ArticlePublisher preview available

Authente-Kente: enabling authentication for artisanal economies with deep learning

Authors:
Kwame Porter Robinson at Wayne State University
Kwame Porter Robinson
Ron Eglash at University of Michigan
Ron Eglash
Audrey Bennett at University of Michigan
Audrey Bennett
Sansitha Nandakumar at University of Michigan
Sansitha Nandakumar
Show all 5 authorsHide
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

The economy for artisanal products, such as Navajo rugs or Pashmina shawls are often threatened by mass-produced fakes. We propose the use of AI-based authentication as one part of a larger system that would replace extractive economies with generative circulation. In this case study we examine initial experiments towards the development of a cell phone-based authentication app for kente cloth in West Africa. We describe the context of weavers and cloth sales; an initial test of a machine learning algorithm for distinguishing between real and fake kente, and an outline of the next stages of development.
Figures - available from: AI & SOCIETY
This content is subject to copyright. Terms and conditions apply.
A preview of this full-text is provided by Springer Nature.
Learn more
Content available from AI & SOCIETY
This content is subject to copyright. Terms and conditions apply.
Vol.:(0123456789)
1 3
AI & SOCIETY (2021) 36:369–379
https://doi.org/10.1007/s00146-020-01055-2
OPEN FORUM
Authente‑Kente: enabling authentication forartisanal economies
withdeep learning
KwamePorterRobinson1 · RonEglash1 · AudreyBennett2 · SansithaNandakumar1 · LionelRobert1
Received: 24 June 2020 / Accepted: 18 August 2020 / Published online: 2 September 2020
© Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract
The economy for artisanal products, such as Navajo rugs or Pashmina shawls are often threatened by mass-produced fakes.
We propose the use of AI-based authentication as one part of a larger system that would replace extractive economies with
generative circulation. In this case study we examine initial experiments towards the development of a cell phone-based
authentication app for kente cloth in West Africa. We describe the context of weavers and cloth sales; an initial test of a
machine learning algorithm for distinguishing between real and fake kente, and an outline of the next stages of development.
Keywords Human–machine collaboration· Machine learning· Artisanal economy· Generative justice· Industrial
symbiosis· Ethnocomputing
1 Introduction
Mass production economies have introduced many ills into
social life, including high rates of mental, and physical ill-
ness from dull repetitive jobs; high rates of environmental
degradation; and a “junk food” approach to both physical
and informational over-consumption (Michelsen and Bildt
2003; Coccia 2017; Hunt etal. 2018). Artisanal fabrication,
in contrast, tends to embody the opposite effect. Artisans
often report that they are drawn to their craft because it is
an enjoyable and rewarding form of labor. Traditional fab-
rication methods often use locally sourced and sustainable
supply chains. And (at least traditionally) artisanal items
were purchased in more thoughtful ways, often establish-
ing a personal relationship between buyer and seller. In
our prior work (Eglash etal. 2019), we suggested that AI,
robotics, and other forms of automation, if properly designed
and implemented, could gradually scale these beneficial sys-
tems towards the development of an artisanal economy. One
small step in that direction might be AI guides that help con-
nect consumers with artisanal producers. In this paper, we
explore a prototype, Authente-Kente, to help guide consum-
ers toward selection of authentic hand-woven kente cloth,
and thus diminish income loss due to mass produced fake
cloth.
