Quality of service (QoS): measurements of image formats in social cloud computing
- PDF / 2,607,185 Bytes
- 26 Pages / 439.37 x 666.142 pts Page_size
- 22 Downloads / 183 Views
Quality of service (QoS): measurements of image formats in social cloud computing Sajida Karim 1 & Hui He 1 & Asif Ali Laghari 1 & Arif Hussain Magsi 2 & Rashid Ali Laghari 1 Received: 26 June 2019 / Revised: 12 August 2020 / Accepted: 22 September 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
With the development of multimedia and the popularity of social cloud (SC), Quality of Service (QoS) is used to measure the technical parameters of service delivery. QoS is mostly used for measuring various kinds of multimedia data. Sharing (uploading) on SC has become the daily activity of end-users. As a result of this activity, it provides an open challenge for service providers. As a service provider, the host delivers productive infrastructure, allowing end-users to upload and share their high-quality images. To evaluate the QoS of image compression, we conduct objective QoS techniques to measure the efficiency of image quality and QoS performance of service delivery via uploading /downloading of images on three popular SC, (Facebook, Twitter, and Instagram). These social clouds and image services are compared by the mean, standard deviation (STDEV),mean square error (MSE), and signal-to-noise ratio (SNR). So we can achieve better results about the SC image quality. The results show that Instagram and Facebook compressed images more as compared to Twitter. However, Twitter less supports image formats and provides an acceptable quality of compressed images as compared to others. Therefore, Facebook supports all image file formats and enhanced (QoE) levels of end-users, but Twitter provided the best QoS of compressed images as compared to Instagram. Further, we found that the decrease of MSE and increase of SNR both have a high impact on image resolutions as compared to original image parameters, which has a higher effect on quality. For the specific quality assurance of image tasks acquired for robust automated image analysis research, we are also using deep learning techniques (DLT) to classify the quality of images. We also have identified the opportunity to simplify the Convolutional Neural network (CNN) with smaller resampling grids and making this process more suitable for at least five thousand (5000) of the enormous datasets available in the future. Keywords Convolution neural network (CNN) . Deep learning (DL) . Quality of service (QoS) . Social cloud (SC) computing . Etc.
* Hui He [email protected] Extended author information available on the last page of the article
Multimedia Tools and Applications
1 Introduction Quality of service (QoS) is mainly used to measure various multimedia data, such as image, audio and video, toimprove web service [9]. Image sharing (uploading) on SC has become the daily activity of end-users [2, 47]. Many users take an interest in capturing, share, and, upload the bulk of images on SC simultaneously [45]. Due to this end-user activity, which raises an open challenge for service providers as SC hosts to provide a productive infrastructure, whi
Data Loading...
