Inhomogeneous Illumination Research Articles

3D Monte Carlo simulation of light distribution in mouse brain in quantitative photoacoustic computed tomography.

Photoacoustic computed tomography (PACT) detects light-induced ultrasound (US) waves to reconstruct the optical absorption contrast of the biological tissues. Due to its relatively deep penetration (several centimeters in soft tissue), high spatial resolution, and inherent functional sensitivity, PACT has great potential for imaging mouse brains with endogenous and exogenous contrasts, which is of immense interest to the neuroscience community. However, conventional PACT either assumes homogenous optical fluence within the brain or uses a simplified attenuation model for optical fluence estimation. Both approaches underestimate the complexity of the fluence heterogeneity and can result in poor quantitative imaging accuracy. To optimize the quantitative performance of PACT, we explore for the first time 3D Monte Carlo (MC) simulation to study the optical fluence distribution in a complete mouse brain model. We apply the MCX MC simulation package on a digital mouse (Digimouse) brain atlas that has complete anatomy information. To evaluate the impact of the brain vasculature on light delivery, we also incorporate the whole-brain vasculature in the Digimouse atlas. k-wave toolbox was used to investigate the effect of inhomogeneous illumination on the reconstructed images and chromophore concentration estimation. The simulation results clearly show that the optical fluence in the mouse brain is heterogeneous at the global level and can decrease by a factor of five with increasing depth. Moreover, the strong absorption and scattering of the brain vasculature also induce the fluence disturbance at the local level. Both global and local fluence heterogeneity contributes to the reduced quantitative accuracy of the reconstructed PACT images of mouse brain. Correcting the optical fluence distribution can improve the quantitative accuracy of PACT.

Light-Powered Directional Nanofluidic Ion Transport in Kirigami-Made Asymmetric Photonic-Ionic Devices.

Nacre-mimetic 2D nanofluidic materials with densely packed sub-nanometer-height lamellar channels find widespread applications in water-, energy-, and environment-related aspects by virtue of their scalable fabrication methods and exceptional transport properties. Recently, light-powered nanofluidic ion transport in synthetic materials gained considerable attention for its remote, noninvasive, and active control of the membrane transport property using the energy of light. Toward practical application, a critical challenge is to overcome the dependence on inhomogeneous or site-specific light illumination. Here, asymmetric photonic-ionic devices based on kirigami-tailored graphene oxide paper are fabricated, and directional nanofluidic ion transport properties therein powered by full-area light illumination are demonstrated. The in-plane asymmetry of the graphene oxide paper is essential to the generation of photoelectric driving force under homogeneous illumination. This light-powered ion transport phenomenon is explained based on a modified carrier diffusion model. In asymmetric nanofluidic structures, enhanced recombination of photoexcited charge carriers at the membrane boundary breaks the electric potential balance in the horizontal direction, and thus drives the ion transport in that direction under symmetric illumination. The kirigami-based strategy provides a facile and scalable way to fabricate paper-like photonic-ionic devices with arbitrary shapes, working as fundamental elements for large-scale light-harvesting nanofluidic circuits.

Comparison of feature point detectors for multimodal image registration in plant phenotyping.

With the introduction of multi-camera systems in modern plant phenotyping new opportunities for combined multimodal image analysis emerge. Visible light (VIS), fluorescence (FLU) and near-infrared images enable scientists to study different plant traits based on optical appearance, biochemical composition and nutrition status. A straightforward analysis of high-throughput image data is hampered by a number of natural and technical factors including large variability of plant appearance, inhomogeneous illumination, shadows and reflections in the background regions. Consequently, automated segmentation of plant images represents a big challenge and often requires an extensive human-machine interaction. Combined analysis of different image modalities may enable automatisation of plant segmentation in “difficult” image modalities such as VIS images by utilising the results of segmentation of image modalities that exhibit higher contrast between plant and background, i.e. FLU images. For efficient segmentation and detection of diverse plant structures (i.e. leaf tips, flowers), image registration techniques based on feature point (FP) matching are of particular interest. However, finding reliable feature points and point pairs for differently structured plant species in multimodal images can be challenging. To address this task in a general manner, different feature point detectors should be considered. Here, a comparison of seven different feature point detectors for automated registration of VIS and FLU plant images is performed. Our experimental results show that straightforward image registration using FP detectors is prone to errors due to too large structural difference between FLU and VIS modalities. We show that structural image enhancement such as background filtering and edge image transformation significantly improves performance of FP algorithms. To overcome the limitations of single FP detectors, combination of different FP methods is suggested. We demonstrate application of our enhanced FP approach for automated registration of a large amount of FLU/VIS images of developing plant species acquired from high-throughput phenotyping experiments.

