System, method and apparatus for retinal absorption phase and dark field imaging with oblique illumination

A method for imaging a tissue of an eye, the method including the steps of providing oblique illumination to the eye by a plurality of light emitting areas of a light delivery device, the plurality of light emitting areas being independently controllable and arranged to direct light towards at least one of a retina and an iris of the eye, causing an output beam from light backscattered from the at least one of the retina and the iris by the oblique illumination, capturing the output beam with an imaging system to provide a sequence of images of a fundus of the eye, and retrieving a phase and absorption contrast image from the sequence of images of the fundus, wherein the sequence of images of the fundus of the step of capturing is obtained by sequentially turning on one or more of the plurality of light emitting areas at a time in the step of providing the oblique illumination.

Realization and study of InAs/GaAs quantum dot superluminescent diodes emitting at 1.3µm

Marco Rossetti

A quantum dot is a semiconductor nanostructure that confines the motion of conduction band electrons and valence band holes in all three spatial directions, thus creating fully discrete energy levels. The confinement in the InAs/GaAs material system is generated by the bandgap difference for the two materials (∼ 0.4 eV for InAs, and ∼ 1.4 eV for GaAs) which provides a minimum of potential energy for both electrons and holes inside the InAs nanoparticle. The appearance of new nanoscale effects modifies the electro-optical properties of such a system which makes possible the improvement of existing device performance or even fascinating novel device applications. For the creation of coherent structures with very high purity and high density, as it is required for application in light emitting devices, self-assembling is a very important pathway. The quantum dot creation mostly relies on the relaxation of the strain accumulated during the growth of lattice mismatched materials, which happens in a non-equlibrium condition. This results in a statistical distribution of all the parameters characterizing the InAs nanocrystals as size, composition and strain, whose direct manifestation is the inhomogeneously broadened light emission deriving from quantum dot ensembles. At the dawning of the quantum dot development the broadening was considered as a limiting factor for application in semiconductor lasers. Today, it is considered favorably for all the applications where a broad gain spectrum is required as optical amplifiers, monolithic and external-cavity tunable lasers, and superluminescent diodes. The objective of this thesis is the application of InAs/GaAs self-assembled quantum dots to the active region of high power and broad-band superluminescent diodes emitting in the 1.3 μm wavelength region. Superluminescent diodes are edge-emitting semiconductor light sources based on the amplification of spontaneous emission along a waveguide where optical gain is achieved through current injection. They combine the high power and brightness of laser diodes with the low coherence of LEDs. The latter is a fundamental feature for the application in optical coherence tomography medical imaging, where the image resolution is related to the spectral bandwidth of the light source used. The achievement of bandwidths larger than 100 nm are demonstrated in this work through the optimization of molecular beam epitaxy growth conditions. This may be done optimizing the statistical size-dispersion of the dot ensemble, or through the use of different dot layers emitting at slightly different wavelengths. The large bandwidth emission is obtained also due to the peculiar energy structure of this material, for which a multi-state emission appears under particular injection conditions. As a term of comparison, best commercial single-chip superluminescent diodes for the imaging application at 1.3 μm show bandwidths around 60 nm. Moreover, the discrete nature of the energy levels in InAs quantum dots together with the implementation of a GaAs-based technology can be beneficial for the device temperature stability (the competing technology, based on InP, suffers from strong thermal degradation). The analysis and understanding of the device temperature characteristics will be detailed in the last chapter of the manuscript. Even though standard devices show an important temperature dependence, a better stability may be achieved through the use of p-doped active regions. In the last few years quantum dot devices have shown outstanding properties if compared to the competing quantum well technology. Low threshold currents, high temperature stability and low chirp in lasers, large bandwidths and low gain saturation in amplifiers have been demonstrated. However, in spite of the large interest among the scientific community, many of the microscopic processes at the origin of the electro-optical characteristics of this material are not completely understood yet. The physics of the quantum dot active material, will be modeled in this work through the use of systems of rate equations with different complexity depending on the system they are applied to. We will provide a mean-field rate equation model allowing to get insights about carrier dynamics in QD lasers. Also, we will develop two different traveling-wave rate equation models, one for the modeling of the L-I characteristics and the other for the modeling of the spectral characteristics of quantum dot superluminescent diodes. Considering the carrier and photon density distributions across the device cavity is essential in superluminescent diodes, where the high single pass gain results in large non-homogeneities of photon and carrier distributions. The work is motivated by an industrial cooperation with EXALOS AG, a leading company in the development, manufacturing, and sales of broadband superluminescent diodes.

