Online Images Research Articles

A data-driven approach to traditional village colors: K-Means clustering of online images

ABSTRACT This study aims to quantify the color characteristics of traditional villages, explore the key aspects of color that attract tourist attention, and propose solutions to enhance the readability and legibility of these villages. The findings are intended to provide practical guidance for the restoration and use of village colors. The research focuses on how tourists perceive the color characteristics of traditional villages. The objective is to conduct color clustering analysis to identify the composition and characteristics of village colors, and further analyze the underlying causes to offer suggestions for optimization and enhancement. It employs a methodology based on urban imagery theory, using clustering and secondary extraction of color data from online images. The results reveal distinct contrasts and harmonies in the colors of the five elements providing a basis for improving spatial readability and visual guidance. These findings not only offer practical design references for optimizing traditional village colors but also provide valuable insights for the protection and restoration of other traditional sites. In particular, the contrast between the colors of boundaries and landmarks can enhance visual recognition, while the complementary color designs of paths and boundaries can improve spatial readability, contributing to the sustainable development of traditional villages.

Differences between Doctors of Dental Medicine and Doctors of Medicine Awareness of Their Online Image and Perception Concerns: a Quantitative Cross-Sectional Study.

Healthcare professionals (HCPs), including doctors of dental medicine (DMDs) and doctors of medicine (MDs), use social networking sites (SNSs) for personal and professional purposes. This study aims to: (i) explore DMDs and MDs awareness of their online presence, (ii) examine how frequently and for what reasons DMDs and MDs monitor their online images, (iii) determine the prevalence of finding inaccurate, incomplete, or unprofessional information about themselves, examine that experience, and investigate actions performed regarding unprofessional or embarrassing posted material, (iv) assess concerns about their own SNSs usage and the content they post. A cross-sectional study was conducted among DMDs and MDs in Croatia using a questionnaire. Out of 1,013 participants, 753 were active SNS users. Facebook (91.6%) and Instagram (63.1%) were the predominant SNSs used. DMDs exhibited a significantly higher inclination to utilize SNSs for professional purposes than MDs. Both groups reported a high level of proficiency in using Facebook (M=3.2, SD=0.827) and Instagram (M=2.44, SD=1.162). Approximately 55.4% of the participants repored monitoring their online presence, typically once a year or less frequently. DMDs emphasized the completeness of posted information more than MDs (U=25,623.0, p=0.001). Their highest level of concern pertained to confidentiality breaches and inaccurate medical and dental information. There are similar results for MDs and DMDs for finding erroneous information posted on SNSs. This study offers insights into HCPs' online presence, and their concerns regarding SNS usage. It underscores the significance of e-professionalism, and provides recommendations for maintaining a positive online presence while mitigating potential risks.

IMAGE CAPTION GENERATOR USING CNN AND LSTM

Machine learning is now all the rage in the AI world. We have recently used AI to construct very clever devices with exceptional performance. Deep learning is a subset of machine learning that produces very accurate findings, which in turn indicates very good performance. Apps for picture description make use of deep learning in our study. Providing a description of a picture's content is what image description is all about. Object and action detection in the input picture is the foundation of the notion. When describing images, there are primarily two methods: bottom-up and top-down. Bottom-up methods create captions by combining the information of an input picture. Using different architectures, such as recurrent neural networks, top-down methods provide a semantic representation of an input picture, which is then translated into a caption. One potential advantage of picture description is that it might aid those with visual impairments in comprehending what is shown in online images. What follows is an explanation of the specifics. Looking at the image below, what can you make out?