A deep Coarse-to-Fine network for head pose estimation from synthetic data

Various applications of human-computer interaction are based on the estimation of head pose, which is challenging due to different facial appearance, inhomogeneous illumination, partial occlusion, etc. In this paper, we propose a deep neural network following the Coarse-to-Fine strategy to estimate head poses. The scheme includes two branches: Coarse classification phase classifying the input image into four categories, and Fine Regression phase estimating the accurate pose parameters. The two sub-networks are trained jointly. To tackle the problem of insufficient annotated data in training process, we design a rendering pipeline to synthesize realistic head images and generate an annotated dataset with a collection of 310k head poses. The results on benchmark datasets and synthetic dataset validate the effectiveness of our approach, as well as the results on images with diverse illumination, occlusion, and motion blur. Moreover, our method can be easily extended to estimate head poses on depth images.

Comparison and extension of three methods for automated registration of multimodal plant images

With the introduction of high-throughput multisensory imaging platforms, the automatization of multimodal image analysis has become the focus of quantitative plant research. Due to a number of natural and technical reasons (e.g., inhomogeneous scene illumination, shadows, and reflections), unsupervised identification of relevant plant structures (i.e., image segmentation) represents a nontrivial task that often requires extensive human-machine interaction. Registration of multimodal plant images enables the automatized segmentation of ’difficult’ image modalities such as visible light or near-infrared images using the segmentation results of image modalities that exhibit higher contrast between plant and background regions (such as fluorescent images). Furthermore, registration of different image modalities is essential for assessment of a consistent multiparametric plant phenotype, where, for example, chlorophyll and water content as well as disease- and/or stress-related pigmentation can simultaneously be studied at a local scale. To automatically register thousands of images, efficient algorithmic solutions for the unsupervised alignment of two structurally similar but, in general, nonidentical images are required. For establishment of image correspondences, different algorithmic approaches based on different image features have been proposed. The particularity of plant image analysis consists, however, of a large variability of shapes and colors of different plants measured at different developmental stages from different views. While adult plant shoots typically have a unique structure, young shoots may have a nonspecific shape that can often be hardly distinguished from the background structures. Consequently, it is not clear a priori what image features and registration techniques are suitable for the alignment of various multimodal plant images. Furthermore, dynamically measured plants may exhibit nonuniform movements that require application of nonrigid registration techniques. Here, we investigate three common techniques for registration of visible light and fluorescence images that rely on finding correspondences between (i) feature-points, (ii) frequency domain features, and (iii) image intensity information. The performance of registration methods is validated in terms of robustness and accuracy measured by a direct comparison with manually segmented images of different plants. Our experimental results show that all three techniques are sensitive to structural image distortions and require additional preprocessing steps including structural enhancement and characteristic scale selection. To overcome the limitations of conventional approaches, we develop an iterative algorithmic scheme, which allows it to perform both rigid and slightly nonrigid registration of high-throughput plant images in a fully automated manner.

