Pedestrian Detection Using Multispectral Images and a Deep Neural Network.

Pedestrian fatalities and injuries most likely occur in vehicle-pedestrian crashes. Meanwhile, engineers have tried to reduce the problems by developing a pedestrian detection function in Advanced Driver-Assistance Systems (ADAS) and autonomous vehicles. However, the system is still not perfect. A remaining problem in pedestrian detection is the performance reduction at nighttime, although pedestrian detection should work well regardless of lighting conditions. This study presents an evaluation of pedestrian detection performance in different lighting conditions, then proposes to adopt multispectral image and deep neural network to improve the detection accuracy. In the evaluation, different image sources including RGB, thermal, and multispectral format are compared for the performance of the pedestrian detection. In addition, the optimizations of the architecture of the deep neural network are performed to achieve high accuracy and short processing time in the pedestrian detection task. The result implies that using multispectral images is the best solution for pedestrian detection at different lighting conditions. The proposed deep neural network accomplishes a 6.9% improvement in pedestrian detection accuracy compared to the baseline method. Moreover, the optimization for processing time indicates that it is possible to reduce 22.76% processing time by only sacrificing 2% detection accuracy.

Analysis of Different Illuminance of the Room Lighting Condition on the Accuracy (Trueness and Precision) of An Intraoral Scanner.

PURPOSE: To measure the influence of illuminance of the ambient light between 1000 lux (room light) and 10 000 lux (chair light) on the accuracy of an intraoral scanner (IOS). MATERIAL AND METHODS: A typodont was digitized using an extraoral scanner to obtain a reference standard tessellation language (STL) file. Ten groups were created based on the different illuminance of the ambient light conditions tested starting from 1000 lux (no chair light) to 10000 lux (chair light) in increments of 1000 lux by increasing the distance between the chair light and the mannequin, with the room light turned on. Ten digital scans per group were obtained (n = 10) using an IOS (Trios 3; 3Shape). The accuracy of the digital scans was evaluated with respect to the reference mesh of the typodont using a 3D mesh processing software. Kruskal-Wallis and pair-wise comparison tests were used to analyze the data (α = 0.05). RESULTS: Significant difference for trueness and precision values were found among the groups (p < 0.001). The 1000-lux group exhibited the lowest discrepancy values with a median of 26.33 µm and an interquartile range (IQR) of 40.04 µm (11.97-52.00) (p < 0.001); while the 5000-lux group obtained the highest discrepancy values with a median of 46.38 µm and an IQR of 99.94 µm (19.05-118.98) (p < 0.001). The pair-wise multi-comparison showed no difference between the 8000- and 4000-lux groups (p = 0.287). In all groups, the IQR was higher than the mean errors from the control mesh, suggesting that the relative precision was low. CONCLUSIONS: A 1000-lux illumination lighting condition is recommended to maximize the scanning accuracy of the IOS tested; the chair light should be avoided. Furthermore, the scanning accuracy response under the illuminance range tested presented a lack of monotonicity.

Clinical Study of the Influence of Ambient Lighting Conditions on the Mesh Quality of an Intraoral Scanner.

PURPOSE: To compute the effect of ambient light illuminance settings on the mesh quality of the digital scans accomplished in a subject. MATERIAL AND METHODS: A subject was recruited. The maxillary dentition did not present any dental restoration. A prosthodontist recorded different complete-arch maxillary digital scans by using an IOS (TRIOS 3; 3Shape) under 4 different illuminance light conditions namely chair light at 10,000-lux illuminance (CL group), room light at 1000-lux illuminance (RL group), natural light at 500-lux illuminance (NL group), and no light at 0-lux luminosity (ZL group). Ten digital scans for each group were consecutively obtained. Mesh quality was examined using the iso2mesh MATLAB package. Shapiro-Wilk test revealed a nonnormally distributed data. Kruskal-Wallis one-way ANOVA, and pair-wise comparison were selected to evaluate the data (α = 0.05). RESULTS: Significant differences in mesh quality values were measured among the groups (p < 0.001). Pair-wise comparisons revealed that significant difference was found across all pairs of lighting groups, except for the RL-NL comparison (p = 0.279). However, the CL condition obtained the highest mean values, followed by RL and NL groups, and the lowest mean values were obtained on the ZL lighting condition. CONCLUSIONS: Chair light at 10,000-lux illuminance condition is recommended to maximize the quality mesh values of the IOS system tested (TRIOS 3; 3Shape).