2 The problem context
Traditional artisanal items often compete with mass-pro-
duced fakes. M’Closkey (2010) estimates that out of roughly
2 billion in annual sales of “Native American” goods, about
50% is not actually of Native origin. Similar problems arise
elsewhere: for example, Mehra (2019) reports that due to
competition from mass produced fakes, the number of Pash-
mina shawl hand weavers had dropped from over 100,000
in 2007 to about 10,000 in 2017. The artisanal product of
concern in our case study is Kente cloth, a traditional fabric
from the West African nation of Ghana. Kente is of great
symbolic importance as a symbol of African heritage, not
only in Ghana but also abroad. African American students
* Kwame Porter Robinson
kwamepr@umich.edu
Ron Eglash
eglash@umich.edu
Audrey Bennett
agbennett@umich.edu
Sansitha Nandakumar
sansitha@umich.edu
Lionel Robert
lprobert@umich.edu
1 University ofMichigan School ofInformation, 3360 North
Quad, 105 S. State St., AnnArbor, MI48109-1285, USA
2 University ofMichigan Penny W. Stamps School ofArt
andDesign, AnnArbor, USA
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
... However, availability of such data and their translation in practical and cost effective applications does not seem viable. In this context, (Iqbal Hussain et al. 2020;Robinson et al. 2021) have shown the use of deep learning solutions for fabric identification from fabric imagery data. Iqbal Husain et al., designed and developed a deep learning-based solution to recognize woven fabrics of varied textures and physical properties. ...
... Iqbal Husain et al., designed and developed a deep learning-based solution to recognize woven fabrics of varied textures and physical properties. Similarly, (Robinson et al. 2021) has developed a deep learning model Authente-Kente to authenticate Kente cloth from imagery data with the accuracy of 88%. These studies laid the foundation of our hypothesis to build a deep learning model for authentication of unadulterated Pashmina fabric. ...
... Deep learning can differentiate two types of fabrics by analysing their unique visual and structural features like colour, texture, pattern, and other attributes that are not visible to the human eye (Robinson et al. 2021). Besides the efficiency and versatility of deep models, they can also process large amounts of data quickly with very little economic implications, making them easy to scale and distribute. ...
Pashmina authentication on imagery data using deep learning
Article
Full-text available
  • May 2023
  • AI Soc
Pashmina is one of the most luxurious and finest fibres in the world. It is a special kind of wool obtained from Cashmere goats. Counterfeiting Pashmina is becoming a prevalent malpractice because of limited supply, expensive pricing and high demand in western markets. Presently, there is a lack of a low-cost and easily available approach for distinguishing authentic Pashmina apparels from other similar-looking products. Because of technological advances and cost reductions in digital image processing, we have been able to implement a variety of image recognition systems on a practical level. In this study, we propose a deep learning model PashmNet for identification of Pashmina fabric using transfer learning. The technique proposed herein classifies Pashmina and non-Pashmina images accurately with classification accuracy of 97.4%.
View
... We call this platform "Ubuntu-AI" (online at https://ubuntu-ai.net/). The term is not original to us; several researchers (Birhane, 2021;Mhlambi, 2020;Robinson et al. 2021) have proposed that the ubuntu principle --often summarized as "I am because we are"--be utilized as a guiding principle for AI. It shares the concept of relational economy envisioned in the generative justice literature (https://generativejustice.org/publications/), ...
... Maybe they want to pay for a high resolution version of the image for a book, or they are looking to pay someone to give a live demo of a weaving technique. Showing the students in your class some kente cloth is nice, but letting them meet the person who created it, and have a discussion with them, is starting to look more like a relational economy (Robinson et al. 2021). ...
Interim Report for Ubuntu-AI: A Bottom-up Approach to More Democratic and Equitable Training and Outcomes for Machine Learning
Conference Paper
Full-text available
  • Sep 2023
Artificial Intelligence (AI) can be a threat to creative arts and design, taking data and images without permission or compensation. But with AI becoming a global portal for human knowledge access, anyone resisting inclusion in its data inputs will become invisible to its outputs. This is the AI double bind, in which the threat of exclusion forces us to give up any claims of ownership to our creative endeavors. To address such problems, this project develops an experimental platform designed to return value to those who create it, using a case study on African arts and design. If successful, it will allow African creatives to work with AI instead of against it, creating new opportunities for funding, gaining wider dissemination of their work, and creating a database for machine learning that results in more inclusive knowledge of African arts and design for AI outputs.
View
... Online marketplaces powered by AI allow local artisans to sell their goods [43]. These platforms utilize machine learning algorithms to determine visitors' likes and recommend locally manufactured crafts, artwork, and souvenirs. ...
