Background Illumination Research Articles

Chassis Assembly Detection and Identification Based on Deep Learning Component Instance Segmentation

Chassis assembly quality is a necessary step to improve product quality and yield. In recent years, with the continuous expansion of deep learning method, its application in product quality detection is increasingly extensive. The current limitations and shortcomings of existing quality detection methods and the feasibility of improving the deep learning method in quality detection are presented and discussed in this paper. According to the characteristics of numerous parts and complex types of chassis assembly components, a method for chassis assembly detection and identification based on deep learning component segmentation is proposed. In the proposed method, assembly quality detection is first performed using the Mask regional convolutional neural network component instance segmentation method, which reduces the influence of complex illumination conditions and background detection. Next, a standard dictionary of chassis assembly is built, which is connected with Mask R-CNN in a cascading way. The component mask is obtained through the detection result, and the component category and assembly quality information is extracted to realize chassis assembly detection and identification. To evaluate the proposed method, an industrial assembly chassis was used to create datasets, and the method is effective in limited data sets of industrial assembly chassis. The experimental results indicate that the accuracy of the proposed method can reach 93.7%. Overall, the deep learning method realizes complete automation of chassis assembly detection.

Moving Vehicle Detection with Shadow Elimination Based on Improved ViBe Algorithm

Moving vehicle detection based on video processing has been widely used in intelligent transportation system recently. However there are also many problems, such as dynamic background, ghost region, and shadow of moving objects. This paper proposes an improved ViBe object detection algorithm. First, an accurate background image is obtained by using the multi-frame averaging method, and then the background model is initialized by this accurate background image, thus effectively reducing the generation of ghost region. Whenever there is no moving object for a fixed number of consecutive frames in the video, this frame is updated to the background image. Conservative update strategy and foreground point counting method are adopted to update the background and reduce the impact of dynamic background on the foreground detection. Next, the foreground image detected by improved ViBe algorithm is input into the shadow elimination method proposed in this paper. Shadows in foreground pixels are detected in RGB color space, and then the pixels determined as shadows are eliminated. Finally, accurate moving vehicles are obtained. Our algorithm can effectively eliminate the shadows of moving vehicles, quickly adapt to background changes and illumination mutation, and get accurate moving objects, which is helpful for vehicle contour extraction and subsequent image processing.

Phase-shifting VU factorization for interferometry

The problem of phase recovery in phase-shifting interferometry (PSI), when the introduced phase-shifts are unknown, has been analyzed by existing methods through the modification or combination of the least-squares and the principal component analysis (PCA) approaches. In this paper, we show that the PSI sequence of images can be written in a matrix form as the product of two matrices V and U, where the matrix V contains the components of the modulating phase and the matrix U contains the components of the phase-shifts. The VU matrix describes the PSI images without restrictions on the background illumination or the phase shifts, thus, the phase-shifts can be given arbitrarily and they do not need to cover the interval [0, 2π]. In particular, we introduce a numerical method that can separate these matrices V and U from a given set of PSI images, which enables us to recover both the modulating phase and the phase-shifts, albeit, we are only interested in the modulating phase. We refer to this process as phase-shifting VU factorization. This factorization method is fast, stable, and converges well to the desired solution. An important feature of the method is that the convergence of the VU factorization is robust to the initial guess and the only parameter it requires is the convergence accuracy. The method can reach a convergence accuracy of the order of 10−3 in a few iterations. Results from our method using both simulated and experimental interferograms are shown and compared with recovered modulating phases obtained by three other state-of-the-art approaches: the PCA, least squares and a combination of both.

A deep-level region-based visual representation architecture for detecting strawberry flowers in an outdoor field

An accurate and robust strawberry flower representation and detection scheme is a key step to enable the reliable forecasting of fruit yield for use in precision agricultural applications. A state-of-the-art deep-level object detection framework which processes images through several layers using a region-based convolutional neural network (R-CNN) was developed to visually represent the instances of strawberry flowers in outdoor fields and improve the detection accuracy. A modified version of the visual geometry group 19 (VGG19) architecture, which had 47 layers, was used to represent the multiple scales of strawberry flower image features. The networks were trained entirely on 400 strawberry flower images and tested on another 100 images. Different region-based object detection methods, including the R-CNN, Fast R-CNN and Faster R-CNN, were used to represent the strawberry flower instances. The Faster R-CNN model achieved a better performance than the R-CNN and Fast R-CNN in detecting the instances and had a lower execution time. The detection accuracy of the Faster R-CNN model was 86.1%, which was higher than those of the R-CNN and Fast R-CNN models (63.4% and 76.7%, respectively). The experimental results showed the effectiveness of the deep-level Faster R-CNN framework for representing the strawberry flower instances under various camera view-points, different distances to flowers, overlaps, complex background illumination, blur, etc. The system developed for automatic and accurate strawberry flower detection provides an important and significant solution that enables subsequent applications to estimate the strawberry yield in outdoor fields.