EPFL2007

Design and characterization of a CMOS 3-D image sensor based on single photon avalanche diodes

Pierre-André Besse, Edoardo Charbon, Cristiano Niclass, Alexis Rochas

The design and characterization of an imaging system is presented for depth information capture of arbitrary three-dimensional (3-D) objects. The core of the system is an array of 32 × 32 rangefinding pixels that independently measure the time-of-flight of a ray of light as it is reflected back from the objects in a scene. A single cone of pulsed laser light illuminates the scene, thus no complex mechanical scanning or expensive optical equipment are needed. Millimetric depth accuracies can be reached thanks to the rangefinder's optical detectors that enable picosecond time discrimination. The detectors, based on a single photon avalanche diode operating in Geiger mode, utilize avalanche multiplication to enhance light detection. On-pixel high-speed electrical amplification can therefore be eliminated, thus greatly simplifying the array and potentially reducing its power dissipation. Optical power requirements on the light source can also be significantly relaxed, due to the array's sensitivity to single photon events. A number of standard performance measurements, conducted on the imager, are discussed in the paper. The 3-D imaging system was also tested on real 3-D subjects, including human facial models, demonstrating the suitability of the approach.

2005

Application of 3D range camera in virtual human-computer interfaces

Huan Du

In this work, we propose new ways to employ 3D-ranging systems in advanced human-computer interfaces and show that they can be used for precise hand-tracking systems aiming at virtual keyboard or mouse applications. We first implement a Structured Light (SL) based 3D-ranging system as the testbed for acquiring depth information. The ranging system uses an off-the-shelf projector combined with an industrial color camera, where the projector projects well-designed line patterns onto the scene and the camera captures the reflected light from the scene and the target. The depth information is recovered from the deformed line pattern and the pre-calibrated projector-camera system. With this system we can capture dense depth maps with moderate resolution and high frame rate as the simulated input to the hand-tracking system. We then explore an approach to 3D hand matching that combines a detailed 3D hand model (skeletal + polygonal model) with a Physical-based Model Fitting technique (PMF). Virtual forces are assigned between the depth measurements and the polygonal vertices of the model. The matching problem is formulated as the minimization of the distances between the measurement and the model surface with the guidance of virtual forces. This can employ 3D information simply and directly onto the parameter space, and yield good results under challenging conditions. To demonstrate the effectiveness of the approach we show that we can track hand motions such as typing or grasping frame by frame. The major limitation of this method is that it requires a good initialization of the hand model and the computation time to achieve convergence with good fitting results is quite long. We therefore investigated more complex hand models and feature-based matching techniques. A three-level hand model (skeletal + primitive + polygonal model) is proposed for the Multi-Level Feature Matching approach (MLFM). The matching is carried out from the bottom to top levels of different hand presentations, and distinctive features for each level extracted from both the measurements and the model are matched with Scaled Conjugate Gradient method (SCG). The matching result from lower level is used as the initial state of the model parameters for higher level matching. This hierarchical structure speeds up the optimization process and enhances the precision of the matching result. We demonstrate our approach with the same hand motion as in the previous method and compare the performance of these two methods. We show that the MLFM method is faster and more robust to hand occlusions than the PMF method. Finally, we present a more powerful approach based on the Back-Constrained Gaussian Process Latent Variable Model (BC-GPLVM) that import prior information of categorized hand motion such as typing. BC-GPLVM provides a low-dimensional embedding of hand motion data, with a density function that gives higher probability to gestures and motions close to normal typing motion. This model is learned from a small set of clusters of vast captured motion data. The prior model is then combined with the previous multi-level models and features to generate a uniform cost function for optimization. We show the tracking result to demonstrate that this method can be used for human-computer interfaces requiring precise 3D hand tracking.