Automatic estimation of mechanical properties from fractographs using optimal anisotropic diffusion and Voronoi tessellation

With the advent of materials informatics there is a high demand of establishing automatic structure-property correlationship of materials which is a popular research topics due to the advancement of image processing and pattern recognition algorithms. Therefore, in this work, an attempt is made to estimate mechanical properties (i.e. yield strength and ductility) of AISI 304LN stainless steel from the SEM images of fracture surfaces obtained from tensile tests carried out at different strain rates using image processing techniques. As the void morphologies of fracture surfaces change systematically with the change in strain rates, the automatic detection of voids and geometrical features estimation from detected voids from the obtained are key goals of this present study. Therefore, in this work, a novel method of optimal anisotropic diffusion technique along with contrast limited adaptive histogram equalization (CLAHE), Otsu's optimal thresholding and morphological thinning operation are applied over the fractographs for edge enhancement, overcoming inhomogeneous illumination, edge segmentation and thinning, respectively, to detect voids, automatically. Then, Voronoi tessellation technique, which is a geometrical texture analysis, is utilized on these edge images of fractographs to extract four features viz. Voronoi edges, mean area, mean elongation and mean perimeter of Voronoi polygons where the linear correlation values (R2) with mechanical properties are found in the range of 0.90–0.99. A high linear correlation of features (i.e. 0.97–0.99) with the ductility is noticed as ductility is a geometrical parameter measured during fracture.

Simultaneous arteriole and venule segmentation with domain-specific loss function on a new public database.

The segmentation and classification of retinal arterioles and venules play an important role in the diagnosis of various eye diseases and systemic diseases. The major challenges include complicated vessel structure, inhomogeneous illumination, and large background variation across subjects. In this study, we employ a fully convolutional network to simultaneously segment arterioles and venules directly from the retinal image, rather than using a vessel segmentation-arteriovenous classification strategy as reported in most literature. To simultaneously segment retinal arterioles and venules, we configured the fully convolutional network to allow true color image as input and multiple labels as output. A domain-specific loss function was designed to improve the overall performance. The proposed method was assessed extensively on public data sets and compared with the state-of-the-art methods in literature. The sensitivity and specificity of overall vessel segmentation on DRIVE is 0.944 and 0.955 with a misclassification rate of 10.3% and 9.6% for arteriole and venule, respectively. The proposed method outperformed the state-of-the-art methods and avoided possible error-propagation as in the segmentation-classification strategy. The proposed method was further validated on a new database consisting of retinal images of different qualities and diseases. The proposed method holds great potential for the diagnostics and screening of various eye diseases and systemic diseases.

Hybrid organic/inorganic position-sensitive detectors based on PEDOT:PSS/n-Si

Various configurations like p-n junctions, metal-semiconductor Schottky barriers, and metal-oxide-semiconductor structures have been widely used in position-sensitive detectors. In this report, we propose a PEDOT:PSS/n-Si heterojunction as a hybrid organic/inorganic configuration for position-sensitive detectors. The influence of the thickness of the PEDOT:PSS layer, the wavelength of incident light, and the intensity of illumination on the device performance are investigated. The hybrid PSD exhibits very high sensitivity (>100 mV/mm), excellent nonlinearity (<3%), and a response correlation coefficient (>0.995) with a response time of <4 ms to the inhomogeneous IR illumination. The presented hybrid configuration also benefits from a straightforward low-temperature fabrication process. These advantages of the PEDOT:PSS/n-Si heterojunction are very promising for developing a new class of position-sensitive detectors based on the hybrid organic/inorganic junctions.

A deeply supervised residual network for HEp-2 cell classification via cross-modal transfer learning

Accurate Human Epithelial-2 (HEp-2) cell image classification plays an important role in the diagnosis of many autoimmune diseases and subsequent treatment. One of the key challenges is huge intra-class variations caused by inhomogeneous illumination. To address it, we propose a framework based on very deep supervised residual network (DSRN) to classify HEp-2 cell images. Specifically, we adopt a residual network of 50 layers (ResNet-50) that is substantially deep to extract rich and discriminative features. The deep supervision is imposed on the ResNet-based framework to further boost the classification performance by directly guiding the training of the lower and upper levels of the network. The proposed method is evaluated using two publicly available datasets (i.e., International Conference on Pattern Recognition (ICPR) 2012 and ICPR2016-Task1 cell classification contest datasets). Different from the previous deep learning models learned from scratch, a cross-modal transfer learning strategy is developed. Namely, we pretrain ICPR2012 dataset to fine-tune ICPR2016 dataset based on our DSRN model since both datasets are similar. Extensive experiments show that the proposed method delivers state-of-the-art performance and outperforms the traditional methods based on deep convolutional neural network (DCNN).