View
... One possible prototype for such functionalities comes from a project in which we developed an AI system for detecting the difference between hand-made textiles and factory fakes (Robinson et al., 2021). This has been extended to examine how that level of identification could make more meaningful connections between specific makers and buyers: The potential of AI to develop a relational economy (Birhane, 2021;Mhlambi, 2020;Nayebare et al., 2023). ...
Computational reparations as generative justice: Decolonial transitions to unalienated circular value flow
Article
Full-text available
  • Jan 2024
The Latin roots of the word reparations are "re" (again) plus "parere" which means "to give birth to, bring into being, pro-duce". Together they mean "to make generative once again". In this sense, the extraction processes that cause labor injustice, ecological devastation, and social degradation cannot be repaired by simply transferring money. Reparations need to take on the full sense of "restorative": the transition to a decolonial system that can support value generators in the control of their own systems of production, protect the value they create from extraction, and circulate value in unalienated forms that benefit the human and non-human communities that produced that value. With funding from the National Science Foundation, we have developed a research framework for this process that starts with "artisanal labor": employee-owned business and worker collectives that have people doing what they love, despite low incomes. Focusing primarily on Detroit's Black-owned urban farms, artisanal textile businesses, Black hair salons, worker collectives, and other community-based production, with additional connections to Indigenous and other communities, we have introduced digital fabrication technologies, sensors, artificial intelligence, server-side apps and other computational support for a transition to unalienated circular value flow. We will report on our investigations with the challenges at multiple scales. At each level, we show how computational supports can act as restorative mechanisms for lost circular value flows, and thus address both past and ongoing disenfranchisement.
View
... Hence, if information scientists ignore the ways of learning in other landscapes by perceiving information literacy as a sequence of objective skills, decontextualised, they risk ignoring the ways of learning in other landscapes in becoming information literate (Lloyd, 2006). Despite that there are explorative studies on Kente (Boateng, 2018;Cohen, 2019;Nunoo et al., 2021;Robinson et al., 2021), no adequate explanation prevails on how learning is enacted to enable novice weavers to become information literate and therefore knowledgeable or competent weavers in the Kente workplace landscape. Also, how novice weavers engage with information to become information literate in the Kente-weaving landscape appears to be non-existent in the literature. ...
workplace-learning-and-information-literacy-in-the-kente-weaving-landscape
Article
Full-text available
  • Jun 2023
  • J LIBR INF SCI
This study investigates how workplace learning is enacted to make a novice Kente (a hand-woven fabric in Ghana) weaver information literate in the Kente-weaving landscape. Ethnography was used as the research design. Interview and participant observation were used as the data collection methods. For the interview, semi-structured interview techniques were used to solicit information from all three levels of weavers (Master, junior and novice weavers) in Bonwire Kente Centre in Ghana. Out of the 62 weavers at the Bonwire Kente Centre, 24 weavers representing 8 each from each level of weavers were purposively chosen. The findings of this study show learning is enacted to make a novice Kente weaver information literate of the Kente-weaving craft by the affordance of guidance provision, conversations, observation and learning by doing. This study contributes to the information literacy literature and workplace learning that learning and becoming information literate do not relate solely to the cognitive activities of the mind, but also to the body through the affordance of the workplace.
View
... Our Detroit participants noted the threats raised for this feature in the shift to online consumption. One of our pilot projects (Robinson et al., 2021) developed AI for distinguishing between hand-woven cloth, and the factory-made fakes that threaten this artisanal tradition by cost cutting. While the first phase was simply a proof-ofconcept (which works; it can now make the distinction with a single photo), the goal is ultimately to help reconnect artisans to consumers: in the case of real cloths, the aim is a future app which will identify the origin and allow "value added" (for example a guest lecture by the weaver for a class, further purchases from a particular family or collective, and so on). ...
