Digital Image Collections Research Articles

In a Perfect World: Exploring the Desires and Realities for Digitized Historical Image Archives

ABSTRACTThe primary goal of this paper is to explore users' desires for digitized historical image collections, examining their desires based on different use purposes and information interaction activities. In addition, we investigate the image attributes that users wish to search from the collection. To accomplish this, we conducted 21 qualitative interviews with active users of a digitized historical photograph archive. Our findings suggest that users' desires relate to three contexts: tools, collection, and socio‐organizational issues. Moreover, our results indicate that users require support for various information interaction activities, not just searching. We found that users' desires vary based on their specific use purposes, and that users prioritize conceptual access points that can already mostly be generated through automated annotation methods. Ultimately, this study contributes to a better understanding of users' real‐life image needs and offers implications for improving access to digital image collections.

Eastern Christian and Islamic Manuscripts in Minnesota: Handwritten, Microfilm, and Digital

Abstract: This article is a brief overview of Eastern Christian and Islamic collections in Minnesota, with a focus on the holdings of the largest such collection, located at the Hill Museum and Manuscript Library (HMML) in Collegeville. Minnesota's manuscripts are largely defined by their digital presence and physical absence, as HMML has amassed the world's largest collection of digital manuscript images while the digitized manuscripts remain in libraries around the world. However, HMML holds a (relatively) small collection of physical manuscripts as well, which is the focus of this survey. The collection includes Islamic manuscripts in Arabic, Persian, and Turkish, along with Christian manuscripts in Amharic, Arabic, Armenian, Church Slavonic, Coptic, Geʻez, Georgian, Greek, Russian, and Syriac. Highlights include twenty-one Geʻez magic scrolls, three copies of Muḥammad al-Jazūlī's Arabic prayer book Dalāʼil al-khayrāt , late antique Coptic and Greek texts on papyrus and wood, and a Georgian palimpsest fragment with two Syriac undertexts. The article describes the history of the institution and its manuscript holdings and gives an outline of the collection's contents.

Development of a grain-size determination method based on digital images of muddy sediments in coastal areas

The grain-size distribution is a fundamental parameter for assessment of coastal environments. In coastal areas, grain-size analysis of sediments is hampered by arduous collection of sediments from the seafloor at each sampling site and long analysis time. Although rapid determination methods using digital imaging have been developed over the last two decades, these methods cannot be applied to muddy sediments in coastal areas. In the present study, a digital image analysis method applicable to muddy sediments was developed based on sediment color analysis combined with an autocorrelation algorithm introduced by Rubin (2004). For digital image collection in the laboratory, a waterproof digital camera (16 megapixels) was used; it was placed 9 cm from the sediment surface. The imaging equipment consisted of the digital camera, an acrylic plate, and a housing filled with tap water, which was used to avoid effects of suspended sediment in the seawater. Model sediments were prepared and images of the sediment surfaces were taken. The mud and coarse fraction of the sediment were successfully estimated through binarization of the image based on brightness values and the grain-size distribution of the coarse fraction was determined using the autocorrelation algorithm introduced by Rubin (2004). In the field, the imaging apparatus consisting of a stainless-steel cage, lighting equipment, a housing filled with tap water, and a wide acrylic plate attached to the bottom of the housing provided a sturdy structure for obtaining clear digital images from uneven field sediments with suspended particles in the bottom water. The ship was moved several hundred meters at a time; the imaging apparatus was allowed to settle on the sediments and images were then taken continuously at set intervals. From these images, grain-size distributions of sediments containing >15% mud were successfully estimated. Thus, the digital image analysis method developed here will be suitable for measuring the fraction of surface area covered by mud and the size of the visible coarse fraction, and will be useful for monitoring of various sediment types, including muddy coastal sediments, demonstrating great promise for elucidating benthic ecosystem structures and functions in coastal areas.

User Study: Implementation of OCLC FAST Subject Headings in the Lafayette Digital Repository

Digital repository migrations present a periodic opportunity to assess metadata quality and to perform strategic enhancements. Lafayette College Libraries implemented OCLC FAST (Faceted Application of Subject Terminology) for its digital image collections as part of a migration from multiple repositories to a single one built on the Samvera Hyrax open-source framework. Application of FAST has normalized subject headings across dissimilar collections in a way that tremendously improves descriptive consistency for staff and discoverability for end users. However, the process of applying FAST headings was complicated by several features of in-scope metadata as well as gaps in available controlled subject authorities.