AI models collapse when trained on recursively generated data

Stable diffusion revolutionized image creation from descriptive text. GPT-2 (ref. 1), GPT-3(.5) (ref. 2) and GPT-4 (ref. 3) demonstrated high performance across a variety of language tasks. ChatGPT introduced such language models to the public. It is now clear that generative artificial intelligence (AI) such as large language models (LLMs) is here to stay and will substantially change the ecosystem of online text and images. Here we consider what may happen to GPT-{n} once LLMs contribute much of the text found online. We find that indiscriminate use of model-generated content in training causes irreversible defects in the resulting models, in which tails of the original content distribution disappear. We refer to this effect as ‘model collapse’ and show that it can occur in LLMs as well as in variational autoencoders (VAEs) and Gaussian mixture models (GMMs). We build theoretical intuition behind the phenomenon and portray its ubiquity among all learned generative models. We demonstrate that it must be taken seriously if we are to sustain the benefits of training from large-scale data scraped from the web. Indeed, the value of data collected about genuine human interactions with systems will be increasingly valuable in the presence of LLM-generated content in data crawled from the Internet.

Pilot study exploring artificial intelligence for facial-image-based diagnosis of Marfan syndrome

BackgroundMarfan Syndrome (MFS), a genetic disorder impacting connective tissue, manifests in a wide array of phenotypes which can affect numerous bodily systems, especially the thoracic aorta. The syndrome often presents distinct facial features that potentially allow for diagnostic clinical recognition. Herein, we explore the potential of Artificial Intelligence (AI) in diagnosing Marfan syndrome from ordinary facial images, as assessed by overall accuracy, F1 score, and area under the ROC curve. MethodsThis study explores the utilization of Convolutional Neural Networks (CNN) for MFS identification through facial images, offering a novel, non-invasive, automated, and computerized diagnostic approach. The research examines the accuracy of Neural Networks in the diagnosis of Marfan Disease from ordinary on-line facial images. The model was trained on 80 % of 672 facial images (182 Marfan and 490 control). The other 20 % of images were used as the test set. ResultsOverall accuracy was 98.5 % (0 % false positive, 2 % false negative). F1 score was 97 % for Marfan facies and 99 % for non-Marfan facies. Area under the ROC curve was 100 %. ConclusionAn Artificial Intelligence (AI) program was able to distinguish Marfan from non-Marfan facial images (from ordinary on-line photographs) with an extremely high degree of accuracy. Clinical usefulness of this program is anticipated. However, due to the limited and preliminary nature of this work, this should be viewed as only a pilot study.

Social Media Users’ Visual and Emotional Preferences of Internet-Famous Sites in Urban Riverfront Public Spaces: A Case Study in Changsha, China

With the increasing online exposure of urban public spaces, the new concept of “internet-famous sites” has emerged in China. Social media users are the main contributors to this new phenomenon. To fully understand social media users’ preferences in such kinds of public spaces, this article took 27 typical riverfront internet-famous sites (RIFSs) in Changsha City (China) as an example. Through social media platform selection, keyword research, text and image data extraction, visual and emotional symbol coding, and manual calculations of coding frequency, this study investigated social media users’ perception of RIFSs, especially on visual and emotional preferences. The online images and review comments were extracted from the popular Chinese social media platform “Xiaohongshu”. We found that (1) the popularity of each RIFS had a significant head effect and there were far more positive emotions than neutral and negative emotions in review comments. (2) RIFSs in Changsha were divided into five categories: commercial RIFSs, art exhibition RIFSs, historical and cultural RIFSs, ecological recreational RIFSs, and uncultivated RIFSs. Social media users had different visual focuses on each kind of RIFS. (3) Social media users provided specific reasons for their emotional preferences towards different types of RIFSs. This study can provide a new perspective on improving waterfront vitality and offer a targeted and attractive method for waterfront regeneration that is different from traditional methods.

