Predicting Facial Attractiveness from Colour Cues: A New Analytic Framework.

Various facial colour cues were identified as valid predictors of facial attractiveness, yet the conventional univariate approach has simplified the complex nature of attractiveness judgement for real human faces. Predicting attractiveness from colour cues is difficult due to the high number of candidate variables and their inherent correlations. Using datasets from Chinese subjects, this study proposed a novel analytic framework for modelling attractiveness from various colour characteristics. One hundred images of real human faces were used in experiments and an extensive set of 65 colour features were extracted. Two separate attractiveness evaluation sets of data were collected through psychophysical experiments in the UK and China as training and testing datasets, respectively. Eight multivariate regression strategies were compared for their predictive accuracy and simplicity. The proposed methodology achieved a comprehensive assessment of diverse facial colour features and their role in attractiveness judgements of real faces; improved the predictive accuracy (the best-fit model achieved an out-of-sample accuracy of 0.66 on a 7-point scale) and significantly mitigated the issue of model overfitting; and effectively simplified the model and identified the most important colour features. It can serve as a useful and repeatable analytic tool for future research on facial impression modelling using high-dimensional datasets.

Optimum display luminance and contrast polarity of desktop head-up display under office lighting level based on visual ergonomic study.

New type desktop head-up display (HUD) can reduce visual fatigue and protect vision through long viewing distance. In this study, participants evaluated visual performance, fatigue, and discomfort of desktop HUD under two contrast polarity (N = 36) and five display luminance levels (N = 21). A positive polarity advantage was found over negative in visual fatigue and discomfort (p < .05). Statistically significant effect of luminance was found on visual performance, fatigue, and discomfort (p < .05). The calculated optimum display luminance by the proposed inverted-U fitted model was 153 cd/m2 under 300 lx, higher than that of traditional desktop displays. It is speculated that higher luminance is required to offset the reduction in contrast sensitivity due to smaller target angular size, which caused by longer viewing distance. These findings suggest that positive polarity and 153 cd/m2 can be used to improve performance and avoid fatigue and discomfort when utilising desktop HUD under 300 lx.

A visual ergonomic study was conducted on new type desktop HUD. With visual performance, fatigue, and discomfort as optimisation goals, positive polarity and 153 cd/m² was recommended under 300 lx. Compared with traditional desktop displays, higher luminance was required in desktop HUD under the same illumination.

Optimum display luminance under a wide range of ambient light for cockpit displays.

The self-luminous cockpit displays need to be adaptive to a wide range of ambient light levels, which changes from very low illuminance to very high levels. Yet, current studies on evaluation and luminance setting of displays in bright surroundings are still limited. In this study, a three-dimensional visual ergonomic experiment was carried out to investigate how bright a cockpit display should be to meet aircrew operational requirements under different illuminance. A lab study with a within-subjects (N = 12) design was conducted in a simulated cockpit. According to the Weber-Fechner's Law, human observers evaluated five display luminance conditions (101, 101.5, 102, 102.5, 103 cd/m2) under five ambient illuminance conditions (10°, 101, 102, 103, 104 lx). Visual performance, visual fatigue and visual comfort were used as evaluation bases, which were measured by d2 task, subjective fatigue questionnaire and visual perception semantic scales. Nonlinear function fitting was used to calculate the optimal luminance under a certain illuminance. Finally, curvilinear regression was used to analyze the illuminance and its corresponding optimal luminance. Based on Silverstein luminance power function, a luminance adjustment model with the form of power function was obtained. The proposed three-dimensional model fits the experimental data well and is consistent with the existing studies. It can be regarded as a supplement and optimization of the previous model under high ambient illuminance. This study can contribute not only to the pleasing luminance setting of panel displays in aircraft cockpits but also to other self-luminous devices, such as tablet devices, outdoor monitoring equipment and advertising screens.

Multi-objective optimization method for reducing mutual interference in cockpit illumination.

The cockpit is a compact space comprised of various light-emitting devices. The light from different devices interferes and overlaps on the target surface. The light distribution requirements of different target surfaces are different. A suitable decision-making process is required to simultaneously meet the requirements of multiple target surfaces. A GPR-NSGA-II framework was proposed in the present study and a corresponding Gaussian process regression prediction model was established to predict and optimize multiple optical quality parameters in the cockpit. The luminous flux and beam angle of the typical luminaires were selected as controlled input parameters in a model case. The average illumination of targets that need lighting were set as constraints, and uniformity of illuminance of these surfaces and vertical illumination (direct light) of the eye position were set as the variables. An orthogonal experiment was conducted using the lighting model and a dataset was generated to validate the proposed framework. The results demonstrate that the solution set of luminescence parameters in cockpit illumination can be specified by GPR-NSGA-II framework.