DeepErythema: A Study on the Consistent Evaluation Method of UV SPF Index Through Deep Learning

This study presents a deep learning-based approach for consistent Sun Protection Factor (SPF) grading by quantifying skin erythema. Despite strict international standards, the current human-based Minimum Erythema Dose (MED) determination method for SPF evaluation suffers from inconsistency due to subjective criteria and the visual characteristics of erythema’s large variance. We propose DeepErythema, a novel method that uses erythema quantification for MED determination. The proposed method comprises pre-processing methods, a deep learning segmentation model, and post-processing methods. The UV irradiation area pointing is a pre-processing method that accurately detects the inspection area as a UV irradiated port, while Median Gradation Reduction eliminates the gradation that arises due to the digital image collection environment. The deep learning segmentation model parts introduce a methodology for improving performance using various feature extractors and methodologies like SeLu, Reverse Attention Gate, and MixedLoss. For post-processing, Perspective Transform Rectangle Restoration restores a distorted inspection area, and Relative Density Evaluation calculates a density score to consider skin characteristics and tone. We verified that utilizing the DeepErythema score in the UV SPF MED Evaluation (USME) dataset reduced the distribution of human-based MED decisions from 36 to 9 in the experimental results. This reduction in MED decision scope leads to improved consistency in SPF Index grading. Overall, the study contributes to the development of an objective and consistent evaluation method for SPF grading.

A Smart Content-Based Image Retrieval Approach Based on Texture Feature and Slantlet Transform

With the advancement of digital storing and capturing technologies in recent years, an image retrieval system has been widely known for Internet usage. Several image retrieval methods have been proposed to find similar images from a collection of digital images to a specified query image. Content-based image retrieval (CBIR) is a subfield of image retrieval techniques that extracts features and descriptions content such as color, texture, and shapes from a huge database of images. This paper proposes a two-tier image retrieval approach, a coarse matching phase, and a fine-matching phase. The first phase is used to extract spatial features, and the second phase extracts texture features based on the Slantlet transform. The findings of this study revealed that texture features are reliable and capable of producing excellent results and unsusceptible to low resolution and proved that the SLT-based texture feature is the perfect mate. The proposed method's experimental results have outperformed the benchmark results with precision gaps of 28.0 % for the Caltech 101 dataset. The results demonstrate that the two-tier strategy performed well with the successive phase (fine-matching) and the preceding phase (coarse matching) working hand in hand harmoniously.

Dolphin Distribution and Habitat Suitability in North Western Australia: Applications and Implications of a Broad-Scale, Non-targeted Dataset

Understanding species’ distribution patterns and the environmental and ecological interactions that drive them is fundamental for biodiversity conservation. Data deficiency exists in areas that are difficult to access, or where resources are limited. We use a broad-scale, non-targeted dataset to describe dolphin distribution and habitat suitability in remote north Western Australia, where there is a paucity of data to adequately inform species management. From 1,169 opportunistic dolphin sightings obtained from 10 dugong aerial surveys conducted over a four-year period, there were 661 Indo-Pacific bottlenose dolphin (Tursiops aduncus), 191 Australian humpback dolphin (Sousa sahulensis), nine Australian snubfin dolphin (Orcaella heinsohni), 16 Stenella sp., one killer whale (Orcinus orca), one false killer whale (Pseudorca crassidens), and 290 unidentified dolphin species sightings. Maximum Entropy (MaxEnt) habitat suitability models identified shallow intertidal areas around mainland coast, islands and shoals as important areas for humpback dolphins. In contrast, bottlenose dolphins are more likely to occur further offshore and at greater depths, suggesting niche partitioning between these two sympatric species. Bottlenose dolphin response to sea surface temperature is markedly different between seasons (positive in May; negative in October) and probably influenced by the Leeuwin Current, a prominent oceanographic feature. Our findings support broad marine spatial planning, impact assessment and the design of future surveys, which would benefit from the collection of high-resolution digital images for species identification verification. A substantial proportion of data were removed due to uncertainties resulting from non-targeted observations and this is likely to have reduced model performance. We highlight the importance of considering climatic and seasonal fluctuations in interpreting distribution patterns and species interactions in assuming habitat suitability.

