RSS 1.0Welcome to ChesterRep - the University of Chester's Online Research Repository
ChesterRep is the University of Chester's institutional repository and an online platform designed to collate, store, and aid discoverability of research carried out at the university to the wider research community
For more information about how to submit material to ChesterRep, see our ChesterRep guides here. You can also find out more about our editorial and open access policies here. Please note that you must be a member of the University of Chester in order to view these pages.
Communities in DSpace
Select a community to browse its collections.
Recently Added
-
When My Work is Found Wanting: Power, intersectionality, postcolonialism, and the reflexive feminist researcherFeminist research emerges out of a struggle with power. Ingrained in feminist studies of religion is the identification and dismantling of religious hierarchies and structures that disempower. Feminist scholarship has contended with the essentialist categories of ‘woman’ and ‘women’s experience’ without questioning that its rendering of ‘religion’ and ‘gender’ was premised on and benefited from its own modes of dominance and suppression, conditioned by Western colonialism. Taking up feminist research is a reflexive position that can assist in upsetting the established hierarchies of power and the binary oppositions of researcher and researched, knower and known, political and personal. However, feminist thinking in religion and gender, like the author own, has not always been reflexively attentive to its almost exclusive focus on the relationships between religion and gender and its own power as the product of Western, colonial, secular discourses.
-
Deep Learning based Human Detection in Privacy-Preserved Surveillance VideosVisual surveillance systems have been improving rapidly over the recent past, becoming more capable and pervasive with incorporation of artificial intelligence. At the same time such surveillance systems are exposing the public to new privacy and security threats. There have been an increasing number of reports of blatant abuse of surveillance technologies. To counteract this, data privacy regulations (e.g. GDPR in Europe) have provided guidelines for data collection and data processing. However, there is still a need for a private and secure method of model training for advanced machine learning and deep learning algorithms. To this end, in this paper we propose a privacy-preserved method for visual surveillance. We first develop a dataset of privacy preserved videos. The data in these videos is masked using Gaussian Mixture Model (GMM) and selective encryption. We then train high-performance object detection models on the generated dataset. The proposed method utilizes state-of-art object detection deep learning models (viz. YOLOv4 and YOLOv5) to perform human/object detection in masked videos. The results are encouraging, and are pointers to the viability of the use of modern day deep learning models for object detection in privacy-preserved videos.
-
A Novel Double-Threshold Neural Classifier for Non-Linearly Separable ApplicationsClassification of data finds applications in various engineering and scientific problems. When real-time operation is desired, hardware solutions tend to be more amenable as compared to algorithmic/heuristic solutions. This paper presents a novel current-mode dual-threshold neuron designed and implemented at 32nm CMOS technology node. Subsequently, a current-mode double-threshold classifier is presented which is capable of classifying input patterns of non-linearly separable problems. Thereafter, application of the current-mode dual-threshold neuron in the realization of the XOR function using only a single neural unit is discussed. The proposed neuron as well as both the applications discussed are capable of operating from sub-1V power supplies. Computer simulations using HSPICE yield promising results with the values of delay and power consumption estimated to be lower than existing circuits.
-
Wasserstein GAN based Chest X-Ray Dataset Augmentation for Deep Learning Models: COVID-19 Detection Use-CaseThe novel coronavirus infection (COVID-19) is still continuing to be a concern for the entire globe. Since early detection of COVID-19 is of particular importance, there have been multiple research efforts to supplement the current standard RT-PCR tests. Several deep learning models, with varying effectiveness, using Chest X-Ray images for such diagnosis have also been proposed. While some of the models are quite promising, there still remains a dearth of training data for such deep learning models. The present paper attempts to provide a viable solution to the problem of data deficiency in COVID-19 CXR images. We show that the use of a Wasserstein Generative Adversarial Network (WGAN) could lead to an effective and lightweight solution. It is demonstrated that the WGAN generated images are at par with the original images using inference tests on an already proposed COVID-19 detection model.