Seasonal Variation in Bright Daylight Exposure, Mood and Behavior among a Group of Office Workers in Sweden.

The purpose of the study was to investigate seasonal variation in mood and behavior among a group of office workers in Sweden (56°N). Thirty subjects participated in this longitudinal study. The subjects kept a weekly log that included questionnaires for ratings of psychological wellbeing and daily sleep-activity diaries where they also noted time spent outdoors. The lighting conditions in the offices were subjectively evaluated during one day, five times over the year. There was a seasonal variation in positive affect and in sleep-activity behavior. Across the year, there was a large variation in the total time spent outdoors in daylight. The subjects reported seasonal variation concerning the pleasantness, variation and strength of the light in the offices and regarding the visibility in the rooms. Finally, the subjects spent most of their time indoors, relying on artificial lighting, which demonstrates the importance of the lighting quality in indoor environments.

The effect of lighting conditions and use of headlights on drivers' perception and appraisal of approaching vehicles at junctions.

Use of Daytime Running Lights (DRL) is mandatory in many countries for motorcycles, and in some for cars. However, in developing countries, DRLs may be optional or compliance low. The effect of car or motorcycle headlights and lighting conditions on Malaysian drivers' ability to perceive and judge the safety of pulling out was investigated. Stimuli were photographs depicting either daytime or nighttime taken at a T-junction with approaching vehicles with headlights on or off. Headlights improved drivers' ability to perceive cars and motorcycles in the nighttime photographs but not the daytime photographs, although this could be due to the bright weather in the photographs. Drivers judged it less safe to pull out when approaching motorcycles had headlights on than off, regardless of the lighting conditions, supporting the utility of DRL for motorcycles. Headlights did not affect judgements for cars, questioning the utility of DRL for cars. Practitioner Summary: The effect of headlights and lighting conditions on drivers' ability to perceive and make judgements about the safety of pulling out was investigated. Daytime Running Lights influenced drivers' decision-making about the safety of pulling out in front of motorcycles, illustrating the importance of having automatic headlights equipped.

Tracking and Classification of In-Air Hand Gesture Based on Thermal Guided Joint Filter.

The research on hand gestures has attracted many image processing-related studies, as it intuitively conveys the intention of a human as it pertains to motional meaning. Various sensors have been used to exploit the advantages of different modalities for the extraction of important information conveyed by the hand gesture of a user. Although many works have focused on learning the benefits of thermal information from thermal cameras, most have focused on face recognition or human body detection, rather than hand gesture recognition. Additionally, the majority of the works that take advantage of multiple modalities (e.g., the combination of a thermal sensor and a visual sensor), usually adopting simple fusion approaches between the two modalities. As both thermal sensors and visual sensors have their own shortcomings and strengths, we propose a novel joint filter-based hand gesture recognition method to simultaneously exploit the strengths and compensate the shortcomings of each. Our study is motivated by the investigation of the mutual supplementation between thermal and visual information in low feature level for the consistent representation of a hand in the presence of varying lighting conditions. Accordingly, our proposed method leverages the thermal sensor's stability against luminance and the visual sensors textural detail, while complementing the low resolution and halo effect of thermal sensors and the weakness against illumination of visual sensors. A conventional region tracking method and a deep convolutional neural network have been leveraged to track the trajectory of a hand gesture and to recognize the hand gesture, respectively. Our experimental results show stability in recognizing a hand gesture against varying lighting conditions based on the contribution of the joint kernels of spatial adjacency and thermal range similarity.

Seeing "the Dress" in the Right Light: Perceived Colors and Inferred Light Sources.

In the well-known "dress" photograph, people either see the dress as blue with black stripes or as white with golden stripes. We suggest that the perception of colors is guided by the scene interpretation and the inferred positions of light sources. We tested this hypothesis in two online studies using color matching to estimate the colors observers see, while controlling for individual differences in gray point bias and color discrimination. Study 1 demonstrates that the interpretation of the dress corresponds to differences in perceived colors. Moreover, people who perceive the dress as blue-and-black are two times more likely to consider the light source as frontal, than those who see the white-and-gold dress. The inferred light sources, in turn, depend on the circadian changes in ambient light. The interpretation of the scene background as a wall or a mirror is consistent with the perceived colors as well. Study 2 shows that matching provides reliable results on differing devices and replicates the findings on scene interpretation and light sources. Additionally, we show that participants' environmental lighting conditions are an important cue for perceiving the dress colors. The exact mechanisms of how environmental lighting and circadian changes influence the perceived colors of the dress deserve further investigation.

