Color Variability Research Articles

An image-based investigation on color associations among 100 Chinese emotion words

ABSTRACT Image data serves as a valuable resource for investigating relationships between colors and emotions. This study conducts an image-based visual corpus analysis on the color associations of 100 Chinese emotion words, aiming to uncover the pivotal roles of colors in understanding emotional concepts. The study addresses two primary objectives: (i) examining the interrelations among four affective properties (valence, arousal, prototypicality, and emotionality) and four image-based color attributes (Jz: a dimension depicting black–white color distinction, Az: a dimension for green-red, Bz: a dimension for blue–yellow, and color variability) associated with these emotion words; and (ii) assessing the efficacy of image-based color information in profiling affective (dis)similarities among different emotion words. The empirical results reveal (i) significant positive correlations between color variability and arousal, Jz and valence, Az and arousal, Bz and valence, as well as a negative correlation between Jz and prototypicality; (ii) the effectiveness of image-based color information in depicting the valence-dominated affective (dis)similarities among the 100 Chinese emotion words. This study contributes image-based empirical support to complement existing research on color-emotion mappings. Moreover, it advances methodological approaches by advocating for the utilisation of image data to address theoretical inquiries in cognitive science.

Highlighted Diffusion Model as Plug-in Priors for Polyp Segmentation.

Automated polyp segmentation from colonoscopy images is crucial for colorectal cancer diagnosis. The accuracy of such segmentation, however, is challenged by two main factors. First, the variability in polyps' size, shape, and color, coupled with the scarcity of well-annotated data due to the need for specialized manual annotation, hampers the efficacy of existing deep learning methods. Second, concealed polyps often blend with adjacent intestinal tissues, leading to poor contrast that challenges segmentation models. Recently, diffusion models have been explored and adapted for polyp segmentation tasks. However, the significant domain gap between RGB-colonoscopy images and grayscale segmentation masks, along with the low efficiency of the diffusion generation process, hinders the practical implementation of these models. To mitigate these challenges, we introduce the Highlighted Diffusion Model Plus (HDM+), a two-stage polyp segmentation framework. This framework incorporates the Highlighted Diffusion Model (HDM) to provide explicit semantic guidance, thereby enhancing segmentation accuracy. In the initial stage, the HDM is trained using highlighted ground-truth data, which emphasizes polyp regions while suppressing the background in the images. This approach reduces the domain gap by focusing on the image itself rather than on the segmentation mask. In the subsequent second stage, we employ the highlighted features from the trained HDM's U-Net model as plug-in priors for polyp segmentation, rather than generating highlighted images, thereby increasing efficiency. Extensive experiments conducted on six polyp segmentation benchmarks demonstrate the effectiveness of our approach.

A fully-coupled algorithm with implicit surface tension treatment for interfacial flows with large density ratios

The stability of most surface-tension-driven interfacial flow simulations is governed by the capillary time-step constraint. This concerns particularly small-scale flows and, more generally, highly-resolved liquid-gas simulations with moderate inertia. To date, the majority of interfacial-flow simulations are performed using an explicit surface-tension treatment, which restrains the performance of such simulations. Recently, an implicit treatment of surface tension able to breach the capillary time-step constraint using the volume-of-fluid (VOF) method was proposed, based on a fully-coupled pressure-based finite-volume algorithm. To this end, the interface-advection equation is incorporated implicitly into the linear flow solver, resulting in a tight coupling between all implicit solution variables (colour function, pressure, velocity). However, this algorithm is limited to uniform density and viscosity fields. Here, we present a fully-coupled algorithm for interfacial flows with implicit surface tension applicable to interfacial flows with large density and viscosity ratios. This is achieved by solving the continuity and momentum equations in conservative form, whereby the density is treated implicitly with respect to the colour function, and the advection term of the interface-advection equation is discretised using the THINC/QQ algebraic VOF scheme, yielding a consistent discretisation of the advective terms. This new algorithm is tested by considering representative surface-tension-dominated interfacial flows, including the Laplace equilibrium of a stationary droplet and the three-dimensional Rayleigh-Plateau instability of a liquid filament. The presented results demonstrate that interfacial flows with large density and viscosity ratios can be simulated and energy conservation is ensured, even with a time step larger than the capillary time-step constraint, provided that other time-step restrictions are satisfied.