Updated typification of the Tasmanian pteridophytes described by Labillardière in his Novae Hollandiae plantarum specimen of 1806

AbstractWe present updated type information on the Tasmanian species of ferns and lycophytes described by Labillardière in his Novae Hollandiae plantarum specimen of 1806. Among the 13 names introduced, we list a total of 80 specimens with type status (13 lectotypes and 67 isolectotypes), mainly from FI, which houses Labillardière's herbarium (and the main set of plants) in FI‐W, but also from B (including B‐W), BM, F, G, GH, K, L, LE, MEL, P (including P‐JU), PRC, RB, and WAG. Another nine specimens are cited as probable isolectotypes, since they lack confident label data. Four lectotypifications and one second‐step lectotypification are newly proposed. Some specimens previously regarded as “holotypes” are better referred to as inadvertently chosen lectotypes. We show evidence that Lindsaea lanceolata Labill. (= L. ensifolia Sw. subsp. ensifolia), although originally described from Tasmania, was probably mislabeled, and the specimen was actually from Amboyna (= Ambon), Indonesia. We provide an Appendix with links to online images of nearly all type specimens.

Applying Digital Technology Opportunities to Education System of Azerbaijan

One of the areas in which the development of information and communication technologies (ICT) has made the greatest contribution is education. The introduction of new technology in education increases the quality of the educational process and creates conditions for the adoption of ICT by the younger generation. Modern technology opportunities have changed education process. It has opened new ways to a massive amount of information through online books, podcasts, images, and videos. Some years ago technology was only used for completing assignments. But nowadays it is used for various aims, from learning to communicating and collaborating. Technology has also opened up numerous opportunities for communication and collaboration. For instance, students in a classroom in rural part of the world can learn from an article published by a other developed countries. Students can communicate through email and even have a lecture, or class virtually via video conference. Furthermore, technology has changed the role of teachers and students. Traditionally, teachers were the centre figure of the learning process, and it was called teacher-centered model. However, with the digital technology sources, information is accessible to everyone, and it gives us more chances for student-centered learning model. In this case the role of teachers is as a guide, and students try to search, to analyze the materials themselves. Of course, the State Program implemented by the Ministry of Education regarding the electronicization of education has created a serious change in this field. During the implementation of the program, modern technological tools were installed in many schools from the capital to the regions. There is no such school that does not have a computer room for the use of students, and does not have an Internet network installed. In most of the capital's schools, the high stage of the electronicization of the educational process has begun. The main purpose of the study is to search literature materials and policy papers about this topic and making description. Keywords: modern education system, digital technology, ITE-Innovative technologies in Education.

Redefining Ormyridae (Hymenoptera, Chalcidoidea) with establishment of subfamilies and description of new genera

AbstractThe circumscription of the family Ormyridae (Hymenoptera: Chalcidoidea) is revised after phylogenetic analysis based on ultra‐conserved elements (UCEs) and comparative morphological assessment of the chalcid ‘Gall Clade’. Six genera are treated in the family, including two new genera, Halleriaphagus van Noort and Burks, gen. nov., and Ouma Mitroiu, gen. nov. One genus, Eubeckerella Narendran, is re‐assigned to the family, and Ormyrulus Bouček is synonymised with Ormyrus Westwood, syn. nov., resulting in the new combination Ormyrus gibbus (Bouček), comb. nov. The six genera are classified in three subfamilies, two of which are newly described, Asparagobiinae van Noort, Burks, Mitroiu and Rasplus, subfam. nov., and Hemadinae van Noort, Burks, Mitroiu and Rasplus, subfam. nov. Halleriaphagus is established for the newly described type species Halleriaphagus phagolucida van Noort and Burks, sp. nov., and Ouma is erected for O. daleskeyae Mitroiu, sp. nov., and O. emazantsi Mitroiu, sp. nov. Asparagobius is revised with description of Asparagobius bouceki van Noort, sp. nov., and Asparagobius copelandi Rasplus and van Noort, sp. nov. Asparagobius and Halleriaphagus are classified in Asparagobiinae, Hemadas in Hemadinae and Eubeckerella, Ormyrus and Ouma in Ormyrinae. The molecular support defining the ormyrid clade is corroborated by the proposed morphological synapomorphy of a foliaceous prepectus overlying the tegula base. Identification keys to the genera of Ormyridae and to the species of Asparagobius and Ouma are provided. Online Lucid identification keys and images of all the species treated herein are available at: http://www.waspweb.org.Zoobank Registration: LSID urn:lsid:zoobank.org:pub:8811695B-EE57-4C18-A6B6-E63D267E2373.

