scholarly journals On the Robustness of Face Recognition Algorithms Against Attacks and Bias

2020 ◽  
Vol 34 (09) ◽  
pp. 13583-13589
Author(s):  
Richa Singh ◽  
Akshay Agarwal ◽  
Maneet Singh ◽  
Shruti Nagpal ◽  
Mayank Vatsa

Face recognition algorithms have demonstrated very high recognition performance, suggesting suitability for real world applications. Despite the enhanced accuracies, robustness of these algorithms against attacks and bias has been challenged. This paper summarizes different ways in which the robustness of a face recognition algorithm is challenged, which can severely affect its intended working. Different types of attacks such as physical presentation attacks, disguise/makeup, digital adversarial attacks, and morphing/tampering using GANs have been discussed. We also present a discussion on the effect of bias on face recognition models and showcase that factors such as age and gender variations affect the performance of modern algorithms. The paper also presents the potential reasons for these challenges and some of the future research directions for increasing the robustness of face recognition models.

2014 ◽  
Vol 10 (2) ◽  
pp. 78-95 ◽  
Author(s):  
Karen Smith ◽  
Francis Mendez ◽  
Garry L. White

A model is developed and tested to explain the relationships among narcissism, privacy concern, vigilance, and exposure to risk on Facebook, with age and gender as controlling variables. Two important constructs are conceptualized and measured in this research. Facebook exposure is defined as the opportunity for privacy and security breaches on Facebook. Facebook vigilance is the extent to which consumers stay focused, attentive, and alert to potential security and privacy risks on Facebook by restricting who can access and post to their Facebook accounts. Data from a survey of 286 adult Facebook users in the U.S. support the hypothesized relationships in the model. Results suggest that narcissism is related to increased Facebook exposure and lower Facebook vigilance, despite greater stated concern for privacy and security. Furthermore, females and younger users have greater risk exposure compared to males and older users. Implications of the findings and future research directions are discussed.


2020 ◽  
Author(s):  
Rizwan Qureshi ◽  
Muhammad Uzair ◽  
Anam Zahra

Hyperspectral imaging systems are well established, for satellite, remote sensing and geosciences applications. Recently, the reduction in the cost of hyperspectral sensors and increase in the imaging speed has attracted computer vision scientists to apply hyperspectral imaging to ground based computer vision problems such as material classification, agriculture, chemistry and document image analysis. Hyperspectral imaging has also been explored for face recognition; to tackle the issues of pose and illumination variations by exploiting the richer spectral information of hyperspectral images. In this article, we present a detailed review on the potential of hyperspectral imaging for face recognition. We present hyperspectral image aquisition process and discuss key preprocessing challenges. We also discuss hyperspectral face recognition databases and techniques for feature extraction from the hyperspectral images. Potential future research directions are also highlighted


2021 ◽  
Vol 21 (5) ◽  
pp. 1513-1530
Author(s):  
Luana Lavagnoli Moreira ◽  
Mariana Madruga de Brito ◽  
Masato Kobiyama

Abstract. Despite the increasing body of research on flood vulnerability, a review of the methods used in the construction of vulnerability indices is still missing. Here, we address this gap by providing a state-of-art account on flood vulnerability indices, highlighting worldwide trends and future research directions. A total of 95 peer-reviewed articles published between 2002–2019 were systematically analyzed. An exponential rise in research effort is demonstrated, with 80 % of the articles being published since 2015. The majority of these studies (62.1 %) focused on the neighborhood followed by the city scale (14.7 %). Min–max normalization (30.5 %), equal weighting (24.2 %), and linear aggregation (80.0 %) were the most common methods. With regard to the indicators used, a focus was given to socioeconomic aspects (e.g., population density, illiteracy rate, and gender), whilst components associated with the citizen's coping and adaptive capacity were slightly covered. Gaps in current research include a lack of sensitivity and uncertainty analyses (present in only 9.5 % and 3.2 % of papers, respectively), inadequate or inexistent validation of the results (present in 13.7 % of the studies), lack of transparency regarding the rationale for weighting and indicator selection, and use of static approaches, disregarding temporal dynamics. We discuss the challenges associated with these findings for the assessment of flood vulnerability and provide a research agenda for attending to these gaps. Overall, we argue that future research should be more theoretically grounded while, at the same time, considering validation and the dynamic aspects of vulnerability.