Counter-hegemonic AI: The Role of Artisanal Identity in the Design of Automation for a Liberated Economy
Chapter
Full-text available
  • Apr 2023
Transformative improvements require systemic change in an economy marked by extreme wealth inequality, stratified by geography, identity, and other social markers. In this chapter, we seek to raise awareness in the technology, design, and scientific communities of the long history of artisans—skilled, independent labor striving to keep a relatively unalienated workplace—that can offer a crucial resource to those interested in designing for a liberated economy. It is a resource as a history of counter-hegemonic movements and identities organized around technology. But it is also a potential site for participatory design, solidarity design and other methods for co-developing innovative strategies by which automation can support the rise of an unalienated economy. Rather than simply a throwback to a romantic past, artisans offer a present-day space for understanding work as a liberated form of expression and a locus for technological innovation grounded in just and sustainable ways of life.
View
View
Prevalence of Counterfeit Fashion Goods in a West African Country. Crime in Perspective
Article
Full-text available
  • Jul 2024
  • DEVIANT BEHAV
This study explores the prevalence of counterfeit fashion goods in a West African country within the context of crime, focusing on the perspectives of producers, distributors, and consumers. Drawing on rational choice theory, the research aims to understand the awareness, motivations and decision-making processes behind the production, distribution, and consumption of counterfeit fashion items in Ghana. A qualitative research approach was adopted, utilizing snowball sampling to gather data from eight producers, eight distributors, and 25 consumers of counterfeit fashion goods. Semi-structured interviews were conducted to explore participants’ awareness, perceptions, rational decisions and behaviors regarding counterfeit fashion goods. The findings underscore the need for stronger regulatory frameworks and public education initiatives to combat the proliferation of counterfeit fashion goods in Ghana, West Africa. Recommendations include implementing stricter laws and enforcement mechanisms, enhancing consumer awareness about the risks associated with counterfeit products, and fostering collaboration among stakeholders to address this complex issue effectively.
View
View
View
Deep convolutional neural network based medical image classification for disease diagnosis
Article
Full-text available
  • Dec 2019
Abstract Medical image classification plays an essential role in clinical treatment and teaching tasks. However, the traditional method has reached its ceiling on performance. Moreover, by using them, much time and effort need to be spent on extracting and selecting classification features. The deep neural network is an emerging machine learning method that has proven its potential for different classification tasks. Notably, the convolutional neural network dominates with the best results on varying image classification tasks. However, medical image datasets are hard to collect because it needs a lot of professional expertise to label them. Therefore, this paper researches how to apply the convolutional neural network (CNN) based algorithm on a chest X-ray dataset to classify pneumonia. Three techniques are evaluated through experiments. These are linear support vector machine classifier with local rotation and orientation free features, transfer learning on two convolutional neural network models: Visual Geometry Group i.e., VGG16 and InceptionV3, and a capsule network training from scratch. Data augmentation is a data preprocessing method applied to all three methods. The results of the experiments show that data augmentation generally is an effective way for all three algorithms to improve performance. Also, Transfer learning is a more useful classification method on a small dataset compared to a support vector machine with oriented fast and rotated binary (ORB) robust independent elementary features and capsule network. In transfer learning, retraining specific features on a new target dataset is essential to improve performance. And, the second important factor is a proper network complexity that matches the scale of the dataset.
View
Automation for the artisanal economy: enhancing the economic and environmental sustainability of crafting professions with human–machine collaboration
Article
Full-text available
  • Sep 2020
  • AI Soc
Artificial intelligence (AI) is poised to eliminate millions of jobs, from finance to truck driving. But artisanal products (e.g., handmade textiles) are valued precisely because of their human origins, and thus have some inherent “immunity” from AI job loss. At the same time, artisanal labor, combined with technology, could potentially help to democratize the economy, allowing independent, small-scale businesses to flourish. Could AI, robotics and related automation technologies enhance the economic viability and environmental sustainability of these beloved crafting professions, perhaps even expanding their niche to replace some job loss in other sectors? In this paper, we compare the problems created by the current mass production economy and potential solutions from an artisanal economy. In doing so, the paper details the possibilities of utilizing AI to support hybrid forms of human–machine production at the microscale; localized and sustainable value chains at the mesoscale; and networks of these localized and sustainable producers at the macroscale. In short, a wide range of automation technologies are potentially available for facilitating and empowering an artisanal economy. Ultimately, it is our hope that this paper will facilitate a discussion on a future vision for more “generative” economic forms in which labor value, ecological value and social value can circulate without extraction or alienation.
