Future University In Egypt (FUE)
Future University is one of most promising private universities in Egypt. Through excellence in teaching, research and service, Future University strives to provide a comprehensive, high-quality education that prepares our graduates to be future leaders.
mainLogo
90th Street
New Cairo
Egypt
Faculty of Engineering & Technology
Apply Now

Nermin Mohamed Fawzy Mahmoud Salem

Basic information

Name : Nermin Mohamed Fawzy Mahmoud Salem
Title: Assistant Lecturer
Personal Info: Eng. Nermin M.Fawzy Salem is a lecturer assistant at the Faculty of Engineering Future University. Cairo, she received her masters' degree from Faculty of Engineering Ain Shams University and is currently teaching the courses of (Microprocessors, Microcontroller, Computer Organization and Digital Logic) for the students of electrical and mechanical engineering. A strong advocate for hands-on, inquiry-based learning, she involves her students in a variety of community service, problem-solving, and technology-infused activities that provide them with opportunities to use their knowledge to help others. Her professional interests focus on communicative approaches, thematic planning, and cooperative learning. She was recently honored From Future University for getting her masters' degree. View More...

Education

Certificate Major University Year
Masters Science Engineering Of Electrical Engineering Ain Shams University - Faculty Of Engineering 2013
Bachelor Electrical Engineering - Computer Engineering Ain Shams - Egypt 2006

Researches /Publications

Random-Shaped Image Inpainting using Dilated Convolution - 01/0

Nermin Mohamed Fawzy Mahmoud Salem

Abbas, Hazem M. Mahdi, Hani M. K.

01/08/2019

Over the past few years, Deep learning-based methods have shown encouraging and inspiring results for one of the most complex tasks of computer vision and image processing; Image Inpainting. The difficulty of image inpainting is derived from its’ need to fully and deeply understand of the structure and texture of images for producing accurate and visibly plausible results especially for the cases of inpainting a relatively larger region. Deep learning methods usually employ convolution neural network (CNN) for processing and analyzing images using filters that consider all image pixels as valid ones and usually use the mean value to substitute the missing pixels. This result in artifacts and blurry inpainted regions inconsistent with the rest of the image. In this paper, a new novel-based method is proposed for image inpainting of random-shaped missing regions with variable size and arbitrary locations across the image. We employed the use of dilated convolutions for composing multiscale context information without any loss in resolution as well as including a modification mask step after each convolution operation. The proposed method also includes a global discriminator that also considers the scale of patches as well as the whole image. The global discriminator is responsible for capturing local continuity of images texture as well as the overall global images’ features. The performance of the proposed method is evaluated using two datasets (Places2 and Paris Street View). Also, a comparison with the recent state-of-the-art is preformed to demonstrate and prove the effectiveness of our model in both qualitative and quantitative evaluations.

Download PDF
Random-Shaped Image Inpainting using Dilated Convolution - 01/0

Nermin Mohamed Fawzy Mahmoud Salem

Hani M. K. Mahdi, and Hazem M. Abbas

01/08/2019

Over the past few years, Deep learning-based methods have shown encouraging and inspiring results for one of the most complex tasks of computer vision and image processing; Image Inpainting. The difficulty of image inpainting is derived from its’ need to fully and deeply understand of the structure and texture of images for producing accurate and visibly plausible results especially for the cases of inpainting a relatively larger region. Deep learning methods usually employ convolution neural network (CNN) for processing and analyzing images using filters that consider all image pixels as valid ones and usually use the mean value to substitute the missing pixels. This result in artifacts and blurry inpainted regions inconsistent with the rest of the image. In this paper, a new novel-based method is proposed for image inpainting of random-shaped missing regions with variable size and arbitrary locations across the image. We employed the use of dilated convolutions for composing multiscale context information without any loss in resolution as well as including a modification mask step after each convolution operation. The proposed method also includes a global discriminator that also considers the scale of patches as well as the whole image. The global discriminator is responsible for capturing local continuity of images texture as well as the overall global images’ features. The performance of the proposed method is evaluated using two datasets (Places2 and Paris Street View). Also, a comparison with the recent state-of-the-art is preformed to demonstrate and prove the effectiveness of our model in both qualitative and quantitative evaluations.

Download PDF
Semantic Image Inpainting using Self-Learning Encoder-Decoder and Adversarial Loss - 01/1

Nermin Mohamed Fawzy Mahmoud Salem

Hani M. K. Mahdi and Hazem M. Abbas

01/12/2018

Images are exposed to deterioration over years due to many factors. These factors may include, but not limited to, environmental factors, chemical processing, improper storage, etc. Image inpainting has gained significant attention from researchers to recover the deteriorated parts in images. In this paper, two new techniques for image inpainting techniques using Deep Convolution Neural Networks (CNN) are proposed. In the first technique, a self-tuned Encoder-Decoder architecture based on a Fully Convolution Network (FCN) is used to generate different sized blocks from non-deteriorated image dataset with L2 being used as a loss measure. On the other hand, the second technique is a two-step technique inspired from Context Encoders. In the first step, Context Encoders are trained on non-deteriorated image dataset to select blocks from training images with minimum L2 loss. In the second step, the selected block is applied to Generative Adversarial Networks (GAN) in order to improve the quality of the recovered image. Several simulation examples were made to proof that the performance self-tuned Encoder-Decoder and GAN is the same. Simulations have also shown that the proposed methods have superior performance in recovering missing regions in deteriorated images over other state-of-art techniques. Paris Street View dataset was used for training and validation to validate our results.

Download PDF
A QoS based framework for efficient web servers - 01/0

Nermin Mohamed Fawzy Mahmoud Salem

01/01/2012

Abstract - The explosive growth in popularity of the World Wide Web is leading to a number of performance problems such as decreased QoS (Quality of Service) when measured in terms of completed sessions and perceived service time by users in case of overloaded servers. This paper contributes to research aiming to improve the QoS of commercial web servers. The QoS is defined in terms of server throughput and service time. The work is a two-sided approach for the problem. The first side is concerned with load balancing of web traffic, while the second tackles admission control for the server itself. The overall approach combines a locality aware solution (distributed workload request distribution policy) for the first problem with a non locality aware solution (predictive admission control) under CODA (Completely Distributed Architecture) for the second problem. The solution methodology for the first part allows all server nodes to participate in session dispatching and exploit at maximum the good features of existing centralized algorithms. That of the second part uses an adaptive time slot scheduling that sets the execution times of the algorithm depending on the burstiness arriving to the system, with the aim of reducing the computational cost of the algorithm. Using a simulation model, we could show that a web server augmented with traffic control and session-based admission control performs better in terms of QoS than a server without these controls. This means it offers a better throughput and is able to provide a fair guarantee of completion, for any accepted session, independent of the session length.

Download PDF

Follow us on

Visit the Faculty

ADS