Methods and advanced tools for the analysis of film colors in digital humanities

A Digital Humanities Approach to Film Colors Barbara Flueckiger (bio) And much of the data mining that leads to visualization … is based on a flawed method that conflates literal discourse and symbolic/interpreted reference. In an art-historical context, this would be the equivalent of counting instances of the color red across a collection of images without discriminating between symbolic and representational functions. The reds are not the same, and cannot be counted the same way, put into the same category, or re-represented as data for visualization in a graph or chart, without monstrous distortion. JOANNA DRUCKER, "GRAPHICAL APPROACHES TO THE DIGITAL HUMANITIES," A NEW COMPANION TO DIGITAL HUMANITIES [End Page 71] There are strong and justifiable objections to the measurement or, more broadly, the computer-assisted analysis of aesthetic phenomena—as Johanna Drucker makes clear.1 One pitfall of quantitative analysis is its potential to disregard the meaningful context of data occurrences across the body of works studied. In fact, any quantitative approach to aesthetics that aims to reduce the inherently ambiguous quality of works of art into measurable units runs the risk of engaging in positivist reductionism and of fundamentally ignoring philosophical aesthetics and its analytical tradition. Given these reservations, it may seem extremely bold to investigate one of the most intangible aspects of film aesthetics, namely, film colors, by computer-assisted tools in the emerging field of digital humanities. Colors are elusive. Our perception of them is deeply influenced by the context of their appearance, their material presentation, the given cultural framework, and each individual spectator's subjective response. However, it was precisely this challenge that led to the development of the projects elaborated and reflected in the digital humanities platform Timeline of Historical Film Colors and, most importantly, the research project FilmColors, which was funded by an Advanced Grant from the European Research Council (ERC). By their very definition, ERC Advanced Grants are meant to be "high risk, high gain" projects that "are designed to allow outstanding research leaders of any nationality and any age to pursue ground-breaking, high-risk projects in Europe."2 This article offers insight into the current state of research in computer-assisted film color analysis and applications that are either available or in development. FilmColors is closely linked to the interactive Timeline of Historical Film Colors, which started in 2012. The platform consists of a comprehensive web resource and online database for all topics related to film colors, with a special focus on their technology, aesthetics, analysis, and restoration. Thus the Timeline is a collection, an archive, a documentation process, and, increasingly, an annotated system for the investigation of the topic, which offers access to a curated body of information to researchers, archivists, film historians, and a broader audience of users from different backgrounds. Four different interdisciplinary approaches investigate the relationship between aesthetics and technical innovation as applied to film colors. First, the database-driven analysis of film color aesthetics, their affective qualities, and their narrative functions aim to identify diachronic aesthetic patterns. Based on this offline database, the research team is developing a computer-assisted tool with a web interface that will allow the crowdsourcing of film color analyses applying recent advancements in digital humanities and custom-made visualizations. Then these aesthetic analyses are connected to the study of film color technology in combination [End Page 72] with chemicophysical analyses of historical color films to understand the influence of film stocks and color processes on films' aesthetic appearance. Third, the team will apply insights gained during the digitization and restoration of historical films by taking into account the requirements of restoration ethics while improving workflows. Finally, three PhD theses in progress will provide case studies from three periods: the emergence of film colors from early applied colors to so-called natural colors (1896–1930), the development of standards in film color technology and aesthetics (1930–55), and the dominance of chromogenic processes (1955–95). The project features a truly interdisciplinary approach to consider and connect all the relevant factors—from technology to perception and aesthetics—because they are closely intertwined. The focus remains firmly rooted in the humanities. METHODOLOGY: STYLISTIC ANALYSIS AND DIGITAL HUMANITIES The planned computer-assisted...