An Improved Automatic Image Annotation Approach Using Convolutional Neural Network-Slantlet Transform

Every day, websites and personal archives generate an increasing number of photographs. The extent of these archives is unfathomable. The ease of usage of these enormous digital image collections contributes to their popularity. However, not all of these databases provide appropriate indexing data. As a result, it’s tough to find information that the user is interested in. Thus, in order to find information about an image, it is necessary to classify its content in a meaningful way. Image annotation is one of the most difficult issues in computer vision and multimedia research. The objective is to convert an image into a single or numerous labels. This necessitates a grasp of the visual content of an image. The necessity for unambiguous information to build semantic-level concepts from raw image pixels is one of the challenges of image annotation. Unlike text annotation, where a dictionary links words to their meaning, raw picture pixels are insufficient to construct semantic-level notions directly. A simple syntax, on the other hand, is well specified for combining letters to form words and words to form sentences. The automatic feature extraction for automatic annotation was the emphasis of this paper. And they employed a deep learning convolutional neural network to build and improve image coding and annotation capabilities. Performance of the suggested technique on the Corel-5K, ESP-Game, and IAPRTC-12 datasets. Finally, experimental findings on three data sets were used to demonstrate the usefulness of this model for image annotation.

Cluttering Reduction for Interactive Navigation and Visualization of Historical Images

Abstract. Iconographic representations, such as historical photos of geographic spaces, are precious cultural heritage resources capable of describing a particular geographical area’s evolution over time. These photographic collections may vary in size, between hundreds and thousands of items. With the advent of the digital era, many of these documents have been digitized, spatialized, and are available online. Browsing through these digital image collections represents new challenges. This paper examines the topic of historical image exploration in a virtual environment enabling the co-visualization of historical photos into a contemporary 3D scene. We address the topic of user interaction considering the potential volume of the input data. Our methodology is based on design guidelines that rely on visual perception techniques to ease visual complexity and improve saliency on specific cues. The designs are additionally implemented following an image-based rendering approach and evaluated in a group of users. Overall, these propositions may be a notable addition to creating innovative ways to visualize and discover historical images in a virtual geographic environment.

Q-S laser micro-drilling and multipass full-beam Q-S laser for tattoo removal \u2014 a case series

The purpose of this study was to evaluate the safety and efficacy of a new combined method of Q-S laser-assisted tattoo removal. Ten patients with 13 professional, mostly mono-chromatic black tattoos were recruited. All tattoos received the same Q-S laser treatment sequence. An objective evaluation of tattoo clearing was assessed by careful analysis of a standardized collection of digital images taken from each tattoo, 2 months after each laser session, with the help of a custom-made pigment-fading percentage photographic ruler. The percentages of pigment clearance and side effects were evaluated by 4 independent dermatologists. Patient satisfaction and perceived discomfort during and post-procedure were evaluated according to specific scales. Clinical evaluators confirmed an average photographic pigment clearance of 97% after a median 4.85 treatment sessions. The Frac-Tat® method required 40% fewer sessions compared to those calculated by Kirby-Desai estimates. Photographic assessment of laser-exposed skin quality performed 2 months after tattoo clearing was considered almost comparable with untreated peripheral skin, confirming a very low side effect score. The Frac-Tat QS laser-assisted tattoo removal sequence used in our study showed a high degree of safety and efficiency, clearing exogenous pigments in a relatively few number of sessions. Preliminary ablative photo-acoustic fractional 1064-nm Q-S laser micro-drilling was considered an essential step in optimizing tattoo removal, increasing wavelength-independent micro-columnar clearing of deeper dermal exogenous pigments. Our preliminary observations also confirmed a significant improvement of tattoo procedure-induced micro-textural changes thanks to a tissue remodeling effect induced by the 1064-nm Q-S fractional laser photo-acoustic ablation.

Figure search by text in large scale digital document collections

SummaryDigital document collections have been created with the transfer of a large number of documents to digital media. These digital archives have provided many benefits to users. As the diversity and size of digital image collections have grown exponentially, it has become increasingly important and difficult to obtain the desired image from them. The images on the document might contain critical information about the subject of it. In this study, an architecture is developed that can work on large‐scale data by creating regular expressions together with full‐text search approaches. The performance of the system has been tested on different academic documents and Elasticsearch and Apache Solr insert times are compared. Compared to Elasticsearch, Apache Solr achieved faster and more successful results.