Polytetrafluoroethylene, by near-ambient pressure XPS

Near-ambient pressure–x-ray photoelectron spectroscopy (NAP-XPS) is a less traditional form of XPS that allows samples to be analyzed at relatively high pressures, i.e., at greater than 2500 Pa. With NAP-XPS, XPS can probe moderately volatile liquids, biological samples, porous materials, and/or polymeric materials that outgas significantly. Polytetrafluoroethylene (PTFE) is an important polymer with many applications in science and industry. It is an insulator that charges under x-ray illumination at high vacuum. In this submission, we show NAP-XPS spectra of PTFE. Survey spectra are shown at different background gas (air) pressures. These spectra contain F 2s, C 1s, O 1s, N 1s, F 1s, and F Auger signals. Also presented are F 1s narrow scans over a range of background pressures and illumination times. Peaks decrease in width, shift toward literature values, and improve in shape with increasing background gas pressure.

Change nonlinear-optical response of electro-optic crystal in conditions of contributions incoherent background illumination

Influence of incoherent background on diffraction divergence of narrow light beams of He-Ne laser with wavelength of 633 nm due to refractive index changes of Fe-doped lithium niobate sample is experimentally studied. Incoherent background is produced by light-emitting diodes (LED´s) of different average wavelengths and laser diode emitting in blue region of visible. It has been experimentally demonstrated that nonlinear diffraction of coherent red beams of He-Ne laser with diameters on full width of half maximum (FWHM) near to 15–20 μm may be totally compensated by means of assistance of incoherent background with shorter average wavelengths 455–465 nm and even much lower intensity compared with that for coherent beam.

LOSA-X: Expandable 3D Motion Tracking System

This paper presents a novel, low-cost, expandable 3D motion tracking system with the ability to adapt to dynamic background illumination. The proposed system consists of multiple tracker modules equipped with a linear optical sensor arrays and a 9-DOF inertial measurement unit for position and orientation tracking of a wireless IR-LED illuminated active marker. This paper presents an extended Kalman filter-based sensor fusion architecture for combining data acquired from multiple trackers and marker for enhanced accuracy and operational range. The marker system adaptively adjusts the LED intensity to optimize energy consumption and measurement accuracy. The proposed system’s ability to extract background illumination allows it to be robust toward differences in scene illuminations, both static and dynamic. Detailed hardware design and performance evaluation through experimentation have been discussed in this paper.

Mid-wavelength infrared InAsSb/InAs nBn detectors and FPAs with very low dark current density

There was a significant improvement in the performance of infrared nBn detectors utilizing InAs/InAsSb absorbers culminating in the development of infrared focal plane arrays (FPAs) with excellent operability (99.7%) and operating at temperature significantly higher than InSb FPAs. In this work, we demonstrated that these detectors can operate with very low dark current densities enabling their use in applications with a low-to-medium level of background illumination. We showed that single pixel photodetectors with a cut-off wavelength of 4.8 μm and a quantum efficiency of QEmax = 35% under backside-illumination have a dark current density of 1 × 10−10 A/cm2 at an operating bias of −0.1 V and temperature T = 100 K. Additionally, we compared the single pixel dark current results with measurements of the dark current in FPAs. The FPA showed excellent performance with an operability of 99.7% and was able to reach a mode dark current density of 5 × 10−11 A/cm2 at 80 K.

Imaging enhancement based on stimulated Brillouin amplification in optical fiber.

The detection of weak optical signals embedded in strong background illumination has broad application prospects. We propose an imaging enhancement method based on stimulated Brillouin scattering (SBS) in a single-mode fiber, which is capable of amplifying the weak optical signal while neglecting the broadband background noise because of its narrow gain bandwidth. In experiment, a high gain of 60 dB was achieved. An imaging enhancement experiment was carried out, where a target which cannot be seen because of transmission loss could be clearly captured with the amplification of SBS in the fiber. Because of the employment of continuous pump rather than a pulsed pump, this system has wide application in the monitoring of non-cooperative targets.

