIN-lenses, allow imaging within extended depths of turbid tissues, however their footprint causes profound mechanical damage to all overlying regions and their imaging performance is very limited. Progress in the domain of complex photonics enabled a new generation of minimally invasive, high-resolution endoscopes by substitution of the Fourier-based image relays with a holographic control of light propagating through apparently randomizing multimode optical waveguides. This form of endo-microscopy became recently a very attractive way to provide minimally invasive insight into hard-to-access locations within living objects. I will review our fundamental and technological progression in this domain and introduce several applications of this concept in bio-medically relevant environments.

Optical wavefront shaping is one of the most effective techniques in focusing light inside a scattering medium. Unfortunately, most of these techniques require direct access to the scattering medium or need to know the scattering properties of the medium beforehand. Through our scheme we develop a novel concept in which both the illumination and the detection is on the same side of the inspected object. We model the scattering medium being a biological tissue as a Matrix having mathematical properties matched to the physical and biological aspects of the sample. In our adaptive optics scheme, we aim to estimate the scattering function and thus to encode the wavefront of the illuminating laser light source using DMD (Digital Micromirror Device) with an inverse scattering function of the scattering medium, such that after passing its scattering function a focused beam is obtained.

Compressive sensing hyperspectroscopy techniques overcome the inherent issue of high data bandwidth spectroscopic imaging. However, common compressive spectrometer layouts have prohibitive losses for low-light levels applications, as it is the case of the spontaneous Raman effect. In this contribution, we present a high-throughput programmable spectrometer exploiting amplitude spectral modulation with a digital micromirror device. This new spectrometer allows conventional and compressive Raman imaging and spectroscopy acquisitions with shot-noise-limited sensitivity over large bandwidths. We demonstrate compressed hyperspectroscopy of biological specimens at faster speeds and at lower costs than traditional cameras, and at 250-nm spatial resolution.

We propose a novel digital micromirror device (DMD) based microscopy method to achieve high-speed and high-resolution phase imaging. By loading different Lee holograms on the DMD conjugated to the sample plane, a varying angle illumination could be achieved. This system is also capable of generating multiple illumination angles simultaneously for interference. By using a high speed camera, we are working on demonstrating > 100 fps synthetic aperture imaging speed.

While there is great interest in 3D printing for microfluidic device fabrication, the challenge has been to achieve feature sizes that are in the truly microfluidic regime (<100 μm). The fundamental problem is that commercial tools and materials, which excel in many other application areas, have not been developed to address the unique needs of microfluidic device fabrication. Consequently, we have created our own stereolithographic 3D printer and materials that are specifically tailored to meet these needs. We show that flow channels as small as 18 μm x 20 μm can be reliably fabricated, as well as compact active elements such as valves and pumps, resulting in highly integrated microfluidic devices as small as a few cubic millimeters.

Microchannels in fused silica have attracted much attention for their broad range of applications in microfluidics. We proposed to combine temporally shaped (double-pulse train) laser pulses with spatially shaped (Bessel beam) laser pulses. Through irradiation with a temporally shaped femtosecond laser Bessel beam, modification efficiency was improved by adjusting the pulse delay. By combining the FLICE method with a temporally shaped Bessel beam, high-throughput microchannels fabrication were achieved. The mechanism of etching depth enhancement was elucidated by localized transient free electrons dynamics-induced structural and morphological changes.This method enables high-throughput and high-aspect-ratio microchannels fabrication in fused silica for potential application in microfluidics.

Predictive visualisation of laser-processing of materials is challenging, as the nonlinear interaction of light and matter is complicated to model. Here, we demonstrate a visualisation approach that relies entirely on a neural network and uses training data obtained via laser machining using a digital micromirror device (DMD) as a spatial light modulator. The technique is able to generate a predicted surface image and 3D depth profile of the ablated surface for a wide range of beam shapes. The predicted visualisations are remarkably effective and almost indistinguishable from real experimental data in appearance.

Additive manufacturing of optical elements with its layered process brings challenges concerning the homogeneity of the material. Depending on the printing process local variations of the refractive index might occur within or between the layers. Using the Scanning Focused Refractive Index Microscopy the distribution of the refractive index within and between the layers is analyzed. The purpose of the presentation is to give a detailed insight into refractive index measurements at the layer interfaces of elements created by additive manufacturing while summarizing the used printing technologies. Possible use cases of the refractive index distributions within the part are also discussed.

This paper presents new methods to achieve parallel nanofabrication and high-resolution multi-photon imaging based on temporal focusing, i.e., the spectrum of a femtosecond laser is spatially separated by a DMD and recombined by an objective lens, forming a-depth resolved light sheet for laser micromachining and multi-photon imaging. Examples of high-throughput nanofabrication will be given in two-photon polymerization processes and 3-D metal printing, achieving a resolution of 500 nm. In terms of imaging, a new two-snapshot structured light illumination (SLI) reconstruction algorithm for fast image acquisition will be reported, demonstrating improved improved axial resolution on a temporal focusing fluorescence microscope.