Inheritance of signs of parent species by hybrids of Russet, Yellow and Speckled ground squirrels (<i>Spermophilus</i>, Sciuridae, Rodentia)

The inheritance of size, color and bioacoustic characteristics of Russet (Spermophilus major), Yellow (S. fulvus) and Speckled (S. suslicus) ground squirrels in hybrids differing in their genetic status were studied. In a hybrid population of Russet and Yellow ground squirrels, 10 individuals of S. major, 10 individuals of S. fulvus and 40 hybrids of S. major × S. fulvus were analyzed; in a hybrid population of Russet and Speckled ground squirrels, 11 individuals of S. major, 11 individuals of S. suslicus and 24 hybrids of S. major × S. suslicus were analyzed. Hybrid individuals different in genetic status have been shown to demonstrate differentiation in the vector space of the variability range of the parental species. At the same time, both a shift in some categories of hybrids towards one of the hybridizing species and a significant deviation are noted. The results obtained indicate an increase in the level of variability in size, color and bioacoustic characters in hybrid populations, associated with the erosion of the divergent hiatus of specific features of the parental forms due to the formation of a differentiated hybrid population. As hybrid individuals with different combinations of parental traits accumulate in the population, the situation arises when a sufficient number of hybrids with an overall combination of parental traits (Hcomb) appear in the hybridogenic population. In the hybrid population of the Russet and Yellow ground squirrels, the proportion of such hybrids, according to our data, is 22.5% (n = 40), and in the hybrid population of the Russet and Speckled ground squirrels it is 12.5% (n = 24). This category of hybrid individuals can be considered as possible material for further microevolutionary events and the process of speciation.

V. Introduction and selection of a new species of Clarkia (C. purpurea) and development of a method for evaluating Clarkia pursh varieties for distinctiveness, uniformity and stability

During 2011–2023 in the south of Western Siberia, with the aim of introducing and selecting a new species of clarkia (C. purpurea), its biological characteristics and morphological characteristics were studied. Based on the data obtained and the use of other species and varieties of the genus Clarkia Pursh. For the first time, a methodology for conducting tests for distinctiveness, uniformity and stability (DUS) for this crop was developed. This technique is applicable to all varieties of the genus Clarkia Pursh, classified into sections: Godetia (C. purpurea (Curtis) A. Nelson & J. F. Macbr.), Phaeostoma (C. unguiculata Lindl.), Rhodanthos (C. amoena (Lehm.) A . Nelson & J. F. Macbr., Clarkia amoena ssp. lindleyi (Douglas) H. F. Lewis & M. Е. Lewis), fireweed family (Onagraceae Juss.). Many species of clarkia are characterized by long-lasting flowering and are promising for floral decoration of urban spaces, as well as for cutting. Most varieties are from very low – up to 25 cm tall, to very high – 96 cm or more; leaves are simple, lanceolate, narrow-lanceolate, wide-lanceolate, ovate, smooth, pubescent. The flowers are very decorative, varying in size from 2.0 to 8.0 cm tall, simple, semidouble, terry and strongly terry with or without fragrance, with a large variability of colors and types of floral pigmentation. Based on the study of the bioecological and morphological properties of the crop under study and in accordance with the rules approved by the Federal State Budgetary Institution “State Varietal Commission”, 36 characteristics were selected by which it is possible to test clarkey varieties. As a result of breeding work, a new variety of C. purpurea, Lilac Fairy, was created. When analyzing crossing, it was established that the lilac (violet) color of flowers is dominant in relation to pale pink (almost white), and when crossing F1 hybrids with light purple color, splitting 15:1 (15 violet from dark purple to pale purple was obtained and 1 is pale pink), which confirms the hypothesis about the polygenic nature of the inheritance of the main color of flowers in C. purpurea. The national methodology RTG/1157/1, created for the first time, for testing the DUS of Clarkia Pursh varieties will serve as a scientific basis for practical breeding.