Decreased Amplitude of the Photopic Negative Response (PhNR) in the Ins2Akita Mouse Model of Type 1 Diabetes Mellitus

Diabetic retinopathy (DR) is a common, severe complication of diabetes mellitus. Current clinical detection of DR is based primarily on visible retinal vascular damage, and earlier detection of impending DR might allow earlier treatment to protect function. Recent studies in both animal models and humans with diabetes have shown that degradation of retinal ganglion cell (RGC) function occurs at an early stage, before visible vascular complications and before significant RGC loss. The photopic negative response (PhNR), obtained from full‐field electroretinogram (ERG) with photopic stimuli is thought to arise directly from RGC activity. Here, we hypothesized that PhNR amplitudes are reduced in a type 1 diabetic mouse model, the Ins2Akita mouse, which develops hyperglycemia by 5 weeks of age and visual function and ocular blood flow deficits early in the disease progression, before overtly visible vascular damage. Diabetic (male heterozygotes) and control (male WT littermates) mice were grouped according to age: young, adult, and old (5–6, 9–11, and 12–14 months of age, respectively). ERG recording was performed with 1.62 log cd/m2 white flashes over a constant 40 cd.s/m2 green background (Diagnosys Espion and ColorDome ganzfeld system) using gold ring contact lens electrodes (LKC Techniologies) under isoflurane anesthesia. Following 1 hour adaptation to the green background illumination, PhNR, a‐wave, and b‐wave amplitudes and implicit times were measured in response to 25 10‐msec flashes.The mean PhNR amplitude of the adult diabetic mice was 36% of their age‐matched controls (p < 0.01). For the old diabetic group, the mean PhNR amplitude was 49% of their age‐matched controls (p < 0.05). However, no significant difference was found in the young group. A‐wave amplitudes were decreased in the diabetic adult group, to a value only 41% of their age‐matched controls (p < 0.05). B‐wave amplitudes were decreased in the diabetic old group, to 35% of their age‐matched controls (p < 0.05). However, no significant difference was found among groups for PhNR, a‐wave, or b‐wave implicit times. Our results suggest that PhNR amplitude is altered relatively early in the Ins2Akita mouse model of diabetes. This suggests PhNR measurements could be used to detect diabetic retinopathy at earlier stages than the current clinical standard.Support or Funding InformationNIH R01 EY023290 and UIWRSO Faculty Development GrantThis abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

Assessing Feasibility of Secure Quantum Communications Involving Underwater Assets

In this paper, we apply a mathematical model of the quantum bit error rate to assess the viability of quantum key distribution (QKD) protocols for secure and covert underwater communications. We consider scenarios that may include passive sensor arrays, coordinated underwater vehicles, and surface and aerial assets. This analysis suggests that single-photon QKD can potentially be used effectively at depths between 60 and 110 m in clear waters, with the precise depth depending in part on ambient illumination levels. We discuss how diurnal variations in background illumination can be exploited in some applications to achieve improved communications rate and/or reliability when using QKD or other secure optical protocol.

Compact Real-Time Inter-Frame Histogram Builder for 15-Bits High-Speed ToF-Imagers Based on Single-Photon Detection

Time-of-flight (ToF) image sensors based on single-photon detection, i.e., SPADs, require some filtering of pixel readings. Accurate depth measurements are only possible if the jitter of the detector is mitigated. Moreover, the time stamp needs to be effectively separated from uncorrelated noise, such as dark counts and background illumination. A powerful tool for this is building a histogram of a number of pixel readings. Future generation of ToF imagers are seeking to increase spatial and temporal resolution along with the dynamic range and frame rate. Under these circumstances, storing the complete histogram for every pixel becomes practically impossible. Considering that most of the information contained by the histogram represents noise, we propose a highly efficient method to store just the relevant data required for the ToF computation. This method makes use of the shifted inter-frame histogram. It requires a memory as low as 128 times smaller than storing the complete histogram if the pixel values are coded on up to 15 bits. Moreover, a fixed 28 words memory is enough to process histograms containing up to 215 bins. In exchange, the overall frame rate only decreases to one half. The hardware implementation of this algorithm is presented. Its remarkable robustness for a low SNR of the ToF estimation is demonstrated by Matlab simulations and FPGA implementation using input data from a SPAD camera prototype.

Standards in Pupillography.