Study of composite Al2O3-Ce:Y3Mg1.8Al1.4Si1.8O12 ceramic phosphors.

The nominal composition of Al2O3-Ce:Y3Mg1.8Al1.4Si1.8O12 (A-Ce:YMASG) ceramic phosphors was fabricated by the vacuum sintering technique. The introduction of Al2O3 as a second phase partially enters the crystal lattice, which was confirmed by the composition changing of the samples through x-ray diffraction measurement. An impurity phase of Y4MgSi3O13 was observed in Ce:YMASG and disappeared with the introduction of Al2O3 at the concentration of 10 wt. %. When the content of Al2O3 increased to 30 and 50 wt. %, the second phase of Al2O3 was measured with actual weight ratios of 7.72 and 20.55 wt. %, respectively. The third phase of MgAl2O4 was found with the further addition of Al2O3at 70 wt. %; the weight ratios of Ce:YMASG, Al2O3, and MgAl2O4 were 68.756, 18.457, and 12.787 wt. %, respectively. The luminescent characters of the samples were measured by the photoluminescence spectra and electroluminescent spectra. With the increase of Al2O3 from 0 to 30 wt. %, the emission wavelength of Ce3+ plummeted from 610 to 552 nm, and the luminous efficacy of the samples increased from 35 to 65 lm/W.

Model predicting discomfort glare caused by LED road lights.

To model discomfort glare from LED road lighting, the effect of four key variables on perceived glare was explored. These variables were: the average glare source luminance (Lg), the background luminance (Lb), the solid angle of the glare source from the perspective of the viewer (ω) and the angle between the glare source and the line of sight (θ). Based on these four variables 72 different light conditions were simulated in a scaled experimental set-up. Participants were requested to judge the perceived discomfort glare of these light conditions using the deBoer rating scale. All four variables and some of their interactions had indeed a significant effect on the deBoer rating. Based on these findings, we developed a model, and tested its general applicability in various verification experiments, including laboratory conditions as well as real road conditions. This verification proved the validity of the model with a correlation between measured and predicted values as high as 0.87 and a residual deviation of about 1 unit on the deBoer rating scale. These results filled the gap in estimating discomfort glare of LED road lighting and clarified similarities of and differences in discomfort glare between LED and traditional light sources.

Cross-cultural variation of memory colors of familiar objects.

The effect of cross-regional or cross-cultural differences on color appearance ratings and memory colors of familiar objects was investigated in seven different countries/regions - Belgium, Hungary, Brazil, Colombia, Taiwan, China and Iran. In each region the familiar objects were presented on a calibrated monitor in over 100 different colors to a test panel of observers that were asked to rate the similarity of the presented object color with respect to what they thought the object looks like in reality (memory color). For each object and region the mean observer ratings were modeled by a bivariate Gaussian function. A statistical analysis showed significant (p < 0.001) differences between the region average observers and the global average observer obtained by pooling the data from all regions. However, the effect size of geographical region or culture was found to be small. In fact, the differences between the region average observers and the global average observer were found to of the same magnitude or smaller than the typical within region inter-observer variability. Thus, although statistical differences in color appearance ratings and memory between regions were found, regional impact is not likely to be of practical importance.

Non-destructive assessment of soluble solids content in kiwifruit using hyperspectral imaging coupled with feature engineering.

The maturity of kiwifruit is widely gauged by its soluble solids content (SSC), with accurate assessment being essential to guarantee the fruit's quality. Hyperspectral imaging offers a non-destructive alternative to traditional destructive methods for SSC evaluation, though its efficacy is often hindered by the redundancy and external disturbances of spectral images. This study aims to enhance the accuracy of SSC predictions by employing feature engineering to meticulously select optimal spectral features and mitigate disturbance effects. We conducted a comprehensive investigation of four spectral pre-processing and nine spectral feature selection methods, as components of feature engineering, to determine their influence on the performance of a linear regression model based on ordinary least squares (OLS). Additionally, the stacking generalization technique was employed to amalgamate the strengths of the two most effective models derived from feature engineering. Our findings demonstrate a considerable improvement in SSC prediction accuracy post feature engineering. The most effective model, when considering both feature engineering and stacking generalization, achieved an RMSEp of 0.721, a MAPEp of 0.046, and an RPDp of 1.394 in the prediction set. The study confirms that feature engineering, especially the careful selection of spectral features, and the stacking generalization technique are instrumental in bolstering SSC prediction in kiwifruit. This advancement enhances the application of hyperspectral imaging for quality assessment, offering benefits that extend across the agricultural industry.