Correlation and Differentiation Based Algorithms for Cell Mobility Quantification

Quantification of cell mobility in confluent cell cultures or on coverslips with high cell density is a challenge, because tracking of individual cells is not readily possible. Two procedures have been developed and tested to solve this problem, namely a differential (MoveDif) and a correlation (MoveCor) algorithm. Both compare successive image frames and calculate a difference parameter between them. Image preprocessing was required to neutralize brightness changes and inhomogeneous illumination. Additionally, structures and membranes in the images had to be enhanced and standardized by edge-detection filters. We showed that both algorithms had advantages in different experimental setups, depending of the complexity of the cellular movement. Besides the determination of global cell mobility, the membrane contractility and movement of single cells, small cell groups and intracellular structures could be quantified, if individual cells were separated and magnified. The discriminatory potency and sensitivity of both algorithms were tested by drug induced changes of cellular movement speeds and their robustness against brightness changes were analyzed too. To demonstrate the mobility effects of ketamine on dexamethasone-treated hippocampal neurons, we found that the correlation algorithm MoveCor could quantify the ketamine-dependent increase in cell movement speed after prior treatment with dexamethasone had reduced it distinctively. We also could measure how staurosporine slowed down the mobility of acidic intracellular organelles in highly magnified glial cells from chicken telencephalon. The differentiation algorithm MoveDif was employed to demonstrate that the contractility of pericyte membranes slowed down after induction of oxygen glucose deprivation. The limitations of both methods were given by their inability to produce absolute mobility values but these drawbacks can be solved by appropriate image preprocessing.

Ab initio multiscale simulation of high-order harmonic generation in solids

High-harmonic generation by a highly non-linear interaction of infrared laser fields with matter allows for the generation of attosecond pulses in the XUV spectral regime. This process, well established for atoms, has been recently extended to the condensed phase. Remarkably well pronounced harmonics up to order ~30 have been observed for dielectrics. We present the first ab-initio multi-scale simulation of solid-state high-harmonic generation. We find that mesoscopic effects of the extended system, in particular the realistic sampling of the entire Brillouin zone, the pulse propagation in the dense medium, and the inhomogeneous illumination of the crystal have a strong effect on the formation of clean harmonic spectra. Our results provide a novel explanation for the formation of clean harmonics and have implications for a wide range of non-linear optical processes in dense media.

A novel method for automated correction of non-uniform/poor illumination of retinal images without creating false artifacts

Retinal images are frequently corrupted by unwanted variations in the brightness that occur due to over-all imperfections in the image acquisition process. This inhomogeneous illumination across the retina can limit the pathological information that can be gained from the image; and can lead to serious difficulties when performing image processing tasks that requires qualitative as well as quantitative analysis of feature presence on the image. On that perspective we have proposed a novel two-step approach for non-uniform and/or poor illumination correction in the context of retinal imaging. A subjective experiment was conducted to ensure that the proposed method did not create visually noticeable false color or artifacts on the images, especially on the areas that did not suffer non-uniform/poor illumination prior to correction. An objective experiment on 25,872 retinal images was performed to justify the significance of the proposed method for automated pathology detection/classification.

Size Measurement of Laser-Induced Damage in Large-Aperture Final Optics

In this paper, a hierarchical kernel extreme learning machine (HK-ELM) is proposed to solve the problem that the size of laser-induced damage (LID) cannot be accurately measured under the condition of inhomogeneous total internal reflection illumination. The proposed method overcomes the limitations of the radiometric method, which depends on a single factor to measure the size of damage. After analyzing the light scattering characteristics of LID with the finite-difference time-domain method, we find that we should consider the influences of multiple factors when measuring the size of damage under the condition of uneven illumination. The experimental results show that HK-ELM can achieve ultra-resolution measurements, and the mean relative error of the measured size predicted by HK-ELM is within 5% on the testing samples. Compared with the traditional technology, the method proposed in this paper effectively improves the measurement accuracy.