View
Automation for the Artisanal Economy: Enhancing the Economic and Environmental Sustainability of Crafting Professions with Human-Machine Collaboration
Article
Full-text available
  • Aug 2019
Artifcial intelligence (AI) is poised to eliminate millions of jobs, from fnance to truck driving. But artisanal products (e.g., handmade textiles) are valued precisely because of their human origins, and thus have some inherent “immunity” from AI job loss. At the same time, artisanal labor, combined with technology, could potentially help to democratize the economy, allowing independent, small-scale businesses to fourish. Could AI, robotics and related automation technologies enhance the economic viability and environmental sustainability of these beloved crafting professions, perhaps even expanding their niche to replace some job loss in other sectors? In this paper, we compare the problems created by the current mass production economy and potential solutions from an artisanal economy. In doing so, the paper details the possibilities of utilizing AI to support hybrid forms of human–machine production at the microscale; localized and sustainable value chains at the mesoscale; and networks of these localized and sustainable producers at the macroscale. In short, a wide range of automation technologies are potentially available for facilitating and empowering an artisanal economy. Ultimately, it is our hope that this paper will facilitate a discussion on a future vision for more “generative” economic forms in which labor value, ecological value and social value can circulate without extraction or alienation.
View
A survey on Image Data Augmentation for Deep Learning
Article
Full-text available
  • Jul 2019
Abstract Deep convolutional neural networks have performed remarkably well on many Computer Vision tasks. However, these networks are heavily reliant on big data to avoid overfitting. Overfitting refers to the phenomenon when a network learns a function with very high variance such as to perfectly model the training data. Unfortunately, many application domains do not have access to big data, such as medical image analysis. This survey focuses on Data Augmentation, a data-space solution to the problem of limited data. Data Augmentation encompasses a suite of techniques that enhance the size and quality of training datasets such that better Deep Learning models can be built using them. The image augmentation algorithms discussed in this survey include geometric transformations, color space augmentations, kernel filters, mixing images, random erasing, feature space augmentation, adversarial training, generative adversarial networks, neural style transfer, and meta-learning. The application of augmentation methods based on GANs are heavily covered in this survey. In addition to augmentation techniques, this paper will briefly discuss other characteristics of Data Augmentation such as test-time augmentation, resolution impact, final dataset size, and curriculum learning. This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing Data Augmentation. Readers will understand how Data Augmentation can improve the performance of their models and expand limited datasets to take advantage of the capabilities of big data.
View
Causes and Effects of the Dwindled State of the Ghana Textile Industry
Article
Full-text available
  • May 2019
Ghana's once booming Textile industry has seen a downturn in recent times with multiplicity of adverse effects on the nation at large. Major research findings of the Ghana Textile Industry reveal poor operational and output performance with less than 4,000 workforce, average production output of 46 million yards per annum and foreign exchange and revenue generation of 3.173 million US dollars across the country. Whiles some stakeholders blame the government for lack of goodwill and policy direction for the industry, others argue that the industry operates on obsolete machinery and technology. This research identifies and examines the causes and effects of the dwindled state of the Ghana textile industry and makes projections toward revitalization of the textiles sub-sector. It targeted key players of the textile sub-sector from industry and academia. Purposively, Textiles Ghana Limited (TGL), Akosombo Textiles Limited (ATL), KNUST and Takoradi Polytechnic were selected for the study. However, stratified sampling technique was used in selecting members from the parent population for data collection through interview and questionnaire. The study combines qualitative and quantitative research approaches through Ex Post Facto analysis to arrive at the results. Stiff competition and trade liberalization were found as the major causes of the dwindled state of the industry. The rippling effects includes; high rate of redundancy, upsurge in graduate unemployment, low exports and high imports of textiles, decline in textiles education, etc. Restructuring, recapitalization, retooling and policy direction are critical projections recommended for the industry to build competitive edge against its offshore counterpart to survive.
