Casamento de modelos baseado em projeções radiais e circulares invariante a pontos de vista.

Este trabalho aborda o problema de casamento entre duas imagens. Casamento de imagens pode ser do tipo casamento de modelos (template matching) ou casamento de pontos-chaves (keypoint matching). Estes algoritmos localizam uma região da primeira imagem numa segunda imagem. Nosso grupo desenvolveu dois algoritmos de casamento de modelos invariante por rotação, escala e translação denominados Ciratefi (Circula, radial and template matchings filter) e Forapro (Fourier coefficients of radial and circular projection). As características positivas destes algoritmos são a invariância a mudanças de brilho/contraste e robustez a padrões repetitivos. Na primeira parte desta tese, tornamos Ciratefi invariante a transformações afins, obtendo Aciratefi (Affine-ciratefi). Construímos um banco de imagens para comparar este algoritmo com Asift (Affine-scale invariant feature transform) e Aforapro (Affine-forapro). Asift é considerado atualmente o melhor algoritmo de casamento de imagens invariante afim, e Aforapro foi proposto em nossa dissertação de mestrado. Nossos resultados sugerem que Aciratefi supera Asift na presença combinada de padrões repetitivos, mudanças de brilho/contraste e mudanças de pontos de vista. Na segunda parte desta tese, construímos um algoritmo para filtrar casamentos de pontos-chaves, baseado num conceito que denominamos de coerência geométrica. Aplicamos esta filtragem no bem-conhecido algoritmo Sift (scale invariant feature transform), base do Asift. Avaliamos a nossa proposta no banco de imagens de Mikolajczyk. As taxas de erro obtidas são significativamente menores que as do Sift original.

The impact of spectral composition and light periodicity on the activity of two antioxidant enzymes (SOD and CAT) in the coral Favia favus

In oligotrophic waters the light spectrum is mostly blue, and therefore the physiological and biochemical responses to blue light occurring in the coral tissue and in the symbiotic algae are important. Examination of the wavelength dependence of two free radical scavenger enzyme activity revealed an increase in activity in the blue light range (440-480 nm) compared to the red (640680 nm) in the full visible light (400-700 nm) range. These data show for the first time the relationship between the action spectra of photosynthesis and the activity of two main antioxidant enzymes in the symbiotic coral Favia favus. It was found that in the animal (host) the enzyme response to the spectral distribution of light was higher than that of the zooxanthellae, probably due to accumulation of free radicals within the host tissue. Furthermore, we found that the activity of these enzymes is affected in nature by the length of the day and night, and in the laboratory, by the duration of the illumination. Changes in the pigment concentrations were also observed in response to growth under the blue region and the whole PAR spectrum, while fluorescence measurements with the fast repetition rate fluorometer (FRRF) showed a decrease in the sigma cross section and a decrease in the quantum yield also in the blue part of the spectrum. These changes of scavenger enzymes activity, pigment concentration and fluorescence yield at different light spectra are vital in acclimatization and survival of corals in shallow water environments with high light radiation. (c) 2005 Elsevier B.V. All rights reserved.

Practical 3D reconstruction based on photometric stereo

Photometric Stereo is a powerful image based 3D reconstruction technique that has recently been used to obtain very high quality reconstructions. However, in its classic form, Photometric Stereo suffers from two main limitations: Firstly, one needs to obtain images of the 3D scene under multiple different illuminations. As a result the 3D scene needs to remain static during illumination changes, which prohibits the reconstruction of deforming objects. Secondly, the images obtained must be from a single viewpoint. This leads to depth-map based 2.5 reconstructions, instead of full 3D surfaces. The aim of this Chapter is to show how these limitations can be alleviated, leading to the derivation of two practical 3D acquisition systems: The first one, based on the powerful Coloured Light Photometric Stereo method can be used to reconstruct moving objects such as cloth or human faces. The second, permits the complete 3D reconstruction of challenging objects such as porcelain vases. In addition to algorithmic details, the Chapter pays attention to practical issues such as setup calibration, detection and correction of self and cast shadows. We provide several evaluation experiments as well as reconstruction results. © 2010 Springer-Verlag Berlin Heidelberg.