The projection of Chinese University online image and social media engagement based on Bayesian model.

Social media platforms provide the public with a forum for interaction and communication with tourism destinations, playing a significant role in the shaping and dissemination of destination images. Similarly, social media plays a vital role in the construction and propagation of online images for higher education institutions. For instance, indicators such as likes, shares, and visits on Weibo can serve as measures of public engagement with university social media. To reveal the triggering rules of social media engagement by projected images of destinations and related factors, this paper builds a Bayesian model using data from posts and interactions on the official Sina Weibo account of a Chinese university from 2018 to 2023. This model simulates to infer the optimal decisions that trigger university social media engagement.

Defining the term watershed to reflect modern uses and functions as inter- and intra-connected socio-hydrologic systems

Despite the many programs focused on watershed education, the watershed concept is poorly understood, which can lead to a number of wide-ranging consequences from poor watershed planning and policy to inaccurate scientific studies. We argue that the definition of the term watershed, and the images that accompany the definition, contribute to these misunderstandings. The definition has remained very much the same since it was first documented in Europe in the mid-18th century. However, watershed uses and functions have evolved considerably in the past three centuries, notably in the many ways water moves within and across traditionally-defined watershed boundaries. In this study, content analyses of the definitions of the term watershed from textbooks and online sources, and the accompanying images, highlight that watersheds are nearly universally defined as an area of land that drains to a point on a river. More than half the images that accompany these definitions depict forested landscapes or high mountain relief. While over one half of online images depict human activities or developed landscapes, few textbook images do. These incomplete definitions and idealized images do not reflect the diversity of landscapes and watershed functions. The narrow description can contribute to an individual’s disconnection to their own watershed. Furthermore, the description of watersheds as lumped areas, rather than inter- and intra-connected socio-hydrologic systems, misses a fundamental element of the watershed concept: that activities at every point in a watershed impact all points downstream and are impacted by activities at all points upstream. Thus, we propose a revised definition of a watershed that better captures this connectivity: a watershed defines the boundaries of a system of hydrologically connected people, places, and things.

Perceptions on Cultural Ecosystem Services of World Cultural Heritages in Beijing: A cross-cultural Comparison of Social Media-based Online Reviews and Images

This study introduces an innovative approach combining social media text and image comments to quantify the perceptual differences of cultural ecosystem services among various sociocultural groups. For five World Heritage Sites in Beijing, natural-language processing classifies visitor perceptions using social media comments and image data, and importance-performance analysis quantifies the differences between Chinese and international visitor perceptions of these cultural ecosystem services. For the six perceived cultural ecosystem service categories, cross-cultural and local visitors exhibit the same satisfaction trends. In particular, international tourists assign higher importance scores for a given cultural ecosystem service than Chinese tourists. The proposed methods incur a lower cost than traditional methods and acquire large volumes of broad-ranging data. Additionally, the importance of‒the performance analysis data-acquisition method is improved. Moreover, World Heritage Site sustainability can be promoted by understanding the different heritage tourism perceptions of different cultural groups. Furthermore, this study provides valuable insights for decision-makers to improve tourism management and conservation strategies.

A Graph-Based Framework for Traffic Forecasting and Congestion Detection Using Online Images From Multiple Cameras

A Graph-Based Framework for Traffic Forecasting and Congestion Detection Using Online Images From Multiple Cameras

Online Depiction of Urticaria Is Often Flawed and Does Not Reflect the Spectrum of Clinical Manifestation