View
No More FOMO: Limiting Social Media Decreases Loneliness and Depression
Article
Full-text available
  • Nov 2018
  • J SOC CLIN PSYCHOL
Introduction: Given the breadth of correlational research linking social media use to worse well-being, we undertook an experimental study to investigate the potential causal role that social media plays in this relationship. Method: After a week of baseline monitoring, 143 undergraduates at the University of Pennsylvania were randomly assigned to either limit Facebook, Instagram and Snapchat use to 10 minutes, per platform, per day, or to use social media as usual for three weeks. Results: The limited use group showed significant reductions in loneliness and depression over three weeks compared to the control group. Both groups showed significant decreases in anxiety and fear of missing out over baseline, suggesting a benefit of increased self-monitoring. Discussion: Our findings strongly suggest that limiting social media use to approximately 30 minutes per day may lead to significant improvement in well-being.
View
Classification Assessment Methods: a detailed tutorial
Article
Full-text available
  • Aug 2018
Classification techniques have been applied to many applications in various fields of sciences. There are several ways of evaluating classification algorithms. The analysis of such metrics and its significance must be interpreted correctly for evaluating different learning algorithms. Most of these measures are scalar metrics and some of them are graphical methods. This paper introduces a detailed overview of the classification assessment measures with the aim of providing the basics of these measures and to show how it works to serve as a comprehensive source for researchers who are interested in this field. This overview starts by highlighting the definition of the confusion matrix in binary and multi-class classification problems. Many classification measures are also explained in details, and the influence of balanced and imbalanced data on each metric is presented. An illustrative example is introduced to show (1) how to calculate these measures in binary and multi-class classification problems, and (2) the robustness of some measures against balanced and imbalanced data. Moreover, some graphical measures such as Receiver operating characteristics (ROC), Precision-Recall, and Detection error trade-off (DET) curves are presented with details. Additionally, in a step-by-step approach, different numerical examples are demonstrated to explain the preprocessing steps of plotting ROC, PR, and DET curves.
View
The Copyright Thing Doesn't Work Here: Adinkra and Kente Cloth and Intellectual Property in Ghana
Book
  • Apr 2011
In Ghana, adinkra and kente textiles derive their significance from their association with both Asante and Ghanaian cultural nationalism. Adinkra, made by stenciling patterns with black dye, and kente, a type of strip weaving, each convey, through color, style, and adornment, the bearer’s identity, social status, and even emotional state. Yet both textiles have been widely mass-produced outside Ghana, particularly in East Asia, without any compensation to the originators of the designs. This book focuses on the appropriation and protection of adinkra and kente cloth in order to examine the broader implications of the use of intellectual property law to preserve folklore and other traditional forms of knowledge. The book investigates the compatibility of indigenous practices of authorship and ownership with those established under intellectual property law, considering the ways in which both are responses to the changing social and historical conditions of decolonization and globalization. Comparing textiles to the more secure copyright protection that Ghanaian musicians enjoy under Ghanaian copyright law, it demonstrates that different forms of social, cultural, and legal capital are treated differently under intellectual property law. The book then moves beyond Africa, expanding the author’s analysis to the influence of cultural nationalism among the diaspora, particularly in the United States, on the appropriation of Ghanaian and other African cultures for global markets.
View
A Survey of Outlier Detection Methodologies
Article
  • Oct 2004
Outlier detection has been used for centuries to detect and, where appropriate, remove anomalous observations from data. Outliers arise due to mechanical faults, changes in system behaviour, fraudulent behaviour, human error, instrument error or simply through natural deviations in populations. Their detection can identify system faults and fraud before they escalate with potentially catastrophic consequences. It can identify errors and remove their contaminating effect on the data set and as such to purify the data for processing. The original outlier detection methods were arbitrary but now, principled and systematic techniques are used, drawn from the full gamut of Computer Science and Statistics. In this paper, we introduce a survey of contemporary techniques for outlier detection. We identify their respective motivations and distinguish their advantages and disadvantages in a comparative review.
View
View