Understanding the cultural concerns of libraries based on automatic image analysis
| Published date | 03 June 2019 |
| Date | 03 June 2019 |
| Pages | 419-434 |
| DOI | https://doi.org/10.1108/EL-11-2018-0229 |
| Author | Heng Ding,Wei Lu,Tingting Jiang |
| Subject Matter | Information & knowledge management,Information & communications technology,Internet |
Understanding the cultural
concerns of libraries based on
automatic image analysis
Heng Ding
Central China Normal University, Wuhan, China, and
Wei Lu and Tingting Jiang
School of Information Management, Wuhan University, Wuhan, China
Abstract
Purpose –Photographs are a kind of cultural heritage and very usefulfor cultural and historical studies.
However, traditional or manual research methods are costly and cannot be applied on a large scale. This
paper aims to present an exploratorystudy for understanding the cultural concerns of librariesbased on the
automaticanalysis of large-scale image collections.
Design/methodology/approach –In this work, an image dataset including 85,023 images preserved
and shared by 28 libraries is collected from the Flickr Commons project. Then, a method is proposed for
representing the culturewith a distribution of visual semantic concepts using a state-of-the-art deep learning
technique and measuring the culturalconcerns of image collections using two metrics. Case studieson this
dataset demonstrated the great potential and promise of the method for understanding large-scale image
collectionsfrom the perspective of cultural concerns.
Findings –The proposed methodhas the ability to discover important cultural units from large-scaleimage
collections. The proposed two metrics are able to quantify the cultural concerns of libraries from different
perspectives.
Originality/value –To the best of the authors’knowledge,this is the first automatic analysis of images for
the purpose of understanding cultural concerns of libraries. The significance of this study mainly consists
in the proposedmethod of understanding the cultural concernsof libraries based on the automatic analysisof
the visual semantic concepts in image collections.Moreover, this paper has examined the cultural concerns
(e.g. importantcultural units, cultural focus, trendsand volatility of cultural concerns) of 28 libraries.
Keywords Digital humanities, Cultural concerns, Image mining,
Photograph-based culture understanding, Visual semantic concepts
Paper type Research paper
1. Introduction
Cultural heritage witnesses where we comefrom and where we are (Serageldin et al.,2001).
In the digital age, libraries are playing an important role in the long-term preservation,
usage and inheritance of cultural heritage in digital forms.Existing studies have discussed
the relationships between culture and libraries. Wang and Frank (2002) highlighted the
important role of academic libraries in cross-cultural communication. Loach et al. (2017)
provided a critique of the sustainability policies and research of museums and libraries.
They believed that the devotion to cultural sustainability is to the future survival of
museums and libraries.
In addition, researchers have been interested in how to use information technologies for
preservation, management, and organisation of cultural heritage (Kalfatovic et al.,2008;
Rimmer et al.,2008). For example, Russel (1967) proposed the standardisation of mark-ups
Automatic
image analysis
419
Received28 November 2018
Revised10 April 2019
Accepted30 May 2019
TheElectronic Library
Vol.37 No. 3, 2019
pp. 419-434
© Emerald Publishing Limited
0264-0473
DOI 10.1108/EL-11-2018-0229
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0264-0473.htm
for the encoding of literary texts in the COCOA programme. In 1998, the famous
“Poughkeepsie Principles”was put forward as the text encoding guideline for literary,
linguistic and historicalresearch. Then, the extensible mark-up language (XML) became the
guideline of text encoding for literary and linguistic computing (Text Encoding Initiative
[TEI] Consortium, 2009).To support the preservation of digitalobjects and ensure long-term
usability, the Library of Congress released the PREMIS Data Dictionary and Schema as an
international metadata standard (Caplan and Guenther, 2005). Rimmer et al. (2008)
established the Arts and Humanities Data Service to help researchers with the discovery
and preservation of digitalresources.
More recently, the Flickr Commonsproject collected thousands of image collections from
many libraries, which might be used to gain a better understanding of the culture and
history of libraries (Springer et al., 2008). However, traditional research methods (i.e.image
collections analysed by humans) are costly and cannot be applied on a large scale. In this
paper, the researchers report an exploratory study to understand the cultural concerns of
libraries based on the automatic analysis of large-scale image collections. The value of this
study is twofold. On the one hand, the methodproposed has the ability to analyse large-scale
image collections quantitatively. On the other hand, the study included an examination of
the cultural units that libraries were concerned with as well as the cultural concerns of
different libraries.
The rest of the paper is organised as follows. In Section 2, the backgroundof this study is
introduced, including the quantification of culture and the automatic understanding of
images, both of which are related to image mining in the digital humanities. In Section 3, a
method is proposed of understanding large-scale image collections from the perspective
of cultural concerns. The method first represents culture with a series of visual semantic
concepts in the image collections. Second, two metrics are used to quantify the cultural
concerns of the image collections. In Section 4, an image dataset collected from the Flickr
Commons project, including 85,023 images preserved and shared by 28 libraries, is
described to test the method. In Section 5, the effectiveness of the method is demonstrated
based on case studies using the collected dataset and the findings about the cultural
concerns of libraries.
2. Background
The main idea of this study was to quantify the cultural concerns based on the automatic
analysis of visual semantic concepts in the image collections. The related studies are thus
reviewed in terms of two topics; the quantification of culture and the automatic
understanding of images.
2.1 Quantification of culture
It has been a long time since people started to use quantitative methods to study culture
from various aspects. To our best knowledge, Hofstede (1984) was the first who used a
quantitative method to study culture from multiple cultural dimensions, such as power
distance, individualism versus collectivism, masculinity versus femininity, uncertainty
versus avoidance, long-term versus short-term orientation and indulgence versus restraint.
Following Hofstede’scultural dimension theory and quantitative method, Stuart-Fox(1986),
Ger and Belk (1996) and Schwartz (2006) quantified and compared culture with manual
methods (e.g. surveys).
During the past few decades, it was popular to use computational methods to study
culture based on special cultural carriers (e.g. books and social media). Stubbs (1996)
conducted an analysis of advertisements, newspapers, and scientific research articles in
EL
37,3
420
To continue reading
Request your trial