Modelling human behaviour and event detection from videos

Automatic analysis of human behaviour in large collections of videos is gaining interest, even more so with the advent of file sharing sites such as YouTube. However, challenges still exist owing to several factors such as inter- and intra-class variations, cluttered backgrounds, occlusion, camera motion, scale, view and illumination changes. This research focuses on modelling human behaviour for action recognition in videos. The developed techniques are validated on large scale benchmark datasets and applied on real-world scenarios such as soccer videos. Three major contributions are made. The first contribution is in the area of proper choice of a feature representation for videos. This involved a study of state-of-the-art techniques for action recognition, feature extraction processing and dimensional reduction techniques so as to yield the best performance with optimal computational requirements. Secondly, temporal modelling of human behaviour is performed. This involved frequency analysis and temporal integration of local information in the video frames to yield a temporal feature vector. Current practices mostly average the frame information over an entire video and neglect the temporal order. Lastly, the proposed framework is applied and further adapted to real-world scenario such as soccer videos. A dataset consisting of video sequences depicting events of players falling is created from actual match data to this end and used to experimentally evaluate the proposed framework.

What is second-order vision for? Discriminating illumination versus material changes

The human visual system is sensitive to second-order modulations of the local contrast (CM) or amplitude (AM) of a carrier signal. Second-order cues are detected independently of first-order luminance signals; however, it is not clear why vision should benet from second-order sensitivity. Analysis of the first-and second-order contents of natural images suggests that these cues tend to occur together, but their phase relationship varies. We have shown that in-phase combinations of LM and AM are perceived as a shaded corrugated surface whereas the anti-phase combination can be seen as corrugated when presented alone or as a flat material change when presented in a plaid containing the in-phase cue. We now extend these findings using new stimulus types and a novel haptic matching task. We also introduce a computational model based on initially separate first-and second-order channels that are combined within orientation and subsequently across orientation to produce a shading signal. Contrast gain control allows the LM + AM cue to suppress responses to the LM-AM when presented in a plaid. Thus, the model sees LM -AM as flat in these circumstances. We conclude that second-order vision plays a key role in disambiguating the origin of luminance changes within an image. © ARVO.

The post-illumination pupil response of melanopsin expressing intrinsically photosensitive retinal ganglion cells in diabetes

Purpose: This study investigates the clinical utility of the melanopsin expressing intrinsically photosensitive retinal ganglion cell (ipRGC) controlled post-illumination pupil response (PIPR) as a novel technique for documenting inner retinal function in patients with Type II diabetes without diabetic retinopathy. Methods: The post-illumination pupil response (PIPR) was measured in seven patients with Type II diabetes, normal retinal nerve fiber thickness and no diabetic retinopathy. A 488 nm and 610 nm, 7.15º diameter stimulus was presented in Maxwellian view to the right eye and the left consensual pupil light reflex was recorded. Results: The group data for the blue PIPR (488 nm) identified a trend of reduced ipRGC function in patients with diabetes with no retinopathy. The transient pupil constriction was lower on average in the diabetic group. The relationship between duration of diabetes and the blue PIPR amplitude was linear, suggesting that ipRGC function decreases with increasing diabetes duration. Conclusion: This is the first report to show that the ipRGC controlled post-illumination pupil response may have clinical applications as a non-invasive technique for determining progression of inner neuroretinal changes in patients with diabetes before they are ophthalmoscopically or anatomically evident. The lower transient pupil constriction amplitude indicates that outer retinal photoreceptor inputs to the pupil light reflex may also be affected in diabetes.