Color–emotion associations by speakers of Spanish and Mandarin in verbal and visual tasks: a comparison

Abstract This study aims to determine if there are differences in color–emotion association between monolingual speakers of Spanish and Mandarin, depending on how colors are presented (verbally or visually). We tested two groups of 25 speakers of these two languages in two different tasks using the Geneva Emotion Wheel, which encompasses 20 types of emotions. In Task 1, 13 colors were presented to participants as color terms in their native language, whereas in Task 2 the same colors were presented as color patches from the Munsell chart. Participants were then asked to associate color terms or color patches to the set of emotion concepts (and intensities of emotion) in the Geneva Emotion Wheel. Overall, differences between languages were not significant, regarding either the type of emotion or individual dimensions of emotion (valence, arousal or power), although significant differences were observed for specific colors. Also, Spanish speakers tended to attribute higher intensity values and higher numbers of emotion values to colors. At the same time, speakers of both languages reacted similarly to color presentation, with color terms being associated with the same emotions as color patches, but eliciting stronger reactions with respect to intensity and the number of emotion values. Finally, we found less variability in color–emotion associations within the Spanish-speaking group. Overall, our study points to a mixed pattern of universality and culture-specificity regarding how colors are used for conveying emotions by people.

Earth observation reveals the shifting patterns of China's lake colour driven by climate change and land cover

Water colour has been recognized as one of the most important Essential Climate Variables of the lake ecosystem, as it is directly related to changes in water constituents and almost all of the lake's ecological changes could alter water colour. Given the high retrieval accuracy from existing Earth observation satellite data, water colour, in terms of Forel Ule Index (FUI), can be a realistic indicator to track the long-term changes in the lake ecosystem and further explore the lake response to environmental changes. This paper aims to comprehensively investigate the spatiotemporal variation patterns of FUI in 159 large lakes (≥25 km2) across China during 2000–2022 based on the MODIS data and detect the climatic and anthropogenic driving forces of these changes. The 23 years of MODIS records revealed an overall downward trend of lake FUI across China, indicating the lakes in China shifted to bluer colour during the past two decades. Through driving factor analyses, the complicated interplay among lake colour, lake morphology, regional climate shifts and human interference dynamics was uncovered. In the long term, it was found the pronounced change in lake colour in the western lake zones was primarily attributed to climate warming and humidification, whereas that in the eastern lake zones was mainly related to the alterations in regional land cover during the past two decades. Seasonally, lake basin's air temperature was identified as the main factor impacting the seasonal patterns of lake colour, followed by wind speed and runoff. Spatially, there was high spatial variability in lake colour across China, which was mainly associated with lake elevation and lake basin's precipitation rate, although the factors exhibited considerable divergence across different zones. Based upon the above findings, the implications for lake environment protection and management in different regions of China were further discussed.

Perceptual effect of color use in patient rooms

Messages are conveyed to users through every element within the space. People receive these messages through sense organs. People and space interact constantly because of the way that they perceive and transmit space. People also establish their action boundaries, identify their motivational criteria, and feel a sense of belonging to the place they experience as a result of this interaction. In this way, the location, with all of its characteristics, influences the mental health of those who inhabit it. The colors chosen for patient rooms have a significant impact on the emotional and mental health of the patients, given the medical procedures and rehabilitation programs that they undergo. Patients' healing processes may vary based on the physical characteristics of the environment in which they are situated. Hospitals are intricate functional structures, so specific design requirements should be established by assessing every department independently. The purpose of this study is to determine how people's perception of spatial quality in general patient rooms where two adults stay during treatment is affected by the use of color. In the research method, adjective pairs were determined in line with spatial quality in order to reveal the perceptual effect of color variables in patient rooms. The survey technique was applied according to this experimental design. By simulating the patient room- which was proposed as the study area- in a digital environment with a one-to-one scale, the colors identified as the dependent variable in the study's methodology were visualized independently. The survey technique was employed to ask questions about the perceptual effect of colors on people who have experienced patient rooms by using adjective pairs based on the spatial quality level. The orange color used in patient room interior design was found to be eye-catching, communication facilitator, inviting, pleasing, encouraging, and sincere, while the blue color was found to be comforting and refreshing, based on data acquired from the survey, which was administered to 168 people in total. Many spaces in hospital buildings serve a variety of purposes, and the patients who use these spaces have a range of needs. For this very reason, it is advised that these hospital units be included in the interior design process utilizing information supported by empirical research, with each space being specially designed to meet its unique requirements and functions.