Evidence for human-centric in-vehicle lighting: part 3-Illumination preferences based on subjective ratings, eye-tracking behavior, and EEG features.

Within this third part of our mini-series, searching for the best and worst automotive in-vehicle lighting settings, we aim to extend our previous finding about white light illumination preferences by adding local cortical area activity as one key indicator. Frontal electrical potential asymmetry, measured using an electroencephalogram (EEG), is a highly correlated index for identifying positive and negative emotional behavior, primarily in the alpha band. It is rarely understood to what extent this observation can be applied to the evaluation of subjective preference or dislike based on luminaire variations in hue, chroma, and lightness. Within a controlled laboratory study, we investigated eight study participants who answered this question after they were shown highly immersive 360° image renderings. By so doing, we first subjectively defined, based on four different external driving scenes varying in location and time settings, the best and worst luminaire settings by changing six unlabeled luminaire sliders. Emotional feedback was collected based on semantic differentials and an emotion wheel. Furthermore, we recorded 120 Hz gaze data to identify the most important in-vehicle area of interest during the luminaire adaptation process. In the second study session, we recorded EEG data during a binocular observation task of repeated images arbitrarily paired by previously defined best and worst lighting settings and separated between all four driving scenes. Results from gaze data showed that the central vehicle windows with the left-side orientated colorful in-vehicle fruit table were both significantly longer fixed than other image areas. Furthermore, the previously identified cortical EEG feature describing the maximum power spectral density could successfully separate positive and negative luminaire settings based only on cortical activity. Within the four driving scenes, two external monotonous scenes followed trendlines defined by highly emotionally correlated images. More interesting external scenes contradicted this trend, suggesting an external emotional bias stronger than the emotional changes created by luminaires. Therefore, we successfully extended our model to define the best and worst in-vehicle lighting with cortical features by touching the field of neuroaesthetics.

A Novel Framework of Manifold Learning Cascade-Clustering for the Informative Frame Selection.

Narrow band imaging is an established non-invasive tool used for the early detection of laryngeal cancer in surveillance examinations. Most images produced from the examination are useless, such as blurred, specular reflection, and underexposed. Removing the uninformative frames is vital to improve detection accuracy and speed up computer-aided diagnosis. It often takes a lot of time for the physician to manually inspect the informative frames. This issue is commonly addressed by a classifier with task-specific categories of the uninformative frames. However, the definition of the uninformative categories is ambiguous, and tedious labeling still cannot be avoided. Here, we show that a novel unsupervised scheme is comparable to the current benchmarks on the dataset of NBI-InfFrames. We extract feature embedding using a vanilla neural network (VGG16) and introduce a new dimensionality reduction method called UMAP that distinguishes the feature embedding in the lower-dimensional space. Along with the proposed automatic cluster labeling algorithm and cost function in Bayesian optimization, the proposed method coupled with UMAP achieves state-of-the-art performance. It outperforms the baseline by 12% absolute. The overall median recall of the proposed method is currently the highest, 96%. Our results demonstrate the effectiveness of the proposed scheme and the robustness of detecting the informative frames. It also suggests the patterns embedded in the data help develop flexible algorithms that do not require manual labeling.

A Novel Distribution for Representation of 6D Pose Uncertainty.

The 6D Pose estimation is a crux in many applications, such as visual perception, autonomous navigation, and spacecraft motion. For robotic grasping, the cluttered and self-occlusion scenarios bring new challenges to the this field. Currently, society uses CNNs to solve this problem. The CNN models will suffer high uncertainty caused by the environmental factors and the object itself. These models usually maintain a Gaussian distribution, which is not suitable for the underlying manifold structure of the pose. Many works decouple rotation from the translation and quantify rotational uncertainty. Only a few works pay attention to the uncertainty of the 6D pose. This work proposes a distribution that can capture the uncertainty of the 6D pose parameterized by the dual quaternions, meanwhile, the proposed distribution takes the periodic nature of the underlying structure into account. The presented results include the normalization constant computation and parameter estimation techniques of the distribution. This work shows the benefits of the proposed distribution, which provides a more realistic explanation for the uncertainty in the 6D pose and eliminates the drawback inherited from the planar rigid motion.