Introduction: The internet is a popular source of health information including images of disease manifestations. Online photographs of skin lesions may aid patients in identifying their disease, if these pictures are of good quality and of the disease they claim to show. If not, patients may be at risk of delayed diagnosis, misdiagnosis, and suboptimal treatment. For urticaria, the mismatch rate and quality of online pictures are unknown. The objective of this study was therefore to evaluate the content and quality of online images of urticaria. Methods: The search term “urticaria” was applied to Google Images and Shutterstock. The top 100 photographs from each search engine were retrieved on October 9th, 2022. Illustrations, drawings, and heavily edited photographs were excluded. Each image was evaluated for patient characteristics, characteristics of urticarial lesions, and image quality. Results: Across 194 unique images of urticaria (after removing duplicates), 35 (18.0%) did not depict urticarial lesions, and 38 (19.6%) were ambiguous. Less than two-thirds of images 121 (62.4%) showed bona fide urticarial lesions. Pictures of urticarial lesions under-represented children and did not reflect female preponderance of the disease. Images predominantly depicted urticaria lesions on Caucasian skin (59.8%) and were typical of spontaneous rather than inducible urticaria. Only 3 (1.5%) pictures showed angioedema, a common clinical sign in patients with urticaria. The overall quality of online urticaria pictures was mostly good or very good. Conclusion: Physicians and patients should be aware that one in five online pictures of urticaria does not show urticarial skin lesions, and children, females, non-Caucasian patients, inducible urticaria, and angioedema are under-represented. These findings should prompt efforts to improve the accuracy and representativeness of online urticaria pictures.

Deep learning-based image analysis for in situ microscopic imaging of cell culture process

Mammalian cell culture is an important bioprocess that directly affects the quality and yield of biopharmaceuticals. Traditionally, condition monitoring of the operation is based on sampling periodically and off-line analysis, which is labor intensive, time consuming, and causing time delays. In this work, in situ microscope is investigated for on-line real-time monitoring of the culture process of Chinese hamster ovary cells with focus on investigation of deep learning-based Mask R-CNN algorithm for image analysis. The model is trained by 184 images with 183,040 cells using data augmentation methods and transfer learning technique. Mask R-CNN segmented the clustered cells more effectively than the conventional one combining edge detection, intensity thresholding, and advanced watershed method as well as the multi-scale edge detection method. Its Dice score, accuracy, precision, sensitivity, F1 score, specificity, and relative volume difference reach 0.862, 0.945, 0.901, 0.827, 0.862, 0.977, and 0.082, respectively. The evolution of geometrical features of cells were further analyzed, including equivalent diameter, circularity, aspect ratio, and eccentricity. The result demonstrated the great potential of deep learning technology in analysis of on-line images for optimization and control of the cell culture process.

Information-seeking Strategies Used and Challenges Encountered by Art and Design Students in Uganda When Seeking Information

Access to relevant information is crucial for Art and Design students to meet their information needs and achieve their educational goals. This paper explores the information-seeking strategies used by Art and Design students in diplomaawarding institutions in Uganda, and the challenges they encounter when seeking information to meet their information needs. This is part of a doctoral study that investigated the information-seeking behaviour of Art and Design students in diploma-awarding institutions in Uganda. The study population comprised Art and Design students in the four Art and Design institutions, library heads in the same institutions, and directors at the National Council for Higher Education (NCHE). Data was collected using questionnaire, interview and observation methods. Data was collected from February to August 2022. The study revealed that Art and Design students use various information-seeking strategies to locate and access information. Browsing online images, photographing activities in the environment, recording videos of activities in the environment, and consulting Art and Design librarians were the most used. The least used strategies included browsing art and design collections on library shelves, using keywords, and using Boolean operators. The study found that Art and Design students encountered varied challenges when seeking information. The key challenges were lack of Internet at the institutions, inadequate and outdated information materials in the library, slow Internet, high cost of Internet data, and lack or inadequate knowledge to evaluate the quality of information on the Internet. The key recommendations included the need for Art and Design institutions to invest heavily in ICT infrastructure, including Internet connectivity, and also to develop information literacy courses for the students; the need for the NCHE prioritise monitoring of the study institutions to ensure compliance with library standards; and the need for the Government of Uganda zero-rate the Internet in the study institutions.