Asymbiotic in vitro seed propagation of Dendrobium.

The ability to germinate orchids from seeds in vitro presents a useful and viable method for the propagation of valuable germplasm, maintaining the genetic heterogeneity inherent in seeds. Given the ornamental and medicinal importance of many species within the genus Dendrobium, this review explores in vitro techniques for their asymbiotic seed germination. The influence of abiotic factors (such as temperature and light), methods of sterilization, composition of basal media, and supplementation with organic additives and plant growth regulators are discussed in context to achieve successful seed germination, protocorm formation, and further seedling growth and development. This review provides both a basis for the selection of optimal conditions, and a platform for the discovery of better ones, that would allow the development of new protocols and the exploration of new hypotheses for germination and conservation of Dendrobium seeds and seedlings.

Skin Tone Estimation under Diverse Lighting Conditions.

Knowledge of a person's level of skin pigmentation, or so-called "skin tone", has proven to be an important building block in improving the performance and fairness of various applications that rely on computer vision. These include medical diagnosis of skin conditions, cosmetic and skincare support, and face recognition, especially for darker skin tones. However, the perception of skin tone, whether by the human eye or by an optoelectronic sensor, uses the reflection of light from the skin. The source of this light, or illumination, affects the skin tone that is perceived. This study aims to refine and assess a convolutional neural network-based skin tone estimation model that provides consistent accuracy across different skin tones under various lighting conditions. The 10-point Monk Skin Tone Scale was used to represent the skin tone spectrum. A dataset of 21,375 images was captured from volunteers across the pigmentation spectrum. Experimental results show that a regression model outperforms other models, with an estimated-to-target distance of 0.5. Using a threshold estimated-to-target skin tone distance of 2 for all lights results in average accuracy values of 85.45% and 97.16%. With the Monk Skin Tone Scale segmented into three groups, the lighter exhibits strong accuracy, the middle displays lower accuracy, and the dark falls between the two. The overall skin tone estimation achieves average error distances in the LAB space of 16.40±20.62.

Render lighting dataset: A collection of rendered images with varied lighting conditions using blender render engines.

The quality of datasets is crucial in computer graphics and machine learning research and development. This paper presents the Render Lighting Dataset, featuring 63,648 rendered images of Blender's primitive shapes with various lighting conditions and engines. The images were created using Blender 4.0's Cycles and Eevee Render Engines, with careful attention to detail in texture mapping and UV unwrapping. The dataset covers six different lighting conditions, including Area Light, Spotlight, Point Light, Tri-Light, HDRI (Sunlight), and HDRI (Overcast), each adjusted using Blender's different options in the Color Management panel. With thirteen unique materials, ranging from Coastal Sand to Glossy Plastic, the dataset provides visual diversity for researchers to explore material properties under different lighting conditions using different render engines. This dataset serves as a valuable resource for researchers looking to enhance 3D rendering engines. Its diverse set of rendered images under varied lighting conditions and material properties allows researchers to benchmark and evaluate the performance of different rendering engines, develop new rendering algorithms and techniques, optimize rendering parameters, and understand rendering challenges. By enabling more realistic and efficient rendering, advancing research in lighting simulation, and facilitating the development of AI-driven rendering techniques, this dataset has the potential to shape the future of computer graphics and rendering technology.

Enhancing work zone crash severity analysis: The role of synthetic minority oversampling technique in balancing minority categories.