Evidence for human-centric in-vehicle lighting: Part 2-Modeling illumination based on color-opponents.

Illumination preference models are usually defined in a static scenery, rating common-colored objects by a single scale or semantic differentials. Recently, it was reported that two to three illumination characteristics are necessary to define a high correlation in a bright office-like environment. However, white-light illumination preferences for vehicle-occupants in a dynamic semi- to full automated modern driving context are missing. Here we conducted a global free access online survey using VR engines to create 360° sRGB static in-vehicle sceneries. A total of 164 participants from China and Europe answered three levels in our self-hosted questionnaire by using mobile access devices. First, the absolute perceptional difference should be defined by a variation of CCT for 3,000, 4,500, and 6,000 K or combinations, and light distribution, either in a spot- or spatial way. Second, psychological light attributes should be associated with the same illumination and scenery settings. Finally, we created four driving environments with varying external levels of interest and time of the day. We identified three key results: (1) Four illumination groups could be classified by applying nMDS. (2) Combinations of mixed CCTs and spatial light distributions outperformed compared single light settings (p < 0.05), suggesting that also during daylight conditions artificial light supplements are necessary. (3) By an image transformation in the IPT and CAM16 color appearance space, comparing external and in-vehicle scenery, individual illumination working areas for each driving scenery could be identified, especially in the dimension of chroma-, partially following the Hunt-Effect, and lightness contrast, which synchronizes the internal and external brightness level. We classified our results as a starting point, which we intend to prove in a follow-up-controlled laboratory study with real object arrangements. Also, by applying novel methods to display high fidelity 360° rendered images on mobile access devices, our approach can be used in the future interdisciplinary research since high computational mobile devices with advanced equipped sensory systems are the new standard of our daily life.

Diurnal Circadian Lighting Accumulation Model: A Predictor of the Human Circadian Phase Shift Phenotype.

Light is an important external factor that affects human circadian rhythms. This study aimed to explore the effects of different dimensions of diurnal light exposure on the physiological circadian phase shift (CPS) of the human body. A strict light exposure experiment with different timing schemes (8:00-12:00, 13:00-17:00, 18:00-22:00), durations (4 h, 8 h) and effective circadian stimulus levels (circadian stimulus: 0.35, 0.55) was performed in an enclosed laboratory. Fourteen participants, including seven males and seven females, with a mean age of 24.29 ± 2.43 (mean ± standard deviation), participated in this experiment and experienced all six lighting schemes. The results showed that both time factor (F 3,40 = 29.079, p < 0.001, the power of the sample size = 0.98) and circadian stimulus levels (T 20 = - 2.415, p = 0.025, the power of sample size = 0.76) significantly affect the CPS. On this basis, a diurnal circadian lighting accumulation (DCLA)-CPS model was proposed in the form of the Boltzmann function, and was validated by experimental data with high correlation (R 2 = 0.9320, RSS = 0.1184), which provides strong support for rationally arranging the light level at different times of the day.

High circadian stimulus lighting therapy for depression: Meta-analysis of clinical trials.

Although bright light therapy (BLT) has been widely used in the clinical treatment of depression, the antidepressant effect of BLT is not well understood. Considering the connection between depression and disrupted circadian rhythm, we assumed the model of human circadian phototransduction could be more accurate in evaluating the efficacy of BLT for depression compared to light level and spectrum. A systematic review and meta-analysis were conducted and the CS (circadian stimulus) model was used to quantify the efficacy of lighting in BLT. Articles published up to June 2022 were searched in COCHRANE, EMBASE, MEDLINE, and Web of Science. Randomized clinical trials included articles using high circadian stimulus (H-CS, CS > 0.1) as lighting therapy for people with depressive disorder vs. a control group (CS < 0.1). The treatment effect was estimated by calculating the mean difference (MD) with 95% confidence intervals (CIs). Seven trials involving 258 participants met the inclusion criteria. In this sample size, H-CS lighting was associated with a significant reduction in depressive symptoms (MD = -5.56, 95% CI = -9.22 to -1.90, P = 0.003, I 2 = 64%). According to the meta-analysis, CS can be employed for the clinical evaluation of BLT for patients with depressive disorder and exposure to H-CS lighting significantly reduced depressive symptoms among adults. A range of CS > 0.57 was obtained, during which different lighting parameter combinations (e.g., light levels, spectra, duration, and light distribution) could achieve better treatment for depression. Systematic review registration: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021253648.