Methods for the illumination of multilevel buildings with vertical light pipes

This paper examines the feasibility of using vertical light pipes to naturally illuminate the central core of a multilevel building not reached by window light. The challenges addressed were finding a method to extract and distribute equal amounts of light at each level and designing collectors to improve the effectiveness of vertical light pipes in delivering low elevation sunlight to the interior. Extraction was achieved by inserting partially reflecting cones within transparent sections of the pipes at each floor level. Theory was formulated to estimate the partial reflectance necessary to provide equal light extraction at each level. Designs for daylight collectors formed from laser cut panels tilted above the light pipe were developed and the benefits and limitations of static collectors as opposed to collectors that follow the sun azimuth investigated. Performance was assessed with both basic and detailed mathematical simulation and by observations made with a five level model building under clear sky conditions.

The Post-Illumination Pupil Response (PIPR)

Purpose The post-illumination pupil response (PIPR) has been quantified using four metrics, but the spectral sensitivity of only one is known; here we determine the other three. To optimize the human PIPR measurement, we determine the protocol producing the largest PIPR, the duration of the PIPR, and the metric(s) with the lowest coefficient of variation. Methods The consensual pupil light reflex (PLR) was measured with a Maxwellian view pupillometer. - Experiment 1: Spectral sensitivity of four PIPR metrics [plateau, 6 s, area under curve (AUC) early and late recovery] was determined from a criterion PIPR to a 1s pulse and fitted with Vitamin A1 nomogram (λmax = 482nm). - Experiment 2: The PLR was measured as a function of three stimulus durations (1s, 10s, 30s), five irradiances spanning low to high melanopsin excitation levels (retinal irradiance: 9.8 to 14.8 log quanta.cm-2.s-1), and two wavelengths, one with high (465nm) and one with low (637nm) melanopsin excitation. Intra and inter-individual coefficients of variation (CV) were calculated. Results The melanopsin (opn4) photopigment nomogram adequately describes the spectral sensitivity of all four PIPR metrics. The PIPR amplitude was largest with 1s short wavelength pulses (≥ 12.8 log quanta.cm-2.s-1). The plateau and 6s PIPR showed the least intra and inter-individual CV (≤ 0.2). The maximum duration of the sustained PIPR was 83.0±48.0s (mean±SD) for 1s pulses and 180.1±106.2s for 30s pulses (465nm; 14.8 log quanta.cm-2.s-1). Conclusions All current PIPR metrics provide a direct measure of the intrinsic melanopsin photoresponse. To measure progressive changes in melanopsin function in disease, we recommend that the PIPR be measured using short duration pulses (e.g., ≤ 1s) with high melanopsin excitation and analyzed with plateau and/or 6s metrics. Our PIPR duration data provide a baseline for the selection of inter-stimulus intervals between consecutive pupil testing sequences.

The post-illumination pupil response of melanopsin expressing retinal ganglion cells