Additive Manufacturing (or 3D printing) is increasingly used in the production of biomedical implants. In particular, there is a growing research interest in using additive manufacturing for producing scaffolds for advanced tissue engineering constructs. This interest originates from the possibility to build 3D structured scaffolds with a user-defined macrostructure and porosity. This enables the production of 3D objects with tuneable mechanical properties and pore sizes, and opens up the possibility to produce patient specific implants. This presentation will focus on my group's work in laser-based additive manufacturing of tissue engineering scaffolds and the use of these structures in healthcare applications for regeneration of both soft and hard tissues.

Adaptive3D is developing a materials, hardware, software and supply-chain solution for a leading auto manufacturers to rapidly produce high-volume, low-cost car seat cushions with internally complex micro-truss structures resulting in polymer parts with sufficient materials properties necessary to withstand the application demands of the automotive industry. Pursuit of this solution will mature the business and technology ecosystems necessary to deliver additively manufactured polymeric parts that compete on cost and time with injection molding and extrusion to produce lighter and stronger car seats for next-generation vehicles. The technology is built upon recent advances in photo-patterning thiol-based resins using massively scaled digital micromirror device systems. Our solution results in engineering foam-like structures with specific morphologies using custom software, iterating materials design, part design and equipment design in combination to arrive at a tailored manufacturing solution driven by patterned light.

Recent development of high-resolution Micro-Continuous Liquid Interface Production (microCLIP) process has enabled 3D printing of biomedical devices with 10 micron-scale precision. 3D bioresorbable vascular scaffolds (BVS) were printed using an antioxidant, photopolymerizable citric acid-based material (B-InkTM). To improve lingering challenges (material strength/stiffness), a PLLA nanophase (2wt.%) was dissolved and secondary initiators were added to the photopolymer resin to improve post-process polymerization. This greatly improved bulk material stiffness and yielded BVSs with 100um strut thickness that could sustain necessary biological radial pressure loadings. This technology and material is a large step forward toward on-the-spot and on-demand fabrication of patient specific BVSs.

This paper presents a holographic fabrication of a new type of photonic crystal, called graded photonic super-crystals with graded basis, dual period and dual symmetry. Pixel-by-pixel phase coding of laser beams in a spatial light modulator can produce the highest resolution in produced photonic super-lattice. Two-level designs in phase pattern are used to generate graded photonic super-crystals where graded square lattice clusters are orientated in four, five or six-fold symmetry. Further phase engineering in a super-cell of 12x8 pixels can produce small-period square lattice orientated in a large period rectangular pattern.

We demonstrate the manipulation of the propagation trace of an optical vortex with a symmetric Airy beam (SAB) through the combination of Lee method and a digital micromirror device (DMD), and found that the SAB experiences self-rotation with the implementation of a topological phase structure of coaxial vortex. Slight misalignment of the vortex and the SAB enables the guiding of the vortex into one of the self-accelerating channels. Multiple off-axis vortices embedded in SAB are also demonstrated to follow the trajectory of the major lobe for the SAB beam.

Single-pixel imaging employs structured illumination to record images with very simple light detectors, which can be very useful in certain applications. In this work, we compare the performance of a single-pixel camera using different alternative sampling functions. We employ binary, real as well as complex functions, all of them codified with a digital micromirror device (DMD) by using different approaches. In particular, in the study we use Hadamard, cosine, Fourier and noiselets functions. We perform both numerical and experimental tests with the different sampling functions and we compare the performance in terms of the efficiency and the signal-to-noise ratio (SNR).

Unit cell orientation information is encoded in electron diffraction patterns of crystalline materials. Traditional transmission electron detectors implemented in the scanning electron microscope are highly symmetric and are insensitive to in-plane unit cell orientation information. Herein we detail the implementation of a transmission electron detector that utilizes a digital micromirror array to select anisotropic portions of a diffraction pattern for imaging purposes. We demonstrate that this detector can be used to map the in-plane orientation of grains in two-dimensional materials. The described detector has the potential to replace and/or supplement conventional transmission electron detectors.

Digital fringe projection technique is a widely used for high-accuracy three-dimensional (3D) shape measurements. However, when the measured object is moving, there will be severe artifacts and phase errors induced by the movement even though a high-speed fringe projection system is used. Meanwhile, there are different kinds of motion such as uniform and non-uniform motion, which makes the problem even harder to address. To this end, this paper proposes motion error compensation algorithms to deal with different motion artifacts.