The number of research groups studying the pupil is increasing, as is the number of publications. Consequently, new standards in pupillography are needed to formalize the methodology including recording conditions, stimulus characteristics, as well as suitable parameters of evaluation. Since the description of intrinsically photosensitive retinal ganglion cells (ipRGCs) there has been an increased interest and broader application of pupillography in ophthalmology as well as other fields including psychology and chronobiology. Color pupillography plays an important role not only in research but also in clinical observational and therapy studies like gene therapy of hereditary retinal degenerations and psychopathology. Stimuli can vary in size, brightness, duration, and wavelength. Stimulus paradigms determine whether rhodopsin-driven rod responses, opsin-driven cone responses, or melanopsin-driven ipRGC responses are primarily elicited. Background illumination, adaptation state, and instruction for the participants will furthermore influence the results. This standard recommends a minimum set of variables to be used for pupillography and specified in the publication methodologies. Initiated at the 32nd International Pupil Colloquium 2017 in Morges, Switzerland, the aim of this manuscript is to outline standards in pupillography based on current knowledge and experience of pupil experts in order to achieve greater comparability of pupillographic studies. Such standards will particularly facilitate the proper application of pupillography by researchers new to the field. First we describe general standards, followed by specific suggestions concerning the demands of different targets of pupil research: the afferent and efferent reflex arc, pharmacology, psychology, sleepiness-related research and animal studies.

Robust visual tracking via global context regularized Locality-constrained Linear Coding

Locality-constrained Linear Coding (LLC) based visual tracking can give a better and faster tracking performance than traditional sparse representation based tracking methods. However, the existing LLC based methods often use the anchor points near the target to build the sparse coding dictionary for local sparse coding. It may cause a problem that it is hard to discriminate the difference between the negative and positive anchor points in sparse coding dictionary when facing severe background clutter, illumination change and occlusion. In this paper, we propose a context aware sparse coding method to achieve robust visual tracking. The proposed method can prevent the negative anchor points from disturbing the classifier accuracy because it uses a global context regularizer to constrain the sparse coding value of those negative anchor points that are similar to the positive anchor points. Experiment results show that our method can achieve a better tracking performance than state-of-the-art tracking methods do.

Head pose estimation using improved label distribution learning with fewer annotations

Head pose estimation in unconstrained environment remains a challenging task due to background clutter, illumination changes, and appearance variabilities. Multivariate label distribution has been successfully applied to head pose estimation. However, it is not applicable to unconstrained environments where assigning reasonable label distributions for images is difficult, and its performance significantly degrades when accurate grid information is unavailable (e.g., only yaw angles are known). To alleviate these problems, we propose an improved label distribution learning approach with fewer annotations. A data-driven weak learning strategy is first developed to construct label distributions to alleviate the problem of unreasonable label distributions. Regularization terms (e.g., L1,2 norm) are then introduced into the loss function induced by weighted Jeffreys divergence to avoid over-fitting. To further ameliorate the performance, positive correlation and negative competition are also introduced into the loss function to fine-tune the parameters of the corresponding model. Extensive experiments have been conducted on public databases: LFW and Pointing04. The proposed method achieves comparable performance over the state-of-art and possesses good generalization ability, but uses only fewer annotations, which suggests that it has strong potential for head pose estimation in unconstrained environments where sufficient annotations are routinely unavailable.

Object detection and tracking under Complex environment using deep learning‐based LPM

Object detection and tracking under complex environment are challenging because of the disturbances induced by background clutter, illumination changes, occlusions and other factors. The bulk of traditional algorithms basically rely on hand‐crafted features, which are not sufficiently robust to a complex environment. Moreover, the processes of detection and tracking are separated, which leads to the overall efficiency not high. In this study, a novel local probability model (LPM)‐based mean shift (MS) algorithm is proposed to integrate object detection and tracking. The main contributions include: (i) a new framework based on the combination of LPM and MS is established for the integration of object tracking and detection. (ii) For object detection, the training and prediction of LPM are built by stacked denoising autoencoders based deep learning. (iii) For object tracking, an MS tracking algorithm leveraging LPM is modified to improve the tracking efficiency under a complex environment. Experimental results demonstrate that the proposed method is superior to the colour histograms based MS and histograms of oriented gradients based MS in terms of robustness and tracking accuracy.