Road work zones are becoming increasingly common due to the aging infrastructure and the need for capacity enhancement. They present significant safety risks due to narrow lanes, uneven traffic flow, lower speed, and reduced visibility. It is particularly important to understand the role of human behavioral factors in WZ crash injury severity due to difficulty navigating such areas. Furthermore, the crash injury data available is mostly imbalanced, primarily due to the lower incidence of high-cost fatal and severe injuries, and can benefit from the use of emerging analysis techniques. This research study examines a unique dataset comprising 7,855 WZ crashes in Tennessee from 2018 to 2022 as a case study to provide useful insight into the behavioral factors associated with injury severity and how they change after adjusting for the underrepresented fatal and serious injuries within the dataset. The study applies frequentist methods and a machine learning technique enhanced with the Synthetic Minority Oversampling Technique (SMOTE), addressing the data imbalance (relatively fewer fatal and serious injuries) for useful inferences and predictions. The study results indicate that aggressive driving, overspeeding, and drunk driving significantly elevate injury severity. Additionally, after balancing the minority categories of crash injury severity levels, the importance of contributing factors changes. The study offers engineers and data analysts a framework for analyzing imbalanced data, a prevalent issue in crash injury severity analysis. By exploring key behavioral factors responsible for injury severity in WZ crashes, the study provides useful insight and valuable information to traffic safety engineers, transportation agencies, and policymakers to implement enhanced safety measures in WZ design and management, ultimately aiming to mitigate injury severity and to improve overall safety for road users.

Identifying roadway departure crash patterns on rural two-lane highways under different lighting conditions: Association knowledge using data mining approach.

INTRODUCTION: More than half of all fatalities on U.S. highways occur due to roadway departure (RwD) each year. Previous research has explored various risk factors that contribute to RwD crashes, however, a comprehensive investigation considering the effect of lighting conditions has been insufficiently addressed. DATA: Using the Louisiana Department of Transportation and Development crash database, fatal and injury RwD crashes occurring on rural two-lane (R2L) highways between 2008-2017 were analyzed based on daylight and dark (with and without streetlight). METHOD: This research employed a safe system approach to explore meaningful complex interactions among multidimensional crash risk factors. To accomplish this, an unsupervised data mining algorithm association rules mining (ARM) was utilized. RESULTS AND CONCLUSIONS: Based on the generated rules, the findings reveal several interesting crash patterns in the daylight, dark-with-streetlight, and dark-no-streetlight, emphasizing the importance of investigating RwD crash patterns depending on the lighting conditions. In daylight condition, fatal RwD crashes are associated with cloudy weather conditions, distracted drivers, standing water on the roadway, no seat belt use, and construction zones. In dark lighting condition (with and without streetlight), the majority of the RwD crashes are associated with alcohol/drug involvement, young drivers (15-24 years), driver condition (e.g., inattentive, distracted, illness/fatigued/asleep), and colliding with animal(s). PRACTICAL APPLICATIONS: The findings also reveal how certain driver behavior patterns are connected to RwD crashes, such as a strong association between alcohol/drug intoxication and no seat belt usage in the dark-no-streetlight condition. Based on the identified crash patterns and behavioral characteristics under different lighting conditions, the findings could aid researchers and safety specialists in developing the most effective RwD crash mitigation strategies.

Real-Time Detection and Classification of Scirtothrips dorsalis on Fruit Crops with Smartphone-Based Deep Learning System: Preliminary Results.

This study proposes a deep-learning-based system for detecting and classifying Scirtothrips dorsalis Hood, a highly invasive insect pest that causes significant economic losses to fruit crops worldwide. The system uses yellow sticky traps and a deep learning model to detect the presence of thrips in real time, allowing farmers to take prompt action to prevent the spread of the pest. To achieve this, several deep learning models are evaluated, including YOLOv5, Faster R-CNN, SSD MobileNetV2, and EfficientDet-D0. EfficientDet-D0 was integrated into the proposed smartphone application for mobility and usage in the absence of Internet coverage because of its smaller model size, fast inference time, and reasonable performance on the relevant dataset. This model was tested on two datasets, in which thrips and non-thrips insects were captured under different lighting conditions. The system installation took up 13.5 MB of the device's internal memory and achieved an inference time of 76 ms with an accuracy of 93.3%. Additionally, this study investigated the impact of lighting conditions on the performance of the model, which led to the development of a transmittance lighting setup to improve the accuracy of the detection system. The proposed system is a cost-effective and efficient alternative to traditional detection methods and provides significant benefits to fruit farmers and the related ecosystem.

Optical and motor changes associated with lighting and near vision tasks in electronic devices.