2020 ◽  
Author(s):  
Rizwan Qureshi ◽  
Muhammad Uzair ◽  
Anam Zahra

Hyperspectral imaging systems are well established, for satellite, remote sensing and geosciences applications. Recently, the reduction in the cost of hyperspectral sensors and increase in the imaging speed has attracted computer vision scientists to apply hyperspectral imaging to ground based computer vision problems such as material classification, agriculture, chemistry and document image analysis. Hyperspectral imaging has also been explored for face recognition; to tackle the issues of pose and illumination variations by exploiting the richer spectral information of hyperspectral images. In this article, we present a detailed review on the potential of hyperspectral imaging for face recognition. We present hyperspectral image aquisition process and discuss key preprocessing challenges. We also discuss hyperspectral face recognition databases and techniques for feature extraction from the hyperspectral images. Potential future research directions are also highlighted


2020 ◽  
Vol 64 (1) ◽  
pp. 31-41
Author(s):  
Tameeka Hunter ◽  
Franco Dispenza ◽  
Mary Huffstead ◽  
Mackenzie Suttles ◽  
Zachary Bradley

This study examined the resilience experiences of 13 sexual and gender minority persons living with disabilities, using phenomenological qualitative methodology. Researchers used in-depth, semi-structured interviews and various trustworthiness standards in the study. Participants endorsed four common resilience maximizers—self-acceptance, advocacy, social support, and a plea for recognition of humanity (an aspirational resilience maximizer)—and three common resilience minimizers—fragmentation, identity concealment, and punishment. Counseling implications and future research directions are also discussed.


2019 ◽  
Vol 22 (2) ◽  
pp. 416-442 ◽  
Author(s):  
Selman Karagoz ◽  
Nezir Aydin ◽  
Vladimir Simic

AbstractWaste management is gaining very high importance in recent years. As automotive is one of the most critical sectors worldwide, which is rapidly increasing, the management of end-of-life vehicles (ELVs) gains importance day by day. Due to legislation and new regulations, actors like users, producers, and treatment facilities are being conferred new responsibilities in the ELV management process. Besides, the ELV management is of vital importance for environment conservation, circular economy and sustainable development. All of these reasons are making the ELV management such a crucial issue to study. Today, the ELV management is a well-positioned and emergent research area. However, the available review papers are focused only on a small area of the ELV management, such as reverse logistics, recovery infrastructure, disassemblability, etc. Besides, a review of state-of-the-art mathematical models for the ELV management is still missing. This paper aims to provide an extensive content analysis overview of studies on the ELV management. A total of 232 studies published in the period 2000–2019 are collected, categorized, reviewed and analyzed. A critical review of the published literature is provided. Gaps in the literature are identified to clarify and suggest future research directions. This review can provide a source of references, valuable insights, and opportunities for researchers interested in the ELV management and inspire their additional attention.


Covid-19 taught us the importance of personalized ICT use in the higher education context. In this scenario, the importance of researching student's ICT behaviour is becoming ever more crucial. This study investigates the influence of student alienation (SAL), socio-economic status, residential background, type of course, and gender on students' ICT use behaviour. 704 Kashmiri university students responded to an offline survey comprising two scales: Students ICT use scale and student alienation scale. The results showed that SAL has a negative relationship with student’s ICT use for education and capital enhancement. Students differed in their ICT behaviour based on gender, type of course, and residential background. Socio-economic status was positively correlated with ICT use for education and entertainment. These findings highlight the nuances of ICT use behaviour among young university students. The implications and future research directions have been discussed.


Sign in / Sign up

Export Citation Format

Share Document