The paper proposes a new method for measuring the 3D shape of fast-moving highly reflective surfaces by using the transitioning state of the digital micro-mirror device (DMD). 1-bit binary patterns are used to achieve kHz measurement speeds and to overcome the rigid camera-projection synchronization requirement. More than one captures of the fringes are made during each cycle of the pattern projection. The dark period between DMD’s ON time is used to alleviate saturation problem of highly reflective surfaces. Experiments conducted will illustrate the success of the proposed method for high-speed applications.

Using DLP projection for 3D surface measurement is well established and very high speed recording has been reported using binary structured light patterns with 20 kHz update rates of DLP systems. Sinusoidal fringes are required for the advanced phase shifting technology in 3D measurement and the corresponding grayscale pattern generation is typically realized by time averaging of binary bit planes. This way, however, phase shifting sequence recording did not run at the 20 kHz rate. The paper describes a novel methodology for line projection with 10 bit grayscale profiles. The basic principle of this new approach has been verified analyzing phase shifting pattern sequences recorded by an integrated camera. More than twenty thousand of precise sinusoidal profiles per second have been generated and the maximum of 50 profiles per second is achieved with 8 bit grayscale. Experimental results are presented that are very promising for future use in 3D surface inspection.

This paper presents a structured light system with two single camera-projector pairs for multi-view 3D shape measurement in real time. A systematic calibration framework is proposed for high quality measurement, and a data fusion algorithm is developed to merge two patches of 3D geometry into one single surface with larger field of view. In addition, we develop a GPU-assisted multi-thread processing software to simultaneously capture, reconstruct and display 3D data in real time from two different views. Experimental results demonstrate that the proposed technique can achieve multi-view real-time 3D shape measurement.

As a scanning process, structured light illumination is not associated with real-time vision without the use of highly expensive light engines, even though commodity 3D-ready projectors are readily available. A significant technical hurdle for commodity projectors is the ability to synchronize the video of the projector to the camera since commercial graphics cards are not real-time systems. As such, we introduce a smart-camera architecture that incorporates HDMI video pass-through to perfectly synchronize camera capture so that any commodity HDMI projector can instantly be converted into an SLI light engine.

We propose a new system for kHz-level high-speed structured light profilometry using a digital micro-mirror device (DMD) based projector and two high-speed cameras. By correctly halftoning input images using an adaptive error diffusion algorithm and by using optical low-pass filtering, we obtained high-quality sinusoidal fringe patterns at the full speed of our DMD without mirror dithering. Working in tandem, the two high-speed cameras allowed a matched image acquisition speed to that of the structured light projector (6 kHz) at one-megapixel resolution. We have achieved profilometry of a rotating 3D object at a speed of 700 rounds per minute.

Many applications of 3D must deal with objects that are highly specularly reflective and have multiple interreflections, resulting in missing and erroneous points. We use a single camera with multiple DMD projectors, resulting in multiple incident lighting angles to deal with both specular and inter-reflections. We explore the effect of varying the number of projectors on the 3D measurement error rate and successfully recovered points. We use a maximum likelihood estimator to fuse together redundant depth measurements and neighboring depth information. Our multi-projector approach and estimator increases 3D measurement accuracy and recoverable points for complex reflective objects, while reducing error.

Structured light scanning is a 3D scanning process that involves projecting a series of striped patterns and reconstructing the 3D shape based on the warping of the patterns across the target object's surface. In a review of the associated literature, numerous variations on the basic camera-projector pair have been proposed including the use of multiple cameras or orthogonal projector pairs; however, one configuration that has not been thoroughly evaluated is the use of two arbitrarily positioned projectors, and as such, this paper takes a look at this novel arrangement, expounding on the configurations benefits related to multipath.

A high-speed and low latency projector DynaFlash that can display 8bit black and white images at up to 1,000 fps with 3 ms delay and a high-speed projector DynaFlash v2 that can display 8bit color images at 947 fps will be explained. Basic architectures of those systems, and devices as well as application systems including dynamic projection mapping onto moving objects and/or flexible objects and three dimensional shape recognition using structured light will be shown. Those systems use high-speed projectors and high-speed vision that can recognize three dimensional shape and/or position of moving objects every 1ms.

Matrix-LED systems offer different functionalities to increase road safety, e.g. glare-free high beam and marking light. Shortly after their introduction efforts to increase the quantity of pixels to implement high resolution pixel light systems were already underway. One approach is the use of LED arrays as a light source to illuminate a DLP. Unlike video projectors which require a homogeneous illumination of the DLP in order to obtain a homogeneous projection, headlamps require a hotspot centric distribution. In this paper, concepts for imaging and non-imaging illumination strategies of a DLP for high resolution headlamps will be introduced.