Purpose: To assess optical and motor changes associated with near vision reading under different controlled lighting conditions performed with two different types of electronic screens. Methods: Twenty-four healthy subjects with a mean age of 22.9±2.3 years (18- 33) participated in this study. An iPad and an e-ink reader were chosen to present calibrated text, and each task lasted 5 minutes evaluating both ambient illuminance level and luminance of the screens. Results: Eye-tracker data revealed a higher number of saccadic eye movements under minimum luminance than under maximum luminance. The results showed statistically significant differences between the iPad (p=0.016) and the e-ink reader (p=0.002). The length of saccades was also higher for the minimum luminance level for both devices: 6.2±2.8 mm and 8.2±4.2 mm (e-ink max vs min), 6.8±2.9 mm and 7.6±3.6 mm (iPad max vs min), and blinking rate increased significantly for lower lighting conditions. Conclusions: Performing reading tasks on electronic devices is highly influenced by both the configuration of the screens and the ambient lighting, meanwhile, low differences in visual quality that are transient in healthy young people, were found.

Effect of Indoors Artificial Lighting Conditions on Computer-Based Learning Performance.

Lighting condition is essential to human performance. With the widespread use of computer-based learning, the performance measurements become difficult, and the effects of artificial lighting conditions towards the new learning forms are not investigated extensively. The current study conducts a subject-within experiment with a 45-min-long online learning along with electroencephalogram (EEG)-based measurements, and a post-interview under five lighting setups respectively (300 lx, 3000 K; 300 lx, 4000 K; 300 lx, 6500 K; 500 lx, 4000 K; 1000 lx, 4000 K). Attention is chosen as the key factor to represent the learning performance. The results show that the attention of people aged in the 20s is not affected by the experimental lighting conditions. The results also demonstrate that people in high illumination at 1000 lx are more inclined to sustain attention despite the discomfort and dissatisfaction. Taking the EEG-based attention measurements and post-interview answers into consideration, lighting conditions at 300 lx, 4000 K are the recommended set points for university architectures among the investigated conditions, providing a practical basis when adjusting the lighting standard for its advantage in energy saving.

The importance of including position and viewing direction when measuring and assessing the lighting conditions of office workers.

BACKGROUND: Light and alertness studies have applied different measurement methodologies to determine lighting conditions. However, it has been demonstrated that researchers rarely measure or describe the lighting conditions of their studies in sufficient detail to generalize conclusions or derive universal guidelines. OBJECTIVE: Part I of this paper summarizes the current measurement methodologies used in light and alertness studies to potentially identify methodological issues. Part II determines the differences in lighting conditions for different viewing directions within an office environment. METHODS: A literature review (part I) and both experimental studies and an observational study (part II) were undertaken in this study. RESULTS: Part I demonstrates that most light and alertness studies include photometric quantities; however, it is recommended that one should measure radiometric quantities as well. Further, the light measurements should be performed at the individual level. Part II demonstrates large differences in lighting conditions between viewing directions. For example, when looking toward the window, vertical illuminances were at least 12 times higher when compared to looking in the opposite direction. CONCLUSIONS: Our findings suggest that when analysing or designing an office environment, office workers' positions and viewing direction should be included in the determination of personal lighting conditions.

Highly Efficient Indoor Organic Solar Cells by Voltage Loss Minimization through Fine-Tuning of Polymer Structures.

Herein, we report a detailed study on the optoelectronic properties, photovoltaic performance, structural conformation, morphology variation, charge carrier mobility, and recombination dynamics in bulk heterojunction solar cells comprising a series of donor-acceptor conjugated polymers as electron donors based on benzodithiophene (BDT) and 5,8-bis(5-bromothiophen-2-yl)-6,7-difluoro-2,3-bis(3-(octyloxy)phenyl)quinoxaline as a function of the BDT's thienyl substitution (alkyl (WF3), alkylthio (WF3S), and fluoro (WF3F)). The synergistic positive effects of the fluorine substituents on the minimization of the bimolecular recombination losses, the reduction of the series resistances (RS), the increment of the shunt resistances (RSh), the suppression of the trap-assisted recombination losses, the balanced charge transport, the finer nanoscale morphology, and the deeper highest occupied molecular orbital (EHOMO) are manifested versus the alkyl and alkylthio substituents. According to these findings, the WF3F:[6,6]-phenyl-C71-butyric acid methyl ester (PC71BM)-based organic photovoltaic device is a rare example that features a high power conversion efficiency (PCE) of 17.34% under 500 lx indoor light-emitting diode light source with a high open-circuit voltage (VOC) of 0.69 V, due to the suppression of the voltage losses, and a PCE of 9.44% at 1 sun (100 mW/cm2) conditions, simultaneously.