A web-based mpox skin lesion detection system using state-of-the-art deep learning models considering racial diversity

The recent ‘Mpox’ outbreak, formerly known as ‘Monkeypox’, has been a significant public health concern and spread to over 110 countries globally. The challenge of clinically diagnosing mpox early on is due, in part, to its similarity to other types of rashes. Computer-aided screening tools have been proven valuable in cases where Polymerase Chain Reaction (PCR) based diagnosis is not immediately available. Deep learning methods excel at learning complex data representations, but their efficacy largely depends on adequate training data. To address this challenge, we present the “Mpox Skin Lesion Dataset Version 2.0 (MSLD v2.0)” as a follow-up to the previously released open-access dataset, one of the first to contain mpox lesion images. This dataset contains images of patients with mpox and five other non-mpox classes (chickenpox, measles, hand-foot-mouth disease, cowpox, and healthy). We benchmark the performance of several state-of-the-art deep learning models, including VGG16, ResNet50, DenseNet121, MobileNetV2, EfficientNetB3, InceptionV3, and Xception, to classify mpox and other infectious skin diseases. We leverage transfer learning implemented with pre-trained weights generated from the HAM10000 dataset, an extensive collection of pigmented skin lesion images. To reduce the impact of racial bias, we employ a color space data augmentation method to increase skin color variability during training and present a rigorous assessment of the technique’s effectiveness in skin lesion classification; we achieved the best overall accuracy of 83.59±2.11%. Finally, the developed models are incorporated within a prototype web application to analyze uploaded skin images by a user and determine whether a subject is a suspected mpox patient. We believe that the presented tool can be of significant impact in screening of mpox in case of a future outbreak.

Water Quality Index by Multivariate Statistical Technique to Estimate the Temporal Decay of the Watercourses

Objective: This study aims to add some variables that characterize the water quality in a water quality index (WQIacp), observing the events of precipitation and land use watersheds. Method: The water samples were characterized in 180 local rivers, and they were monthly analyzed. To ellaborate the WQIacp, a multifactorial analysis statistic and the weighted least squares method were used. Results and Discussion: The first factor explained 42.68% of variance, and variables apparent color and total coliform variables showed higher scores in the formation of the WQIacp. According to the analysis, it´s observed that there is a strong correlation between areas that have higher rates of remaining with the municipalities that have a better WQIacp. Cluster analysis of the municipalities, considering the remaining native vegetation, shows that in the extreme north of Goiás are the cities with the largest reserves. In central and southern part of the state lie the most deforested municipalities. Finally, in the northwestern, part of central and southern of Goiás State, are those cities that present inequality related to the the amount of native vegetation. The worst WQIacp indexes for municipalities evaluated, -1.00 and -0.96, are in the center of Goiás, in Meia Ponte River and João Leite River, respectively, near the city of Goiânia, while the best is in northern of Goiás, 0.68, in Campos Belos in Santa Tereza River. It is still observed that for all the years analyzed, there is an improvement in water quality in the dry season. The average rates grow from April on and get lower from October on, in the beginning of the rainy season. This loss of water quality during flooding is represented mainly by varying color, total coliforms and turbidity, which have scores 0.200, 0.187 and 1.179, respectively. Research Implications: The creation of a water quality index helps in diagnosing and comparing the quality of water sources, facilitating decision-making by managers of Sanitation Concessionaires. Originality/Value: This study contributes to the literature, proposing a methodology based on descriptive statistics to create a water quality index, which can assist managers in decision-making.

Communicating the diagnosis of spinal muscular atrophy in endogamous vs. non-endogamous regions