Transfer learning for hostel image classification

PurposeBecause of the fast-growing digital image collections on online platforms and the transfer learning ability of deep learning technology, image classification could be improved and implemented for the hostel domain, which has complex clusters of image contents. This paper aims to test the potential of 11 pretrained convolutional neural network (CNN) with transfer learning for hostel image classification on the first hostel image database to advance the knowledge and fill the gap academically, as well as to suggest an alternative solution in optimal image classification with less labour cost and human errors to those who manage hostel image collections.Design/methodology/approachThe hostel image database is first created with data pre-processing steps, data selection and data augmentation. Then, the systematic and comprehensive investigation is divided into seven experiments to test 11 pretrained CNNs which transfer learning was applied and parameters were fine-tuned to match this newly created hostel image dataset. All experiments were conducted in Google Colaboratory environment using PyTorch.FindingsThe 7,350 hostel image database is created and labelled into seven classes. Furthermore, its experiment results highlight that DenseNet 121 and DenseNet 201 have the greatest potential for hostel image classification as they outperform other CNNs in terms of accuracy and training time.Originality/valueThe fact that there is no existing academic work dedicating to test pretrained CNNs with transfer learning for hostel image classification and no existing hostel image-only database have made this paper a novel contribution.

Minor Politics, Major Consequences

Abstract Metadata is part of our knowledge systems and, so, represents and perpetuates political hierarchies and perceptions of relevance. While some of these have come up for scrutiny in the discourses on digitization, some ‘minor’ issues have gone unnoticed and a few new mechanisms of imbalance have escaped attention as well. Yet, all of these, too, influence the usability of digital image collections. This paper traces three fields of ‘minor politics’ and their epistemic consequences, both in general and in particular, with respect to the study of architecture and its visual representation: first, the intrinsic logic of the original collections and their digital representation; second, the role of support staff in the course of digitization and data transfer; and, third, keywording as a matter of disciplinary habitus. It underlines the ‘political’ role of metadata within the context of knowledge production, even on the local level of a single database, and connects to the implementation of contemporary technologies like computer vision and artificial intelligence for image content classification and the creation of metadata. Given the abundance of digitally available (historical) images, image content recognition and the creation of metadata by artificial intelligence are sheer necessities in order to make millions of hitherto unexplored images available for research. At the same time, the challenge to overcome existing colonial and other biases in the training of AI remains. Hence, we are once again tasked to reflect on the delicate criterion of objectivity. The second part of this paper focuses on research done in the ArchiMediaL project (archimedial.eu); it demonstrates both the potentials and the risks of applying artificial intelligence for metadata creation by addressing the three fields mentioned above through the magnifying glass of programming.

A ten-year literature review of content-based image retrieval (CBIR) studies in the tourism industry

Purpose Due to the worldwide growth of digital image sharing and the maturity of the tourism industry, the vast and growing collections of digital images have become a challenge for those who use and/or manage these image data across tourism settings. To overcome the image indexing task with less labour cost and improve the image retrieval task with less human errors, the content-based image retrieval (CBIR) technique has been investigated for the tourism domain particularly. This paper aims to review the relevant literature in the field to understand these previous works and identify research gaps for future directions. Design/methodology/approach A systematic and comprehensive review of CBIR studies in tourism from the year 2010 to 2019, focussing on journal articles and conference proceedings in reputable online databases, is conducted by taking a comparative approach to critically analyse and address the trends of each fundamental element in these research experiments. Findings Based on the review of the literature, the trends of CBIR studies in tourism is to improve image representation and retrieval by advancing existing feature extraction techniques, contributing novel techniques in the feature extraction process through fine-tuning fusion features and improving image query of CBIR systems. Co-authorship, tourist attraction sector and fusion image features have been in focus. Nonetheless, the number of studies in other tourism sectors and available image databases could be further explored. Originality/value The fact that no existing academic review of CBIR studies in tourism makes this paper a novel contribution.

Content-Based Image Retrieval Using Discrete Wavelet Transform and Convolutional Neural Network

Due to the development of internet and digital media techniques the size of digital image collection is increasing rapidly. Various techniques for storing, searching and retrieving images become essential for large image archives. This leads to an existence of Image Retrieval System. Content-Based Image Retrieval (CBIR) is a process of retrieving the expected images from large image databases based on the features of the query image and this is performed by various feature extraction techniques. In this work, two different methods of CBIR have been proposed. The first method is CBIR using Discrete Wavelet Transform (DWT), which extracts the texture feature of an image and histogram extracts the color feature of an image. And also the results will be ranked according to the chi-square distance calculation. The second method is CBIR using Convolutional Neural Network (CNN). In CNN method the features are extracted using VGG16 pre-trained model and canny edge detection is used for detecting the edges of an image. Finally, the performances of DWT and CNN based methods are analyzed and compared.