Protection of online images against theft using robust multimodal biometric watermarking and T-norms

Protection of online images against theft using robust multimodal biometric watermarking and T-norms

Online versus offline tourism emotion: uncovering the bias pattern in online tourism emotion within theme parks

ABSTRACT Online texts and images have become an important data source for investigating tourism emotion. However, tourism emotion extracted from online data cannot reflect the emotions in the real world. Few scholarly attempts have focused on the complex biases in tourism emotion based on online data. By comparing online and offline emotion, this study quantified biases in tourism emotion and explored patterns among biases. Facial expressions and texts within Disney Resort were collected. Based on facial expression recognition and text mining, several indices were proposed to measure tourism emotion and biases. Our results reveal that tourists tend to share stronger happiness, increased anger in commerce areas and exaggerated disgust in waiting areas. In addition, males show “surprise-suppress”; while females show “surprise-exaggeration” in recreation areas. Our research can help scholars reexamine previous conclusions based on online tourism emotion and provides a theoretical basis for improving the quality of online tourism emotion.

Palaearctic distributions of Anthrenus pimpinellae (Fabricius) and Anthrenus isabellinus Küster (Coleoptera: Dermestidae)

Distributions of Anthrenus pimpinellae (Fabricius, 1775) and A. isabellinus Küster, 1848, are presented using data from verifiable online images, posted by citizen scientists. Anthrenus pimpinellae is distributed from north-western Europe into eastern and south-eastern Europe, and just into eastern Asia. Anthrenus isabellinus in Europe (and north Africa) is largely distributed across regions around the Mediterranean Sea. The species distributions overlap in south-eastern Europe, but do not overlap in western Europe and northern Africa. Until recently, the taxonomy of these two species was confused – consequently our knowledge of their distributions poor. Anthrenus pimpinellae is still considered by some authorities to be cosmopolitan; the current study shows that this is probably not the case.

Deep Gated Multi-modal Fusion for Image Privacy Prediction

With the rapid development of technologies in mobile devices, people can post their daily lives on social networking sites such as Facebook, Flickr, and Instagram. This leads to new privacy concerns due to people’s lack of understanding that private information can be leaked and used to their detriment. Image privacy prediction models are developed to predict whether images contain sensitive information (private images) or are safe to be shared online (public images). Despite significant progress on this task, there are still some crucial problems that remain to be solved. Firstly, images’ content and tags are found to be useful modalities to automatically predict images’ privacy. To date, most image privacy prediction models use single modalities (image-only or tag-only), which limits their performance. Secondly, we observe that current image privacy prediction models are surprisingly vulnerable to even small perturbations in the input data. Attackers can add small perturbations to input data and easily damage a well-trained image privacy prediction model. To address these challenges, in this article, we propose a new decision-level Gated multi-modal fusion (GMMF) approach that fuses object, scene, and image tags modalities to predict privacy for online images. In particular, the proposed approach identifies fusion weights of class probability distributions generated by single-modal classifiers according to their reliability of the privacy prediction for each target image in a sample-by-sample manner and performs a weighted decision-level fusion, so that modalities with high reliability are assigned with higher fusion weights while ones with low reliability are restrained with lower fusion weights. The results of our experiments show that the gated multi-modal fusion network effectively fuses single modalities and outperforms state-of-the-art models for image privacy prediction. Moreover, we perform adversarial training on our proposed GMMF model using multiple types of noise on input data (i.e., images and/or tags). When some modalities are failed by input data with noise attacks, our approach effectively utilizes clean modalities and minimizes negative influences brought by degraded ones using fusion weights, achieving significantly stronger robustness over traditional fusion methods for image privacy prediction. The robustness of our GMMF model against data noise can even be generalized to more severe noise levels. To the best of our knowledge, we are the first to investigate the robustness of image privacy prediction models against noise attacks. Moreover, as the performance of decision-level multi-modal fusion depends highly on the quality of single-modal networks, we investigate self-distillation on single-modal privacy classifiers and observe that transferring knowledge from a trained teacher model to a student model is beneficial in our proposed approach.