Is the use of high correlated color temperature light at night related to delay of sleep timing in university students? A cross-country study in Japan and China.

BACKGROUND: Blue-enriched white light at night has the potential to delay the circadian rhythm in daily life. This study was conducted to determine whether the use of high correlated color temperature (CCT) light at home at night is associated with delay of sleep timing in university students. METHODS: The survey was conducted in 2014-2015 in 447 university students in Japan and 327 students in China. Habitual sleep timing and type of CCT light at home were investigated by using a self-administered questionnaire. The Japanese students were significantly later than the Chinese students in bedtime, wake time, and midpoint of sleep. They were asked whether the lighting in the room where they spend most of their time at night was closer to warm color (low CCT) or daylight color (high CCT). The amount of light exposure level during daily life was measured for at least 1 week by the use of a light sensor in 60 students in each country. RESULTS: The percentages of participants who used high CCT lighting at night were 61.6% for Japanese students and 80.8% for Chinese students. Bedtime and sleep onset time on school days and free days were significantly later in the high CCT group than in the low CCT group in Japan. The midpoint of sleep in the high CCT group was significantly later than that in the low CCT group on free days but not on school days. On the other hand, none of the sleep measurements on school days and free days were significantly different between the high CCT and low CCT groups in China. Illuminance level of light exposure during the night was significantly higher in Japanese than in Chinese, but that in the morning was significantly higher in China than in Japan. CONCLUSIONS: The use of high CCT light at night is associated with delay of sleep timing in Japanese university students but not in Chinese university students. The effects of light at night on sleep timing and circadian rhythm may be complicated by other lifestyle factors depending on the country.

Eye movement and pupil size constriction under discomfort glare.

PURPOSE: Involuntary physiological responses offer an alternative means to psychophysical procedures for objectively evaluating discomfort glare. This study examined eye movement and pupil size responses to glare discomfort using new approaches to analysis: relative pupil size and speed of eye movement. METHODS: Participants evaluated glare discomfort using the standard de Boer rating scale under various conditions manipulated to influence glare discomfort. Eye movement was recorded using an electro-oculogram (EOG), and pupil size was recorded using Tobii glasses. Ten young (mean age: 24.5 years old) and 10 senior (mean age: 61 years old) participants were recruited for this experiment. RESULTS: Subjective evaluation of glare discomfort was highly correlated with eye movement (multiple correlation coefficient [R(2)] of >0.94, P < 0.001) and pupil constriction (R(2) = 0.38, P < 0.001). Severe glare discomfort increased the speed of eye movement and caused larger pupil constriction. Larger variations of eye movement were found among seniors. CONCLUSIONS: The two physiological responses studied here to characterize discomfort glare under various lighting conditions had significant correlation with the subjective evaluation. The correlation between discomfort glare and physiological responses suggests an objective way to characterize and evaluate discomfort glare that may overcome the problems of conventional subjective evaluation. It also offers an explanation as to why long-term exposure to discomfort glare leads to visual fatigue and eyestrain.

Transient effects of harsh luminous conditions on the visual performance of aviators in a civil aircraft cockpit.

The aim of this work was to examine how harsh luminous conditions in a cockpit, such as lightning in a thunderstorm or direct sunlight immediately after an aircraft passes through clouds, may affect the visual performance of pilots, and how to improve it. Such lighting conditions can result in the temporary visual impairment of aviators, which may greatly increase the risk of accidents. Tests were carried out in a full-scale simulator cockpit in which two kinds of dynamic lighting scenes, namely pulse changed and step changed lighting, were used to represent harsh luminous conditions. Visual acuity (VA), reaction time (RT) and identification accuracy (IA) were recorded as dependent variables. Data analysis results indicate that standardized VA values decreased significantly in both pulsing and step conditions in comparison with the dark condition. Standardized RT values increased significantly in the step condition; on the contrary, less reaction time was observed in the pulsing condition. Such effects could be reduced by an ambient illumination provided by a fluorescent lamp in both conditions. The results are to be used as a principle for optimizing lighting design with a thunderstorm light.