Djamel Brahmi

Analysis of parametric images derived from genomic sequences using neural network based approaches

The exploration of DNA genomic huge sequences (up to several megabases) needs new kind of data representation allowing robust analyses. With the help of the chaos game representation method (CGR), fractal images can be generated, which allow to observe, at a glance, frequencies of words (small sequences of the four bases: G, A, T, C) in DNA sequences. Classification of CGR images and extraction of main features are the issues addressed in this work, using a classical statistical analysis (principal component analysis) and neural networks grounded on curvilinear component analysis algorithm and Kohonen map.

Fast MTF and aberrations analysis of MWIR and LWIR imaging systems using quadri wave interferometry

We present the application of Quadri-Wave Lateral Shearing Interferometry (QWLSI), a wave front sensing technique, to characterize optical beams at infrared wavelengths from 2 to 16μm with a single instrument. We apply this technique to qualify optical systems dedicated to MWIR (λ within 3 and 5μm) and LWIR (λ within 8 and 14μm) wavelength ranges. The QWLSI offers the crucial advantage that it yields an analyzed wave front without the use of a reference arm and consequent time consuming alignment. The qualification of an optical system with QWLSI gives a complete diagnostic, from the aberration cartography to the PSF and MTF curves for every direction in one single measurement. In this paper, we first present the QWLSI technology and its main features, we also detail an experimental comparison between our MTF measurement and the results given by a classical MTF test bench. We finally show the experimental analysis of an infrared lens at two different wavelengths, one in the MWIR range (λ=3.39μm) and the other in the LWIR range (λ=10.6μm).

A Learning by Sample Approach for the Detection of Features in Medical Images.

The detection of specific features in medical images is often a key support for diagnosis. Taking advantage of large bases of images where features of interest have been localized by clinicians, a modular system has been developed to spot similar features on new images. Images are scanned through a window, the size of which being previously fitted to the feature of interest. The recognition process involves a coding phase followed by a classification phase. These phases rely on unsupervised and supervised learning respectively for their implementation. Applications in Dermatology and Ophthalmology are presented.

Application of curvilinear component analysis to chaos game representation images of genome

Curvilinear component analysis (CCA) is performed by an original self-organized neural network, which provides a convenient approach for dimension reduction and data exploration. It consists in a non-linear, preserving distances projection of a set of quantizing vectors describing the input space. The CCA technique is applied to the analysis of CGR fractal images of DNA sequences from different species. The CGR method produces images where pixels represent frequency of small sequences of bases revealing nested patterns in DNA sequences.

Evaluation of a Neural Network Classifier for Detection of Microcalcifications and Opacities in Digital Mammograms

A computer-aided diagnosis platform has been developed in our laboratory for automatic detection of microcalcifications and opacities in mammograms. A complete description of the method has been given in previous articles [1]–[3]. In the present work, we report on the evaluation of results which were obtained for two mammography databases.

On-axis and off-axis characterization of MWIR and LWIR imaging systems using quadri-wave interferometry

The Quadri-Wave Lateral Shearing Interferometry (QWLSI) is an innovative wave front sensing technique that is commercially available for MWIR and LWIR applications. We present this technology and its application to the metrology, on and off-axis, of infrared imaging systems. The bench is only composed of a collimated reference beam that creates a source point at infinity, the objective to analyze and the sensor placed a few millimeters after the focal spot. Thanks to this direct measurement configuration, the alignment process is very simple and fast. A complete characterization (aberrations, MTF, field curvature) for several field points is possible within a few minutes.

MWIR and LWIR wavefront sensing with quadri-wave lateral shearing interferometry

We present the application of Quadri-Wave Lateral Shearing Interferometry (QWLSI), a wave front sensing technique, to characterize optical beams at infrared wavelengths from 2 to 16μm with a single instrument. This technique can be used to quantify the quality of optical systems (like thermal infrared lenses) by measuring their aberrations. It can also evaluate laser sources in the infrared range like some gas lasers (HeNe laser or CO2 laser), infrared Optical Parametric Oscillator laser sources or Quantum Cascade Laser sources. In all cases, QWLSI offers the crucial advantage that it yields an analyzed wave front without the use of a reference arm and consequent time consuming alignment. In this paper, we first present the single interferometer which can be used with wavelength within 2 and 16μm, covering this way the MWIR (λ within 3 and 5μm) and LWIR (λ within 8 and 14μm) ranges. We then present the characterization of two gas lasers: an infrared HeNe lasers (λ=3.39μm) and a CO2 laser (λ=10.6μm) with this instrument. We finally show the experimental analysis of an infrared lens at two different wavelengths, one in the MWIR range (λ=3.39μm) and the other in the LWIR range (λ=10.6μm). λ

High numerical aperture focal spot diagnostic and optimization

Adaptive optics loops are now widely implemented on high power laser chains and several industrial solutions exist which are optimized for their specific constrains. However the main goal for adaptive optics loop implementation is to optimize the interaction between laser and matter (gas or solid). In particular, one parameter to optimize for loop convergence is the intensity at the focus of the off-axis parabola, which is the final optics seen by the laser. However it is in practice not possible to place any diagnostic directly in the beam in the interaction chamber. This is why the solution consists in creating a diagnostic beam from a beam pick-up somewhere upstream the focusing optics. The deformable mirror is then driven from the diagnostic beam wave front.

Wave Front Sensor Test Bench for Optical Design Prototyping and Manufacture Check in Production

Optical design produces complex systems with non spherical diopters. We propose a process that accurately compares wave front measurement to optical design. We present experimental results on calibrated aberrated lenses and prospects on freeform optics.

Thermal Infrared Optical Metrology using Quadri-Wave Lateral Shearing Interferometry

We present the application of Quadri-Wave Lateral Shearing Interferometry (QWLSI), a wave front sensing technique, to characterize thermal infrared lenses for wavelengths within 8 and 14μm. Wave front sensing is not only a tool to quantify optical quality, but also to map the local (dust, scratches) or global possible defects. This method offers the crucial advantage that it yields an analyzed wave front without the use of a reference arm and consequent time consuming alignment. Moreover thanks to the acceptance of QWLSI to high numerical aperture beams, no additional optics is required. This makes lens characterization convenient and very fast. We particularly show the experimental characterization of single Germanium lens and finally present the characterization of complex optical imaging systems for high-performance infrared cameras. The analysis is made in conditions that are very close to the usual conditions of the camera use; that is to say, directly in the convergent beam and in polychromatic (black body) light.