Guest Editors' Foreword:Digital Humanities and/in Film archives Dimitrios Latsis (bio) and Grazia Ingravalle (bio) The continuous redefinition of the role and purview of archivists and curators of moving image media has been driven, in no small measure, by the development of digital tools and networks. To better understand this shift and start mapping its current impact on archives and film preservation, this special issue of The Moving Image assembles perspectives from leading curators, archivists, academics, and digital humanists who have developed innovative platforms to disseminate the work done in moving image archival collections. They provide new tools and resources for both research and pedagogy, share best practices, discuss opportunities for collaboration, and address challenges from leading digital humanities (DH) projects in the audiovisual archival field. Most of these projects are still in progress, so the reader will find that this collection of feature articles and Forum pieces signals the current developing status—or "iterative" nature, to quote Charles Tepperman—of the digital humanities. The contributions compellingly reflect the state of the field, while still leaving open critical questions for future discussion. One such question is certainly whether DH methodologies and tools advance new epistemologies and practices for research in film and media studies and in archival moving image collections. While our contributors reject the idea that by incorporating (partially) automated tasks, DH methodologies lend increased scientific credibility to media analyses and histories, they all highlight the heuristic value of its applications. As the articles in this special issue emphasize, many of the tasks involved in DH projects (creating a database, segmenting a film sequence, annotating, selecting variables, etc.) in fact force us to interrogate established vocabularies, prompting us [End Page xi] once more to (re)define, for instance, "race film," a film "shot," or an "archival record." The digital humanities, as these examples show, encourage interinstitutional and interdisciplinary collaborations among scholars and archivists, inviting them to open up the results of their work to awareness, criticism, and debate. Particular focus is placed on outreach initiatives that give access and visibility to nonfiction, amateur, and nontheatrical film: programming, platforms for user-contributed content, crowdsourcing, and original ways of annotating, sharing, and cross-referencing time-based media. We also address pedagogy that utilizes primary sources, facilitates (under)graduate research, and encourages broader stakeholders, such as K–12 instruction and local and community-based organizations. Shane O'Sullivan approaches these goals by exploring institutional projects that in the last fifteen years have granted access to British audiovisual archival materials for education. O'Sullivan particularly concentrates on the pioneering work of current British Film Institute (BFI) head of education Paul Gerhardt and the collaboration between the BFI and Kingston University on a pedagogical project teaching students to reuse moving image archive material in video essays. Philipp Dominik Keidl expands a thorough consideration of the exhibition strategies of the Australian Centre for the Moving Image in Melbourne into a broader consideration of what media archaeological thinking and practices can mean outside the purview of academia, specifically as enabled by digital technologies and environments. An expanded and crucially public iteration of media archeology can stand as an equal partner and generator of discourse that leads to more conceptual contributions to film and media history. Liliana Melgar Estrada, Eva Hielscher, Marijn Koolen, Christian Gosvig Olesen, Julia Noordegraaf, and Jaap Blom collaboratively survey different video annotation and editing tools widely used in media studies and production. The authors examine two kinds of video software, ELAN and NVivo, to analyze a sequence from People on Sunday (Robert Siodmak et al., 1930). They assess the advantages and drawbacks of each, along with the larger implications for moving image scholars, professionals, and archivists. What would a digital humanities approach to film colors look like? In answering this question, Barbara Flueckiger reflects on the ERC Advanced Grant FilmColors project at the University of Zurich, an extension of one of the best-known DH projects to deal with film history, the Timeline of Historical Film Colors. With computer-assisted tools, such as video annotation and a database of color patterns from a wide array of films, FilmColors aims to merge quantitative and qualitative approaches to demonstrate what [End Page xii] a...

Cinemetrics has emerged as an interdisciplinary field within Digital Humanities, undergoing a developmental trajectory from Barry Salt’s scientistic foundation through Yuri Tsivian’s platform building to the current computational phase driven by computer vision technologies. This article traces the theoretical genealogy of Cinemetrics, with particular attention to its paradigm shift from “container” measurements such as cutting rate and shot duration to “content” analysis of color, sets, props, and costumes. We synthesize research across four thematic dimensions: epistemological debates and methodological paradigms; technical evolution and multimodal expansion; localized development in China and cross-cultural adaptations; and research gaps and future possibilities. Chinese scholars have conducted significant localized explorations in directorial style analysis, genre evolution studies, narrative quantification, production design research, and industrial mode investigations. Yet the field still faces persistent challenges, including small sample sizes, limited causal inference methods, and insufficient integration with film theory. This review advances two core propositions: Cinemetrics represents a fundamental paradigm transformation from meaning-oriented interpretation to form-based measurement; successful research requires integrating three dimensions—technological capability, cultural contextualization, and methodological rigor. Future work should move from descriptive statistics to causal inference, develop localized datasets and open-science platforms, and build constructive dialogue between quantitative measurement and interpretive meaning-making, thereby transforming Cinemetrics from a marginal tool into a core research paradigm.