IntroductionThe high prevalence of endogamy, or inbreeding, in northeastern Brazil, is due to historical and cultural factors, with large families living in cities far from the coast and subject to low socioeconomic and infrastructure levels. This breeding practice results in low genetic variability with an increased prevalence of rare autosomal recessive and neurodegenerative diseases, such as spinal muscular atrophy (SMA).ObjectiveUnderstanding the impact of communicating the diagnosis of SMA on the mental health of patients and their families and the differences between the Northeast (endogamous region) and the other regions of Brazil (non-endogamous ones).MethodsCross-sectional study obtained through a structured questionnaire about the moment of receiving the SMA diagnosis, containing the Impact of Event Scale-Revised.Results and discussionThe sample consisted of 100 volunteers from all regions of Brazil, 47 patients diagnosed with SMA and 53 family members present at the time of the diagnosis. There was a predominance of females (83%) and homogeneity between the groups for the variables gender, age, color, education, religion, and SMA subtype (1, 2, 3, and 4). The Northeast region, representing 43% of the sample, despite being less economically favored, showed greater satisfaction with medical care and inclusion in health services, with less self-reported psychological trauma and fewer signs of post-traumatic stress disorder (PTSD) related to the moment of receiving the diagnosis. The non-endogamous regions, in turn, reported the presence of strong waves of emotion, sleep problems, feelings of irritability, anger, and the presence of bad thoughts related to this situation.ConclusionThe feeling of inclusion in health services and satisfaction with medical care in the endogamous region had a positive impact on the mental health of those involved, reducing psychological trauma and signs of PTSD arising from the communication of the SMA diagnosis.

Anthocyanin‐Enhanced Bacterial Cellulose Nanofibers for Sustainable Hg(II) Ion Sensing

AbstractHg(II), poses a significant threat to both the environment and overall well‐being. The present indicator utilizes anthocyanin extract from roselle (Hibiscus sabdariffa), RSE, and incorporates it into bacterial cellulose nanofibers (BCNF) to detect Hg(II) ions. Through examination of the CIE Lab color variables, it is determined that the binding of Hg(II) leads to a discernible alteration in the color. The visible color changes as the Hg(II) concentration increases. ATR‐FTIR(Attenuated Total Reflectance‐Fourier transform infrared spectroscopy) and SEM(scanning electron microscopy) are utilized to analyze the bacterial cellulose nanofibers containing anthocyanin. The selectivity of the proposed sensor is assessed by conducting experiments with different types of metal ions including Co2+, Al3+, Na+, Ba2+, Ni2+, Mg2+, Cd2+, Pb2+, Fe2+, Fe3+, and Cu2+. The outcomes indicate that the sensor exhibits a remarkable ability to distinguish and respond specifically to Hg(II) ions. This particular sensing technique allows for the precise and reliable detection of Hg(II), with a detection limit of 0.72 ppm, by simply observing with the naked eye. For the effective treatment of a tap water sample, the BCNF‐RSE film provides successful results in the detection of Hg(II) ions. This innovative metallochromic sensor offers a promising solution for the on‐site, cost‐effective, portable, easy to use, and real‐time detection of Hg(II) ions.

Atmospheric Waves Driving Variability and Cloud Modulation on a Planetary-mass Object

Planetary-mass objects and brown dwarfs at the transition (T eff ∼ 1300 K) from relatively red L dwarfs to bluer mid-T dwarfs show enhanced spectrophotometric variability. Multiepoch observations support atmospheric planetary-scale (Kelvin or Rossby) waves as the primary source of this variability; however, large spots associated with the precipitation of silicate and metal clouds have also been theorized and suggested by Doppler imaging. We applied both wave and spotted models to fit near-infrared (NIR), multiband (Y/J/H/K) photometry of SIMP J013656.5+093347 (hereafter SIMP0136) collected at the Canada–France–Hawaii Telescope using the Wide-field InfraRed Camera. SIMP0136 is a planetary-mass object (12.7 ± 1.0 M J) at the L/T transition (T2 ± 0.5) known to exhibit light-curve evolution over multiple rotational periods. We measure the maximum peak-to-peak variability of 6.17% ± 0.46%, 6.45% ± 0.33%, 6.51% ± 0.42%, and 4.33% ± 0.38% in the Y, J, H, and K bands, respectively, and find evidence that wave models are preferred for all four NIR bands. Furthermore, we determine that the spot size necessary to reproduce the observed variations is larger than the Rossby deformation radius and Rhines scale, which is unphysical. Through the correlation between light curves produced by the waves and associated color variability, we find evidence of planetary-scale, wave-induced cloud modulation and breakup, similar to Jupiter’s atmosphere and supported by general circulation models. We also detect a 93.°8 ± 7.°4 (12.7σ) phase shift between the H − K and J − H color time series, providing evidence for complex vertical cloud structure in SIMP0136's atmosphere.

