Facial Privacy Preservation using FGSM and Universal Perturbation attacks

<p>Recent research has established the possibility of deducing soft-biometric attributes such as age, gender and race from an individual’s face image with high accuracy. Many techniques have been proposed to ensure user privacy, such as visible distortions to the images, manipulation of the original image with new face attributes, face swapping etc. Though these techniques achieve the goal of user privacy by fooling face recognition models, they don’t help the user when they want to upload original images without visible distortions or manipulation. The objective of this work is to implement techniques to ensure the privacy of user’s sensitive or personal data in face images by creating minimum pixel level distortions using white-box and black-box perturbation algorithms to fool AI models while maintaining the integrity of the image, so as to appear the same to a human eye.</p><div><br></div>

Facial Privacy Preservation using FGSM and Universal Perturbation attacks

<p>Recent research has established the possibility of deducing soft-biometric attributes such as age, gender and race from an individual’s face image with high accuracy. Many techniques have been proposed to ensure user privacy, such as visible distortions to the images, manipulation of the original image with new face attributes, face swapping etc. Though these techniques achieve the goal of user privacy by fooling face recognition models, they don’t help the user when they want to upload original images without visible distortions or manipulation. The objective of this work is to implement techniques to ensure the privacy of user’s sensitive or personal data in face images by creating minimum pixel level distortions using white-box and black-box perturbation algorithms to fool AI models while maintaining the integrity of the image, so as to appear the same to a human eye.</p><div><br></div>

A novel approach for detection and classification of re-entrant crack using modified CNNetwork

Purpose In the purpose of the section, the cracks that are in the construction domain may be common and usually fixed with the human inspection which is at the visible range, but for the cracks which may exist at the distant place for the human eye in the same building but can be captured with the camera. If the crack size is quite big can be visible but few cracks will be present due to the flaws in the construction of walls which needs authentic information and confirmation about it for the successful completion of the wall cracks, as these cracks in the wall will result in the structure collapse. Design/methodology/approach In the modern era of digital image processing, it has captured the importance in all the domain of engineering and all the fields irrespective of the division of the engineering, hence, in this research study an attempt is made to deal with the wall cracks which are found or searched during the building inspection process, in the present context in association with the unique U-net architecture is used with convolutional neural network method. Findings In the construction domain, the cracks may be common and usually fixed with the human inspection which is at the visible range, but for the cracks which may exist at the distant place for the human eye in the same building but can be captured with the camera. If the crack size is quite big can be visible but few cracks will be present due to the flaws in the construction of walls which needs authentic information and confirmation about it for the successful completion of the wall cracks, as these cracks in the wall will result in the structure collapse. Hence, for the modeling of the proposed system, it is considered with the image database from the Mendeley portal for the analysis. With the experimental analysis, it is noted and observed that the proposed system was able to detect the wall cracks, search the flat surface by the result of no cracks found and it is successful in dealing with the two phases of operation, namely, classification and segmentation with the deep learning technique. In contrast to other conventional methodologies, the proposed methodology produces excellent performance results. Originality/value The originality of the paper is to find the portion of the cracks on the walls using deep learning architecture.

Pandemic, Therapy, and a Meditation on How: A Creative-Relational Inquiry

“A creative-relational inquiry,” writes Dan Harris, “might be so small as to be imperceptible to the human eye . . . [b]ut it might also simultaneously be so big that whole universes exist within, between, and through it.” In this brief article, I take up Harris’ claim, writing into questions of scope and pace at this pandemic time, like how the small and the slow of an Edinburgh lockdown flutters into writing, poetic and therapeutic life, while around, within and beyond there is the multiplicity of the small, vast and fast of a virus. The text mutates with one small word, “how.”

Evaluation of Amebicidal and Cysticidal Activities of Antifungal Drug Isavuconazonium Sulfate against Acanthamoeba T4 Strains

Acanthamoeba species of amebae are often associated with Acanthamoeba keratitis, a severe corneal infection. Isavuconazonium sulfate is an FDA-approved drug for the treatment of invasive aspergillosis and mucormycosis. This prodrug is metabolized into the active isavuconazole moiety. Isavuconazole was previously identified to have amebicidal and cysticidal activity against Acanthamoeba T4 strains, but the activity of its prodrug, isavuconazonium sulfate, against trophozoites and cysts remains unknown. Since it is not known if isavuconazonium can be metabolized into isavuconazole in the human eye, we evaluated the activities of isavuconazonium sulfate against trophozoites and cysts of three T4 genotype strains of Acanthamoeba. Isavuconazonium displayed amebicidal activity at nanomolar concentrations as low as 1.4 nM and prevented excystation of cysts at concentrations as low as 136 μM. We also investigated the cysticidal activity of isavuconazonium sulfate in combination with a currently used amebicidal drug polyhexamethylene biguanide (PHMB). Although combination of isavuconazonium with PHMB did not elicit an obvious synergistic cysticidal activity, the combination did not cause an antagonistic effect on the cysts of Acanthamoeba T4 strains. Collectively, these findings suggest isavuconazonium retains potency against Acanthamoeba T4 strains and could be adapted for Acanthamoeba keratitis treatment.

Study on Aqueous Humour Hydrodynamics of Glaucoma Condition Using 3D Printed Model and Particle Image Velocimetry (PIV)

This study aims to explore the knowledge on fluid flow properties of the aqueous humour (AH), specifically on the anterior segment (AS) of the human eye for a medical condition called Glaucoma. The research objectives are to study on fluid flow characteristics of velocity and pressure of the AH on the AS of the eye using enlarged 3D printed model and computational analysis, and also to analyse the suitability of the 3D generated anterior AS in fluid flow analysis application on particle image velocimetry (PIV). Using polyvinyl alcohol (PVA) as a water-soluble 3D printing filament, a 3D model of the AS of the human eye was generated with SolidWorks 2018 and printed using Creality Ender 3. This printed model serves as the pattern for silicon rubber mould production using vacuum casting process. Analysis of AH flow hydrodynamics are conducted with computational analysis using ANSYS Workbench 19.2. Key findings support that use of PVA material suite the creation of specific shapes and patterns for 3D modelling applications alike, and silicon rubber moulding creates a non-reactive and long-lasting mould for PIV applications. Computational analysis findings support the use of the generated model for PIV applications. Overall, the study successfully supports the use of 3D printed model for PIV application and future work that can be induced include direct experimentation of the mould with PIV.