PurposeThis research aims to analyse the current state of research on the application of Artificial Intelligence (AI) in libraries by examining document type, publication year, keywords, country and research methods. The overarching aim is to enrich the existing knowledge of AI-powered libraries by identifying the prevailing research gaps, providing direction for future research and deepening the understanding needed for effective policy development.Design/methodology/approachThis study used advanced tools such as bibliometric and network analysis, taking the existing literature from the SCOPUS database extending to the year 2022. This study analysed the application of AI in libraries by identifying and selecting relevant keywords, extracting the data from the database, processing the data using advanced bibliometric visualisation tools and presenting and discussing the results. For this comprehensive research, the search strategy was approved by a panel of computer scientists and librarians.FindingsThe majority of research concerning the application of AI in libraries has been conducted in the last three years, likely driven by the fourth industrial revolution. Results show that highly cited articles were published by Emerald Group Holdings Ltd. However, the application of AI in libraries is a developing field, and the study highlights the need for more research in areas such as Digital Humanities, Machine Learning, Robotics, Data Mining and Big Data in Academic Libraries.Research limitations/implicationsThis study has excluded papers written in languages other than English that address domains beyond libraries, such as medicine, health, education, science and technology.Practical implicationsThis article offers insight for managers and policymakers looking to implement AI in libraries. By identifying clusters and themes, the article would empower managers to plan ahead, mitigate potential drawbacks and seize opportunities for sustainable growth.Originality/valuePrevious studies on the application of AI in libraries have taken a broad approach, but this study narrows its focus to research published explicitly in Library and Information Science (LIS) journals. This makes it unique compared to previous research in the field.

The article focuses on the role of Digital Humanities in the creation of a new cyberinfrastructure that relies on the concept of Translation Memory (TM) to create advanced tools in Computer Assisted Translation (CAT). In this new cyberinfrastructure, the source text is broken down into segments and cannot exist in isolation at the same time as older translations act as new source material. The article explores the possibility of establishing a transmedial cartography of the transfer of meaning in literary translations using the notion of a “memographie” prevalent in Digital Humanities. Modelled on hybrid databases, facilitating the creation of tropes in literary language not only from the source text but also from past translations, memographies can facilitate the enhancement of the creative power of the translator in future literary translations. The article uses short extracts from Anglo-American translations of Molière in its case studies, starting from The Playhouse to be Let by Sir William D'Avenant, staged in 1663, up to more recent translations by Richard Wilbur. This is done to illustrate the functioning of the source text and target text coupling in a memorographic plexus that could serve as a new starting point in literary information retrieval systems.

This study aimed to evaluate the effect of pectin addition on the physical and mechanical characteristics of carrageenan-based edible films as an environmentally friendly food packaging material. Pectin addition was carried out at various concentrations of 0–7%, and characterization included tensile strength, elongation at break, thickness, water content, water solubility, film color, and surface morphology analysis using SEM. The results showed that pectin addition increased tensile strength from 24.12 MPa to 38.12 MPa and elongation at break from 59.14% to 70.79%, indicating improvements in mechanical properties and film flexibility. Film thickness, moisture content and water solubility also increased with pectin addition, indicating the hydrophilic nature of pectin in the film matrix. Color analysis showed a decrease in brightness and a slight increase in reddish and yellowish hues, but the color changes remained within acceptable limits (ΔE < 3). SEM analysis showed a denser film surface and indicated aggregation. Overall, the addition of pectin to carrageenan edible film can improve the mechanical and functional quality without films sacrificing visual appearance, thus having the potential to be a biodegradable food packaging material that supports sustainability.

This paper describes a fundamental method for measuring the color of the gingiva with a multilayer color film. The Color Discrimination Index (CDI) which is defined from the photometric results using color film is introduced to express the color of the gingiva.The CDI is independently determined from the color temperature of light sources or the variations, shown both theoretically and experimentally, of exposure value for photographs.It is also shown that the CDI can be successfully applied to analyzing the gingival color with a digital image processing technique.

The research on facial animation is growing and became more sensible in terms of three dimensional data. The emergence of laser scans and advanced 3D tools imparted further towards the rapid development of complex facial model. This paper has given an overview on facial animation techniques such as blend shape interpolation, facial action coding system and parameterization. The major challenges, excitement, applications, recent progress and the future of various diversified facial animation methods are high-lighted as well. The emotional theory, involving the cognition as one of the elements of emotion and emphasizing on somatic factors describing emotional expressions and its perception as the other elements are presented. The essential factors responsible for emo-tional colors manifestations to an individual are the key issues. The notable theories and models of Goethe, Claudia Cortes, Naz Kaya, Color Wheel Pro, Shirley Willet and Yau Xueon facial colors expressions are critically commented and a facile comparison is car-ried out. Furthermore, a detailed discussions on facial colors appearance are provided using image based skin color analysis and synthesis, reflectance based model, bidirec-tional texture function, physiology measurement based and emotional appearance mod-els. The paper completed with the remarkable features official expression comprised of facial action coding system and MPEG-4 facial animation that enable considerable im-provement on facial animation technology. Our in-depth appraisal on facial animation may contribute as taxonomy towards the development of human facial skin colors on 3D avatar including facial action coding system, emotion theory, color theory, animation via blending and image based color analysis useful for various applications.