The effects of continuously wetting-drying cycles on pre-weathered and oil-coated rowan (Sorbus torminalis) wood

The naturally weathered and then oil-coated (Teak oil) Rowan wood (Sorbus torminalis) was subjected to distilled (A) and swimming pool water (B) aging treatments with continuous wetting-drying up to the third cycle. Treatment conditions and water aging types seem to influence color variables somewhat. The highest lightness (ΔL*), redness (Δa*), and yellowness (Δb*) changes were measured in samples that were continuously aged at the third cycle procedure in swimming pool water, respectively. Similar results have also been observed for total color changes, such that pool water had higher discoloration effects on color values compared to distilled water at similar conditions, while increasing the cycle usually further lowered discoloration. The highest color change value of ΔEB3: 10.49 (metric) was found in the third cycle in pool water treatment, followed by a value of ΔEB2: 9.42 (metric) in the second cycle and a value of ΔEB1: 1.71 (metric) in the first cycle, respectively. Color changes could be employed as a sensitive physical indicator of coating aging, which can be correlated to the associated chemical changes. All aging processes have lower gloss values than control in all aging procedures, regardless of water type. The lowest gloss value of 1.2 GU was noticed at the third cycle in swimming pool treatments. It is also notable that the most dramatic effect on surface hardness (Shore D) was found after the first cycle, while the lowest hardness values of 38 (metric) and 33 (metric) were found in swimming pools and distilled pure water-treated samples, respectively.

Efficacy of 785-nm Picosecond Titanium Sapphire Laser for Treatment of Brown Nevus of Ota Lesions in FST II-V: A Retrospective Analysis.

Nevus of Ota (NOTA) is a dermal melanocytosis acquired in early childhood or pregnancy. Given their color variability, NOTA often require a combination of wavelengths for successful treatment. Quality-switched lasers have consistently shown efficacy in targeting dermal pigment, while picosecond lasers (PSLs) are an emerging technology for pigmentary disorders. To further elucidate its efficacy, the authors conducted a retrospective review of 17 patients with NOTA treated with a 785-nm PSL for brown NOTA lesions between 2021 and 2023. The primary end point analyzed clinical improvement based on before and after photography reviewed by 3 board-certified dermatologists using a five-point visual analog scale. Seventeen patients of Fitzpatrick skin types (FSTs) II to V, ranging from ages 14 to 38 years, were included in this study. Patients were treated for an average of 3.2 sessions in 2 to 3-month intervals. Visual analog scale scores demonstrated a mean clearance of 51% to 75%. No pigmentary alterations were noted. Because NOTA is common in higher FSTs, the authors believe that the 785-nm PSL is an excellent treatment option for brown NOTA in these skin types. This study highlights the need for further investigation to determine optimal treatment parameters for the color-based laser treatment approach for NOTA.

Comparative transcriptomic and metabolomic profiles reveal fruit peel color variation in two red pomegranate cultivars.

Pomegranate (Punica granatum L.) which belongs to family Lythraceae, is one of the most important fruit crops of many tropical and subtropical regions. A high variability in fruit color is observed among different pomegranate accessions, which arises from the qualitative and quantitative differences in anthocyanins. However, the mechanism of fruit color variation is still not fully elucidated. In the present study, we investigated the red color mutation between a red-skinned pomegranate 'Hongbaoshi' and a purple-red-skinned cultivar 'Moshiliu', by using transcriptomic and metabolomic approaches. A total of 51 anthocyanins were identified from fruit peels, among which 3-glucoside and 3,5-diglucoside of cyanidin (Cy), delphinidin (Dp), and pelargonidin (Pg) were dominant. High proportion of Pg in early stages of 'Hongbaoshi' but high Dp in late stages of 'Moshiliu' were characterized. The unique high levels of Cy and Dp anthocyanins accumulating from early developmental stages accounted for the purple-red phenotype of 'Moshiliu'. Transcriptomic analysis revealed an early down-regulated and late up-regulated of anthocyanin-related structure genes in 'Moshiliu' compared with 'Hongbaoshi'. Alao, ANR was specially expressed in 'Hongbaoshi', with extremely low expression levels in 'Moshiliu'. For transcription factors R2R3-MYB, the profiles demonstrated a much higher transcription levels of three subgroup (SG) 5 MYBs and a sharp decrease in expression of SG6 MYB LOC116202527 in high-anthocyanin 'Moshiliu'. SG4 MYBs exhibited two entirely different patterns, LOC116203744 and LOC116212505 were down-regulated whereas LOC116205515 and LOC116212778 were up-regulated in 'Moshiliu' pomegranate. The results indicate that specific SG members of the MYB family might promote the peel coloration in different manners and play important roles in color mutation in pomegranate.