New Approach on the Techniques of Content-Based Image Retrieval (CBIR) Using Color, Texture and Shape Features

Digital image collection as rapidly increased along with the development of computer network. Image retrieval system was developed purposely to provide an efficient tool for a set of images from a collection of images in the database that matches the user’s requirements in similarity evaluations such as image content similarity, edge, and color similarity. Retrieving images based on the content which is color, texture, and shape is called content based image retrieval (CBIR). The content is actually the feature of an image and these features are extracted and used as the basis for a similarity check between images. The algorithms used to calculate the similarity between extracted features. There are two kinds of content based image retrieval which are general image retrieval and application specific image retrieval. For the general image retrieval, the goal of the query is to obtain images with the same object as the query. Such CBIR imitates web search engines for images rather than for text. For application specific, the purpose tries to match a query image to a collection of images of a specific type such as fingerprints image and x-ray. In this paper, the general architecture, various functional components, and techniques of CBIR system are discussed. CBIR techniques discussed in this paper are categorized as CBIR using color, CBIR using texture, and CBIR using shape features. This paper also describe about the comparison study about color features, texture features, shape features, and combined features (hybrid techniques) in terms of several parameters. The parameters are precision, recall and response time.

L'Edizione digitale dell'epistolario di Alcide De Gasperi: Alcune riflessioni su un progetto in corso

This article presents the National Edition of Alcide De Gasperi’s correspondence project (www.epistolariodegasperi.it). An introduction about digital editions, where some examples of the publication of letters in digital form are offered, is followed by a description of the project, of the reasons why the online edition was chosen and of the technical features of the software for the collection of metadata, digital images and transcriptions. Finally, after a description of the methods for the publication of correspondence on the open access platform of the National Edition, the future developments of the publishing initiative, especially in terms of browsing the records in the database, are summarized.

PHAROS: A digital research space for photo archives

The PHAROS consortium of fourteen international art historical photo archives is digitizing the over 20 million images (with accompanying documentation) in its combined collections and has begun to construct a common access platform using Linked Open Data and the ResearchSpace software. In addition to resulting in a rich and substantial database of images for art-historical research, the PHAROS initiative supports the development of shared standards for mapping and sharing photo archive metadata, as well as for best practices for working with large digital image collections and conducting computational image analysis. Moreover, alongside their digitization efforts, PHAROS member institutions are considering the kinds of art-historical questions the resulting database of images could be used to research. This article indicates some of the prospective research directions stimulated by modern technologies, with the aim of exploring the epistemological potential of photographic archives and challenging the boundaries between the analogue and the digital.

Content based Image Retrieval System using Combination of Color and Shape Features, and Siamese Neural Network

With an advent of technologya huge collection of digital images is formed as repositories on world wide web (WWW). The task of searching for similar images in the repository is difficult. In this paper, retrieval of similar images from www is demonstrated with the help of combination of image features as color and shape and then using Siamese neural network which is constructed to the requirement as a novel approach. Here, one-shot learning technique is used to test the Siamese Neural Network model for retrieval performance. Various experiments are conducted with both the methods and results obtained are tabulated. The performance of the system is evaluated with precision parameter and which is found to be high.Also, relative study is made with existing works.

Billedindeksering og sociale medier

This article focuses on the methodologies, organization, and communication of digital image collection research that utilizes social media content. “Image” is here understood as a cultural, conventional, and commercial—stock photo—representation. Two methodologies i.e. PRISMA and Grounded Theory are employed to examine, categorize and analyze images and comprehend how humans consider them. The literature review covers research since 2005, when major social media platforms emerged. It demonstrates that the images on social media have not changed the overall direction of research into image indexing and retrieval, though new topics on crowdsourcing and tagging have emerged in it. A citation analysis includes an overview of co-citation maps that demonstrate the nexus of image research literature and the journals in which they appear. The results point to new possibilities influencing the research by providing large image collections as new testbeds for improving or testing research hypothesis on a new scale.