In this study, biodegradable and active films based on sodium alginate incorporated with different concentrations of oils (25% and 50%) from fruit seeds were developed for potential applications in food packaging. The ultraviolet and visible (UV-VIS) spectra of raspberry seed oil (RSO) and black currant seed oil (BCSO) indicated differences in bioactive compounds, such as tocopherols, phenolic compounds, carotenoids, chlorophyll, and oxidative status (amounts of dienes, trienes, and tetraenes) of active components added to alginate films. The study encompassed the color, structure, and thermal stability analysis of sodium alginate films incorporated with RSO and BCSO and their mixtures. The color of alginate films before and after the addition of oils from both fruit seeds was evaluated by measuring color coordinates in the CIELab color space: L* (lightness), a* (red-green), and b* (yellow-blue). The lightness values ranged between 94.21 and 95.08, and the redness values varied from -2.20 to -2.65, slightly decreasing for the films enriched with oils. In contrast, yellowness values ranged between 2.93 and 5.80 for the obtained active materials, significantly increasing compared to the control alginate film (L* = 95.48, a* = -1.92, and b* = -0.14). Changes in the structure and morphology of the alginate films after incorporating bioactive-rich oils were observed using scanning electron microscopy (SEM). Films with RSO and oil mixtures had more developed surfaces than films with BCSO. Moreover, the cross-sections of the films with RSO showed holes evenly distributed inside the films, indicating traces of volatile compounds. Thermal decomposition of the alginate films loaded with oils showed five separate stages (to 125 °C, 125-300 °C, 310-410 °C, 410-510 °C, and 750-1000 °C, respectively) related to the oil and surfactant decomposition. The shape of the thermogravimetric curves did not depend on the oil type. The added oils reduced the efficiency of alginate decomposition in the first stage. The obtained results showed that new functional and thermally stable food packaging films based on sodium alginate with a visual appearance acceptable to consumers could be produced by utilizing oils from fruit seed residues.

The use of carrageenan-based edible film has increased since it can be functionalized with other biopolymers and active ingredients. Dillenia serrata peel pectin and curcumin were mixed at various proportions to form a carrageenan-based edible film by casting method. In this work, the main objectives are to utilize the peel of Dillenia serrata fruit as a source of pectin combined with curcumin for carrageenan-based edible film production and to assess the effect of various concentrations of pectin and curcumin on the physical, mechanical, barrier, antibacterial properties, and functional group of films. Nine carrageenan-based edible films produced by the casting method were designed on the basis of a complete factorial design with three concentrations of pectin and curcumin. Tensile strength, thickness, WVTR, swelling, colour, antibacterial activity, and FTIR analysis were measured. The results revealed that the concentration of pectin significantly influenced the thickness, WVTR, and swelling, while the addition of curcumin presents significantly influenced the WVTR and colour of films. The edible film containing high pectin and curcumin gave the lowest thickness and WVTR. All films showed a lower inhibitory zone for Escherichia coli than Staphylococcus aureus when higher curcumin was incorporated into the biopolymer matrix. FTIR analysis revealed that curcumin can be used along with Dillenia serrata pectin to form a good-quality carrageenan-based edible film. These findings suggested that carrageenan-based edible film with addition of pectin and curcumin improved overall performance. This approach can be a good strategy to encourage sustainable utilization of endemic fruit wastes (Dillenia serrata fruit peel) for development of edible packaging film.

The aim of this study was to cover biopolymeric packaging films based on PLA/PHBV blend with a functional composite coating (to retain their ecological character) and to investigate their antimicrobial properties before and after UV irradiation. As an active coating, the carrier hydroxypropyl methyl cellulose (HPMC), as well as its modified form with Achillea millefolium L., Hippophae rhamnoides L., and Hypericum L. extract (E) and a combined system based on the extracts and nano-ZnO (EZ), was used to obtain active formulations. Additionally, film surface morphology (SEM, FTIR-ATR) and color (CIELab scale) analysis of the pre- and post-UV-treatment samples were performed. The results confirmed that the E and EZ-modified films exhibited antibacterial properties, but they were not effective against phage phi6. Q-SUN irradiation led to a decrease in the activity of E coating against Staphylococcus aureus, Pseudomonas syringae, and Candida albicans. In this case, the effectiveness of EZ against C. albicans at 24 h and 72 h UV irradiation decreased. However, the irradiation boosted the antiviral effectiveness of the EZ layer. SEM micrographs of the film surface showed that UV treatment did not significantly influence the native film morphology, but it had an impact on the coated film. FTIR analysis results showed that the coatings based on HPMC altered the IR absorption of the nonpolar groups of the biopolyester material. The applied coatings only marginally affected film color changes and increased their yellowness after UV irradiation, whereas a composite layer of nano-ZnO limited these changes.