Partially shaded III-V concentrator solar cell performance

Currently, an important concern in CPV is inhomogeneity of the light distribution on the cell introduced by the applied optical systems, which may affect system performance. In BICPV applications, the inhomogeneities can be much more severe because of design constraints introduced by the building incorporation. Additionally, one of the predominant loss mechanisms in CPV solar cells is perimeter recombination. In this study, the electrical parameters of CPV cells are investigated under inhomogeneous illumination intensity profiles. Partial shading is used as a model for extremely inhomogeneous illumination, while several shadow patterns are used to study the effect of perimeter recombination on the cell performance. As the latter occurs most strongly in GaAs subcells, shallow and deep junction GaAs CPV cells have been developed and subjected to these experiments, as well as commercial triple junction CPV cells. Deep junction GaAs cells are shown to perform significantly better under concentrated light than their shallow junction counterparts. A large degree of shading exceeding 70% has been found to cause only minor losses in the cell performance of 4%. Also, the cell performance is found to be independent of the location of illumination, in spite of perimeter recombination effects, because the current density spreads out. Clearly, increased illumination inhomogeneities caused by elaborate BICPV optical systems, do not inhibit the electrical performance strongly. As a consequence, a large degree of design freedom exists for the optical systems, which offers good opportunities to develop BICPV that meet all the design challenges of the built environment.

A target detectability assessment metric for the underwater electro-optic imaging system

Compared with the atmospheric imaging, in underwater electro-optic imaging (UEOI), there exist difficulties to describe the target discrimination level within the visibility range, because of the media scattering and inhomogeneous illumination. The large-scale contrast is used to describe the target discrimination traditionally, while it is not linear to human perception. In this paper, the detectability of a UEOI system is firstly defined to quantify the perceptive target discrimination level in full space. And we present a target detectability assessment metric (TDAM), which is used to illustrate the detectability distribution of a UEOI system. Additionally based on the large-scale contrast, an automatic detectability calculation pipeline is proposed, and a trapper-shaped function fits the contrast and detectability well, with the Spearman rank order correlation coefficient (SROCC = 0.8852), Pearson linear correlation coefficient (PLCC = 0.8965) and root-mean-square error (RMSE = 1.6532).

Applications of least-squares pre-stack depth migration in complex geology around the world

Pre-stack depth migration has been an industrial imaging standard for decades, starting from the adoption of Kirchhoff migration in the early Nineties to the emergence of reverse-time migration (RTM) in the late 2000s. These algorithms map the recorded seismic reflection energy from a surface location to a subsurface location, through either a ray-based travel-time table or a waveequation-based propagation engine. In principle, by combining high-fidelity migration algorithms and accurate subsurface velocity models, we can achieve the ultimate goals of seismic imaging: the correct positioning and focusing of the seismic reflectivity of the subsurface geological structures. In reality, however, there are several challenges and issues that need to be addressed before we are able to achieve such an objective, not to mention the fact that most of the time we do not record all subsurface reflections sufficiently owing to limited coverage and sampling of the seismic recording spread. We can consider recorded seismic data to be the result of forward modelling experiments through subsurface structures. To image the reflectivity of the subsurface, we need to reverse the forward wave-propagation effects with an inverse of the forward modelling operator. Essentially, this is an inversion process. However, conventional imaging algorithms are formulated as adjoint operators rather than as true inverses (Claerbout, 1992). This approximation caused image degradation due to irregular and aliased acquisition sampling, limited receiver coverage, noise, and inhomogeneous or poor illumination caused by complex overburden. As a result, standard pre-stack depth migration algorithms suffer from migration artifacts with uncancelled swings, limited bandwidth, and distorted amplitude on subsurface reflectors (Gray, 1997). This is true even for stateof-the-art imaging technology such as RTM (Baysal et al., 1983; Zhang and Zhang, 2009).