An Optimized Semantic Segmentation Framework for Human Skin Detection

The study incorporating optimization strategy in semantic segmentation is underexplored in dermatology. Existing approaches used complex and various heuristic designs of image processing algorithms and deep models customized for skin detection problems. This paper demonstrates Particle Swarm Optimization (PSO)-incorporated AlexNet framework for the skin segmentation task. The results from testing the trained model are promising. The model produced satisfactory performances even with a strict split of 50 %, confirming the high efficiency of the proposed framework. The mean Jaccard index and Dice similarity measures evaluated between the annotated and predicted mask ranged from 0.80 to 0.93 in the binary classification of pixels as “skin” versus “background”. This work identified that the location and color variability of skin pixels in the training data are crucial to obtaining a good skin segmentation performance. Further works that can be explored in this area include adopting a robust preprocessing strategy to increase data variability and improve model generalization or implementing an optimization-enhanced strategy on the existing segmentation models for comparison.

Pengendalian Kualitas Produksi Keripik Pisang pada UMKM Pekopen Menggunakan Peta Kendali NP dan Kapabilitas Proses

Pekopen MSMEs produce banana chips that always strive to produce good products, but in fact they still find some products produced are not in accordance with standards. These problems can be solved using nP control maps, Ishikawa diagrams, and process capabilities, as well as calculations using minitab software. By using some of these methods, it is expected to find the factors that cause product defects. There are several variables to assess the defects of banana chip products, including shape, color, and size variables. The results of this study show the factors that cause defects in banana chip products, namely method factors caused by the manual banana chip cutting process, inaccurate selection of raw materials, manual frying process, and manual packaging process. In addition, it was found that the level of specifications and standards for process capabilities was not less than 95% and no more than 3% of the inability of MSMEs to produce banana chips in accordance with specifications and standards.

Image Data-Centric Visual Feature Selection on Roll-to-Roll Slot-Die Coating Systems for Edge Wave Coating Defect Detection.

Roll-to-roll (R2R) manufacturing depends on a system's capability to deposit high-quality coatings with precise thickness, width, and uniformity. Therefore, consistent maintenance requires the immediate and accurate detection of coating defects. This study proposes a primary color selection (PCS) method to detect edge defects in R2R systems. This method addresses challenges associated with training data demands, complexity, and defect adaptability through a vision data-centric approach, ensuring precise edge coating defect detection. Using color information, high accuracy was achieved while minimizing data capacity requirements and processing time. Precise edge detection was facilitated by accurately distinguishing coated and noncoated regions by selecting the primary color channel based on color variability. The PCS method achieved superior accuracy (95.8%), outperforming the traditional weighted sum method (78.3%). This method is suitable for real-time detection in manufacturing systems and mitigates edge coating defects, thus facilitating quality control and production optimization.

Explainable Automatic Detection of Fiber–Cement Roofs in Aerial RGB Images

Following European directives, asbestos–cement corrugated roofing tiles must be eliminated by 2025. Therefore, identifying asbestos–cement rooftops is the first necessary step to proceed with their removal. Unfortunately, asbestos detection is a challenging task. Current procedures for identifying asbestos require human exploration, which is costly and slow. This has motivated the interest of governments and companies in developing automatic tools that can help to detect and classify these types of materials that are dangerous to the population. This paper explores multiple computer vision techniques based on Deep Learning to advance the automatic detection of asbestos in aerial images. On the one hand, we trained and tested two classification architectures, obtaining high accuracy levels. On the other, we implemented an explainable AI method to discern what information in an RGB image is relevant for a successful classification, ensuring that our classifiers’ learning process is guided by the right variables—color, surface patterns, texture, etc.—observable on asbestos rooftops.