Purpose The post-illumination pupil response (PIPR) has been quantified in the literature by four metrics. The spectral sensitivity of only one metric is known and this study quantifies the other three. To optimize the measurement of the PIPR in humans, we also determine the stimulus protocol producing the largest PIPR, the duration of the PIPR, and the metric(s) with the lowest coefficient of variation. Methods The consensual pupil light reflex (PLR) was measured with a Maxwellian view pupillometer (35.6° diameter stimulus). - Experiment 1: Spectral sensitivity of four PIPR metrics [plateau, 6 s, area under curve (AUC) early and late recovery] was determined from a criterion PIPR (n = 2 participants) to a 1 s pulse at five wavelengths (409-592nm) and fitted with Vitamin A nomogram (ƛmax = 482 nm). - Experiment 2: The PLR was measured in five healthy participants [29 to 42 years (mean = 32.6 years)] as a function of three stimulus durations (1 s, 10 s, 30 s), five irradiances spanning low to high melanopsin excitation levels (retinal irradiance: 9.8 to 14.8 log quanta.cm-2.s-1), and two wavelengths, one with high (465 nm) and one with low (637 nm) melanopsin excitation. Intra and inter-individual coefficients of variation (CV) were calculated. Results The melanopsin (opn4) photopigment nomogram adequately described the spectral sensitivity derived from all four PIPR metrics. The largest PIPR amplitude was observed with 1 s short wavelength pulses (retinal irradiance ≥ 12.8 log quanta.cm-2.s-1). Of the 4 PIPR metrics, the plateau and 6 s PIPR showed the least intra and inter-individual CV (≤ 0.2). The maximum duration of the sustained PIPR was 83.4 ± 48.0 s (mean ± SD) for 1 s pulses and 180.1 ± 106.2 s for 30 s pulses (465 nm; 14.8 log quanta.cm-2.s-1). Conclusions All current PIPR metrics provide a direct measure of intrinsic melanopsin retinal ganglion cell function. To measure progressive changes in melanopsin function in disease, we recommend that the intrinsic melanopsin response should be measured using a 1 s pulse with high melanopsin excitation and the PIPR should be analyzed with the plateau and/or 6 s metrics. That the PIPR can have a sustained constriction for as long as 3 minutes, our PIPR duration data provide a baseline for the selection of inter-stimulus intervals between consecutive pupil testing sequences.

Light- and cytokinin-induced changes in the levels of leucine and tyrosine isoaccepting tRNA species and modified nucleotide contents of total tRNA in cucumber seedlings

We have determined relative levels of chloroplast leucine and tyrosine isoaccepting tRNAs and modified nucleotide contents from total tRNAs isolated from dark-grown, light-grown, N6-isopentenyladenine (i6A)-treated dark-grown and i6A-treated light-grown cucumber seedlings. Significant increases in the relative amounts of tRNA(Leu)2 and tRNA(Leu)3 were observed in the i6A-treated dark-grown seedlings compared to dark-grown, light-grown and i6A-treated light-grown seedlings. On the other hand, i6A-treated light-grown seedlings tRNA(Tyr)1 increased to 85% of total tRNAs(Tyr) from about 9% in light-grown seedlings and tRNA(Tyr)2 decreased to 15% compared with 91% in light-grown seedlings. Analysis of modified nucleotide of total tRNAs indicated that pT, pI, pm1A, pm5C, pGm, pm1G, pm2G and pm7G contents were significantly higher in the total tRNA of i6A-treated dark-grown seedlings than those from untreated dark-grown seedlings. Illumination of 8-day-old dark-grown seedlings for 12 h increased the contents of pT, pI, pGm and pm1G when compared to 8-day-old dark-grown seedlings with extended growth for 12 h in dark. On the contrary, i6A had no stimulatory effect in the contents of modified nucleotide in the light-grown seedlings.

Photo induced optical changes in Sb/As2S3 multilayered film and (As2S3)(0.93)Sb-0.07 film of equal thickness

The increase in optical band gap (photo bleaching) due to light illumination was studied at room temperature as well as at low (4.2 K) temperature for Sb/As2S3 multilayered film of 640 nm thickness by Fourier Transform Infrared Technique. The interdiffusion of Sb into As2S3 matrix results the formation of Sb-As2S3 ternary solid solutions which is explained by the change in optical band gap (E-g), absorption coefficient (alpha), Tauc parameter (B-1/2), Urbach edge (E-e). At the same time, photo darkening phenomena was observed in (As2S3)(0.93)Sb-0.07 film of same thickness both at low and room temperatures. From our X-ray Photoelectron Spectroscopy measurements,we are able to show that some of the As-As, S-S and Sb-Sb bonds are converted into As-S and S-Sb bonds in case of multilayers. We found that the energetically favoured heteropolar bond formation take place by a phonon-assisted mechanism using the lone pair pi electrons of S-2(0). But in case of (As2S3)(0.93)Sb-0.02 film, the homopolar bonds are playing a major role. (C) 2010 Elsevier B.V. All rights reserved.