Effects of Dietary and Lighting Conditions on Diurnal Locomotor Activity and Body Temperature in Microminipigs.

The effects of dietary and lighting conditions on diurnal rhythm of locomotor activity (LA) and body temperature (BT) using four adult male microminipigs were investigated. Different feeding times, diet and lighting conditions were applied sequentially for 3 weeks in each phase as follows: Phase I: Morning mealtime, normal diet, 12-h lights on; phase II: mealtime changed to afternoon; phase III: diet changed to high-fat diet; phase IV: lighting changed to 20-h on; and phase V: phase I repeated. LA was measured by an actigraph which was worn on the body of each pig. A BT recording module (Thermochron Type-SL) was implanted in the neck subcutaneously. Phase II increased BT compared with phase I. Phase III increased LA and BT compared with phase II. Phase IV increased LA compared with phase III. LA in phase V was higher compared with phase I. These results can be extrapolated to other diurnal animals such as humans. This study provides an example of the effects of diet and lighting on biological activities in microminipigs under low-invasive procedures measuring LA and BT, leading to low variations in these measures.

Gene expression profiling of astaxanthin and fatty acid pathways in Haematococcus pluvialis in response to different LED lighting conditions.

Haematococcus pluvialis is a green microalga of major interest to industry based on its ability to produce large amounts of astaxanthin. Biosynthesis of astaxanthin and its mono- and di-esters was significantly stimulated under 150 µmol m-2 s-1 of white LED (W-150) compared with lower light intensities, but the highest astaxanthin amounts were produced under 70 µmol m-2 s-1 of blue LED (B-70). Transcripts of astaxanthin biosynthesis genes psy, crtO, and bkt2 were upregulated under W-150, while psy, lcy, crtO, and crtR-B were upregulated by B-70. Total fatty acid content and biosynthesis genes fata and all dgat genes were induced under W-150, while C18:3n6 biosynthesis and dgat2a expression were specifically stimulated by B-70 which was correlated to astaxanthin ester biosynthesis. Nitrogen starvation, various LEDs and the identified upregulated genes may provide useful tools for future metabolic engineering to significantly increase free astaxanthin, its esters and fatty acid precursors in H. pluvialis.

Dim Light at Night and Constant Darkness: Two Frequently Used Lighting Conditions That Jeopardize the Health and Well-being of Laboratory Rodents.

The influence of light on mammalian physiology and behavior is due to the entrainment of circadian rhythms complemented with a direct modulation of light that would be unlikely an outcome of circadian system. In mammals, physiological and behavioral circadian rhythms are regulated by the suprachiasmatic nucleus (SCN) of the hypothalamus. This central control allows organisms to predict and anticipate environmental change, as well as to coordinate different rhythmic modalities within an individual. In adult mammals, direct retinal projections to the SCN are responsible for resetting and synchronizing physiological and behavioral rhythms to the light-dark (LD) cycle. Apart from its circadian effects, light also has direct effects on certain biological functions in such a way that the participation of the SCN would not be fundamental for this network. The objective of this review is to increase awareness, within the scientific community and commercial providers, of the fact that laboratory rodents can experience a number of adverse health and welfare outcomes attributed to commonly-used lighting conditions in animal facilities during routine husbandry and scientific procedures, widely considered as "environmentally friendly." There is increasing evidence that exposure to dim light at night, as well as chronic constant darkness, challenges mammalian physiology and behavior resulting in disrupted circadian rhythms, neural death, a depressive-behavioral phenotype, cognitive impairment, and the deregulation of metabolic, physiological, and synaptic plasticity in both the short and long terms. The normal development and good health of laboratory rodents requires cyclical light entrainment, adapted to the solar cycle of day and night, with null light at night and safe illuminating qualities during the day. We therefore recommend increased awareness of the limited information available with regards to lighting conditions, and therefore that lighting protocols must be taken into consideration when designing experiments and duly highlighted in scientific papers. This practice will help to ensure the welfare of laboratory animals and increase the likelihood of producing reliable and reproducible results.