The light-dark task in zebrafish confuses two distinct factors: Interaction between background shade and illumination level preference

The light-dark preference task has been commonly used in rodents to screen for anxiogenic and anxiolytic drugs. However, recent adaptations of the light-dark preference test for zebrafish have produced inconsistent results. Several studies have reported that zebrafish exhibit a preference for light, while others have found a preference for black. We suggest the inconsistencies may be the result of confusing certain parameters of the test leading to improper interpretation. For example, researchers often use “light” interchangeably with “white” and “dark” with “black” when these are two distinct factors: level of illumination vs. background shade. In the current study, we use specifically designed preference tanks to investigate the influence of background shade (i.e. white vs. black) and level of illumination (i.e. light vs. dark) on preference and anxiety-like behaviour. Furthermore, we pharmacologically validate our results by quantifying the effects of ethanol, a drug with known anxiety-altering properties, on anxiety-like behaviours. Here we report that zebrafish's preference varies depending upon background shade and level of illumination. We also found that ethanol administration altered behavioural responses in an illumination- and background shade-dependent manner. Our findings reinforce the need to correctly differentiate between these factors when interpreting results obtained with this behavioural paradigm. Lastly, our results show that simple modifications to the experimental tank in which anxiety-related responses are measured can significantly alter behaviour of zebrafish, supporting the need for standardized testing procedures and/or for detailed description of experimental procedures and the apparatus.

Robust Human Activity Recognition Using Multimodal Feature-Level Fusion

Automated recognition of human activities or actions has great significance as it incorporates wide-ranging applications, including surveillance, robotics, and personal health monitoring. Over the past few years, many computer vision-based methods have been developed for recognizing human actions from RGB and depth camera videos. These methods include space-time trajectory, motion encoding, key poses extraction, space-time occupancy patterns, depth motion maps, and skeleton joints. However, these camera-based approaches are affected by background clutter and illumination changes and applicable to a limited field of view only. Wearable inertial sensors provide a viable solution to these challenges but are subject to several limitations such as location and orientation sensitivity. Due to the complementary trait of the data obtained from the camera and inertial sensors, the utilization of multiple sensing modalities for accurate recognition of human actions is gradually increasing. This paper presents a viable multimodal feature-level fusion approach for robust human action recognition, which utilizes data from multiple sensors, including RGB camera, depth sensor, and wearable inertial sensors. We extracted the computationally efficient features from the data obtained from RGB-D video camera and inertial body sensors. These features include densely extracted histogram of oriented gradient (HOG) features from RGB/depth videos and statistical signal attributes from wearable sensors data. The proposed human action recognition (HAR) framework is tested on a publicly available multimodal human action dataset UTD-MHAD consisting of 27 different human actions. K-nearest neighbor and support vector machine classifiers are used for training and testing the proposed fusion model for HAR. The experimental results indicate that the proposed scheme achieves better recognition results as compared to the state of the art. The feature-level fusion of RGB and inertial sensors provides the overall best performance for the proposed system, with an accuracy rate of 97.6%.

Mobile system for road sign detection and recognition with template matching

This paper explores the effective approach to road sign detection and recognition based on mobile devices. Detecting and recognising road signs is a challenging matter because of different shapes, complex background and irregular sign illumination. The main goal of the system is to assist drivers by warning them about the existence of road signs to increase safety during driving. In this paper, the system for detection and recognition of road signs was implemented and tested with the use of Open Source Computer Vision Library (OpenCV). The system consists of two parts. The first part is the detection stage, which is used to detect the signs from the whole image frame and includes the modules: data-image acquisition, image pre-processing and sign detection. During this stage, the impact of Canny edge detector and Hough transform parameters on the quality-level of sign detection was tested. The second part is the recognition stage, whose role is to match the detected object with a priori models of signs in the dataset. In the research, the authors also compared the influence of various image processing algorithms parameters to the time of road sign recognition. The discussion part answers also the question whether the mobile system (smartphone) is robust enough to detect and recognise road sings in real time.

A ROBUST STATISTICAL ALGORITHM FOR BOUNDARY DETECTION IN LIQUID SPRAYS

This work presents a statistical algorithm to detect phase boundaries in liquid sprays. Experimental data were obtained in a nonevaporating acetone spray at atmospheric conditions to minimize the refractive index gradients and associated beam steering. Mie scattering and diffuse background illumination (DBI) diagnostics are implemented to demonstrate the robustness of the approach. The analysis shows that the present algorithm yields similar results for the two different diagnostics, unlike existing threshold-based methodologies providing inconsistent results. The algorithm can be used to identify partial liquid boundaries between single-phase liquid and vapor boundaries. Thus, for each case, liquid boundaries were defined in terms of liquid probability of 100% or less than 100%. Sprays from two different injectors were analyzed to demonstrate the versatility of the methodology. A parametric study was conducted to demonstrate that motion blur can affect an accurate detection of the phase boundaries, and hence, the exposure time in the experiment must be sufficiently small.