Photo-induced optical bleaching in Ge12Sb25S63 amorphous chalcogenide thin films: effect of 532 nm laser illumination

The photo-induced effects of Ge12Sb25S63 films illuminated with 532 nm laser light are investigated from transmission spectra measured by FTIR spectroscopy. The material exhibits photo-bleaching (PB) when exposed to band gap light for a prolonged time in a vacuum. The PB is ascribed to structural changes inside the film as well as surface photooxidation. The amorphous nature of thin films was detected by x-ray diffraction. The chemical composition of the deposited thin films was examined by energy dispersive x-ray analysis (EDAX). The refractive indices of the films were obtained from the transmission spectra based on an inverse synthesis method and the optical band gaps were derived from optical absorption spectra using the Tauc plot. The dispersion of the refractive index is discussed in terms of the single-oscillator Wemple-DiDomenico model. It was found that the mechanism of the optical absorption follows the rule of the allowed non-direct transition. Raman and x-ray photoelectron spectra (XPS) were measured and decomposed into several peaks that correspond to the different structural units which support the optical changes.

Stability of GaAs photocathodes under different intensities of illumination

The photocurrent curves of reflection-mode GaAs photocathodes as a function of time, when were illuminated by white light with an intensity of 0, 33 and 100 Ix, respectively, were measured using a multi-information measurement system. The calculated lifetimes of cathodes are 320, 160 and 75 min, respectively, showing that the stability of cathodes degraded with the increase of light intensity. The lifetime of cathode, illuminated by white light with an intensity of 100 Ix, while no photocurrent was being drawn during the illumination, was 100 min. Through comparison, we found that the influence of illumination on cathodes stability is greater than that of photocurrent. The quantum-yield curves of cathodes as a functions of time, when illuminated by white light with an intensity of 33 Ix, were measured also. The measured results show that the shape of the yield curves changes with increasing illumination time due to the faster quantum-yield degradation rate of low energy photons. Based on the revised quantum-efficiency equations for the reflection-mode cathodes, the variation of yield curves are analyzed to be due to the intervalley diffusion of photoelectrons and the evolution of the surface potential barrier profile of the photocathodes during degradation process.

PHOTOINDUCED CHANGES OF HYDROGEN-BONDING IN SEMIINSULATING IRON-DOPED INDIUM-PHOSPHIDE

After illumination with 1-1.3 eV photons during cooling-down, metastable PH modes are observed by IR absorption at 5 K in semi-insulating InP:Fe. They correlate with the photo-injection of holes, but not with a change of the charge state of the K-related centres present at equilibrium. They are explained by a change of the bonding of H, induced by hole trapping, from IR-inactive centres to PH-containing centres, stable only below 80 K. One metastable centre has well-defined geometrical parameters and the other one could be located in a region near from the interface with (Fe,P) precipitates.

Skin Color-Based Video Segmentation under Time-Varying Illumination

A novel approach for real-time skin segmentation in video sequences is described. The approach enables reliable skin segmentation despite wide variation in illumination during tracking. An explicit second order Markov model is used to predict evolution of the skin-color (HSV) histogram over time. Histograms are dynamically updated based on feedback from the current segmentation and predictions of the Markov model. The evolution of the skin-color distribution at each frame is parameterized by translation, scaling and rotation in color space. Consequent changes in geometric parameterization of the distribution are propagated by warping and resampling the histogram. The parameters of the discrete-time dynamic Markov model are estimated using Maximum Likelihood Estimation, and also evolve over time. The accuracy of the new dynamic skin color segmentation algorithm is compared to that obtained via a static color model. Segmentation accuracy is evaluated using labeled ground-truth video sequences taken from staged experiments and popular movies. An overall increase in segmentation accuracy of up to 24% is observed in 17 out of 21 test sequences. In all but one case the skin-color classification rates for our system were higher, with background classification rates comparable to those of the static segmentation.