Gene function discovery in plants, as other plant science quests, is aided by tools that image, document, and measure plant phenotypes. Tools that acquire images of plant organs and tissues at the microscopic level have evolved from qualitative documentation tools, to advanced tools where software-assisted analysis of images extracts quantitative information that allows statistical analyses. They are useful to perform morphometric studies that describe plant physical characteristics and quantify phenotypes, aiding gene function discovery. In parallel, non-destructive, versatile, robust, and user friendly technologies have also been developed for surface topography analysis and quality control in the industrial manufacture sector, such as optoelectronic three-dimensional (3D) color microscopes. These microscopes combine optical lenses, electronic image sensors, motorized stages, graphics engines, and user friendly software to allow the visualization and inspection of objects of diverse sizes and shapes from different angles. This allow the integration of different automatically obtained images along the Z axis of an object, into a single image with a large depth-of-field, or a 3D model in color. In this work, we explored the performance of an optoelectronic microscope to study plant morphological phenotypes and plant surfaces in different model species. Furthermore, as a “proof-of-concept,” we included the phenotypic characterization (morphometric analyses at the organ level, color, and cell size measurements) of Arabidopsis mutant leaves. We found that the microscope tested is a suitable, practical, and fast tool to routinely and precisely analyze different plant organs and tissues, producing both high-quality, sharp color images and morphometric and color data in real time. It is fully compatible with live plant tissues (no sample preparation is required) and does not require special conditions, high maintenance, nor complex training. Therefore, though barely reported in plant scientific studies, optoelectronic microscopes should emerge as convenient and useful tools for phenotypic characterization in plant sciences.

Biobased and biodegradable polymeric materials are a sustainable alternative to the conventional plastics used in food packaging. This study investigated the possible effect of biobased cling films derived from renewable and circular and sustainable sources on key cheese sensory parameters (appearance and odor) able to influence consumer acceptance or rejection of a food product over time. For this purpose, a semi-hard cheese was selected as food model and stored for 14 days at 5 °C wrapped with five cling films: two bio-plastic materials from renewable circular and sustainable sources (R-BP1 and R-BP2), one bio-plastic film from a non-renewable source (NR-BP), and two conventional cling films (LDPE and PVC). Three analytical approaches (image analysis, electronic nose, and sensory test) were applied to evaluate the variation and the acceptability in terms of appearance and odor of the cheese. In preserving cheese color, the R-BP1 and RBP2 films were comparable to LDPE film, while NR-BP film was comparable to PVC film. In terms of odor preservation, R-BP2 film was comparable to LDPE and PVC. The consumer test showed that appearance and odor scores were higher for cheeses stored in R-BP1 and R-BP2 films than NR-BP film. Moreover, in terms of odor, R-BP1 film performed better than conventional films. This study shows how biodegradable cling films from renewable circular and sustainable resources could have comparable performance to conventional plastics (LDPE and PVC) used in the food sector.

This paper proposes a method for quantitative colour analysis aimed at determining colour variation between different sets of films with a common selection criterion. The process is exemplified by comparing films selected at the Cannes Film Festival and the highest-grossing films in the years 1988 and 2018. The method will try to answer several questions. Firstly, whether there are any differences between the films competing in Cannes Film Festival and those most watched by audiences. Secondly, if these differences exist in both the year 1988 and 2018. Thirdly, if the colour of films exhibited at Cannes and highest-grossing film have evolved during the 30-year difference between samples. To analyse these films, one image every 14 s is extracted and all frames are used to quantify the amount of each colour present. Values for each category undergo a Kolmogorov–Smirnov test to determine whether the distribution of each colour is different for both categories compared. The relative difference between colours is computed to find out if those colours where variations are present have increased or decreased their presence within those sets compared. This methodology enables film researchers to statistically prove that there are colour changes in those films included in their research.

Deep learning tools for foreground-aware analysis of film colors