Adaptive headlamps with innovative lighting functionalities can increase traffic safety. Area-based light modulators such as DMD, LCD or LCoS are considered to be used as an implementation with a high resolution. In order to realize the regulated light distribution as well as to improve the optical efficiency and on-road projection quality of such headlamp systems, an inhomogeneous illumination on the modulator and whereafter lower distortion projection optics are considered. In this paper we present a methodical approach with which the necessary light distribution on the light modulator can be determined. Subsequently, optical concepts of inhomogeneous illumination for headlamps are described.

Electron beam halo and dark current from high repetition rate photoinjectors can be problematic for continuous wave super conduction linear accelerators. This contribution describes an adaptive coronagraph employing a DMD as a programmable occulting mask in an imaging relay.

Texas Instruments’ digital mirror device (DMD) uses thousands to millions of individual mirror pixels to direct light as a Spatial Light Modulator (SLM). The Tilt-Roll-Pixel (TRP) is currently the smallest DLP pixel node at 5.4μm. The small pixel size, which enables fast switching speed, and precise tilt angles, exploits this speed on a system level to double or quadruple the resolution by using super resolution projection. Super resolution projection overlays multiple low-resolution images, each shifted on the screen by a fraction of a pixel, and as long as the shifting occurs at a rate faster than the critical flicker fusion threshold, the human visual system will act as a temporal low pass filter and naturally integrate all low resolution SLM images into a single super resolution result. This presentation will discuss the operation of the TRP node, how this node can be operated more quickly, how super resolution projection works, and how we modified the optical architecture to leverage the switching speed for super resolution projection.

A development test chip of a Phase Spatial Light Modulator (PLM) device has been developed and demonstrated using DLP MEMS based technology. Designed for a red HeNe laser, this device uses an array of individually-addressable, digitally-controlled plate actuators that can be addressed multiple discrete vertical positions. The MEMS superstructure process flow used for DLP Micromirrors was adapted to enable manufacturing this device on top of an existing DLP Technology CMOS device. The test chips have demonstrated good uniformity across the array and the ability to steer light using phase light modulation. We will discuss some initial performance metrics as well as potential applications.

The authors have developed a small, rugged, in-line, broadband Near Infrared Spectrometer (NIRS) for industrial process control. Their solution leverages two proven optical innovations: the Digital Micromirror Device (DMD) and the High-Throughput Virtual Slit (HTVS). This paper documents the development of a next-generation programmable NIRS designed specifically for long-term, unattended, industrial monitoring of manufactured goods. Included are: 1) a brief discussion on the sensing requirements for tomorrow’s Smart Factory; 2) a detailed discussion on the design requirements of their programmable NIRS; and 3) a high-level review of the system attributes and industrial data, enabled by the combination of the DMD and the HTVS.

For the space-borne platforms with limited payload capacity such as the emerging CubeSat, the substantial amount of raw data acquired by hyperspectral imaging sensors may overwhelm the data transmission capacity between the spacecraft and the ground station. In this paper, we will present a new hyperspectral imaging sensor that is based on the compressive line sensing imaging concept. The initial simulation results are compared with the state-of-the-art solutions, and the experimental results obtained from the prototype using a TI DLP NIRscan™ Nano Evaluation Module are discussed.

Digital micromirror devices (DMDs) are well suited for highly multiplexed spectroscopy applications. In astronomy, DMDs can be used as object selectors in multi-object spectrometers (MOS). Moreover, DMDs are the only current alternative to microshutter arrays for space-based multi-object spectrometers. Over the past several years, we have carried out a program to evaluate the viability of XGA DMDs for operation in space, including their ability to survive the launch environment. The DMDs we tested did not show any failures or adverse effects after mechanical vibration and shock testing (which followed the NASA General Environment Variables Specification). Using heavy ion irradation, we found that DMDs are susceptible to single event upsets (SEUs), though all SEUs are non-destructive and can be cleared by loading a new pattern. The estimated SEU rate for "worst week" conditions in interplanetary space was 5 upset micromirrors (out of 786,432) per 24 hours. Using high energy protons, we found that DMDs started to show failures at a total ionizing dose of 20 krad (which is well above a 4 times safety margin of 8 krad estimated for a 4 year mission). In this work, we present the total ionizing dose testing performed using a gamma rays from a Co-60 source, at NASA GSFC. We tested 15 XGA devices and found that individual micromirrors began failing after the devices accumulated a total dose of 17-18 krad. Six of the devices completely recovered after a 24 hour anneal at room temperature, while the rest showed varying degrees of adverse effects. We also tested unbiased (powered off) devices, which showed no effects up to a dose of 50 krad (which is the highest TID we achieved during our testing). This work concludes our efforts to space-qualify XGA DMDs, and shows that these devices are well-suited for deployment in space, except in the harshest radiation environments.