Photoluminescence Imaging of Silicon Wafers and Solar Cells With Spatially Inhomogeneous Illumination

Photoluminescence imaging is a fast and powerful spatially resolved characterization technique, commonly used for silicon wafers and solar cells. In conventional measurements, homogeneous illumination is used across the sample. In this paper, we present a photoluminescence imaging setup that enables inhomogeneous illumination with arbitrary illumination patterns to determine various parameters of solar cells and solar cell precursors. To demonstrate the strength of the proposed inhomogeneous illumination imaging, a set of proof-of-concept measurements have been conducted; these measurements include contactless series resistance imaging, emitter sheet resistance, and diffusion length measurements. The results indicate that the use of inhomogeneous illumination significantly extends the range of photoluminescence imaging applications for the characterization of silicon wafers and solar cells.

Temporally Coherent Illumination Normalization for Indocyanine Green Video Angiography.

Indocyanine green (ICG) angiography is an imaging method for doctors to observe choroidal abnormalities in human eyes. The ICG angiograms typically exhibit inhomogeneous illumination, which poses serious difficulties for the development of computer-aided diagnostic tools. In this paper, we propose a novel illumination normalization method to alleviate the inhomogeneous illumination in ICG video angiograms. In particular, we first align the viewpoint of the input ICG video angiogram using an image registration method. Then, we acquire temporal information using time-dependent intrinsic image and compute the corresponding illumination image. Finally, we correct inhomogeneous illumination from the illumination image by estimating contrast and luminosity distortion. We have conducted extensive evaluation using ICG video angiograms of 60 patients. Two video quality assessment methods are utilized to evaluate the performance of our proposed illumination normalization method. The results show that our proposed method can help improve the visual quality of ICG video angiogram. Visual evaluation by a human expert also confirms that our method yields better illumination normalization results.

The operational methane retrieval algorithm for TROPOMI

Abstract. This work presents the operational methane retrieval algorithm for the Sentinel 5 Precursor (S5P) satellite and its performance tested on realistic ensembles of simulated measurements. The target product is the column-averaged dry air volume mixing ratio of methane (XCH4), which will be retrieved simultaneously with scattering properties of the atmosphere. The algorithm attempts to fit spectra observed by the shortwave and near-infrared channels of the TROPOspheric Monitoring Instrument (TROPOMI) spectrometer aboard S5P.The sensitivity of the retrieval performance to atmospheric scattering properties, atmospheric input data and instrument calibration errors is evaluated. In addition, we investigate the effect of inhomogeneous slit illumination on the instrument spectral response function. Finally, we discuss the cloud filters to be used operationally and as backup.We show that the required accuracy and precision of < 1 % for the XCH4 product are met for clear-sky measurements over land surfaces and after appropriate filtering of difficult scenes. The algorithm is very stable, having a convergence rate of 99 %. The forward model error is less than 1 % for about 95 % of the valid retrievals. Model errors in the input profile of water do not influence the retrieval outcome noticeably. The methane product is expected to meet the requirements if errors in input profiles of pressure and temperature remain below 0.3 % and 2 K, respectively. We further find that, of all instrument calibration errors investigated here, our retrievals are the most sensitive to an error in the instrument spectral response function of the shortwave infrared channel.

Monitoring the effect of transglutaminase in semi-hard cheese during ripening by hyperspectral imaging

Semi-hard cheese samples made from pasteurized 3.5% and 5% fat cow’s milk were partly enzyme treated with microbial transglutaminase (mTG). Cheese colour and yield was greatly affected by mTG at both fat levels. The cross-linking effect of mTG also led to improved protein content and hardness in cheese. As Texture Profile Analysis (TPA) could only distinguish according to enzyme treatment we studied hyperspectral imaging (HSI) to differentiate also according to fat level and ripening status. HSI was used on a weekly basis during 4–10 weeks period of ripening time. In case of uneven surface, using the spatial information and making proper normalization on the spectra of ROI pixels, the hyperspectral method can provide even better signal-to-noise ratio then the conventional spectrophotometric method. Statistical analysis with Partial Least Squares (PLS) regression and Monte Carlo Cross Validation (MCCV) led to determination of significant wavelengths both for the detection of enzyme treatment (1387 nm) and for fat content (1190, 1234 nm) independent of the actual age of examined semi-hard cheese. Hyperspectral measurement method allowed remote inspection of the product through transparent vacuum foil. The spectrum normalization allowed the elimination of effects originating from inhomogeneous illumination caused by cheese holes.