Browsing by Author "Choupani, R."

Now showing 1 - 10 of 10

Citation - WoS: 0
Citation - Scopus: 3
A Robust Watermarking Scheme Over Quadrant Medical Image in Discrete Wavelet Transform Domain
(Institute of Electrical and Electronics Engineers Inc., 2018) Goker, O.; Nazli, N.; Erol, M.M.; Choupani, R.; Dogdu, E.; 21259
The diffusion of digital content is very fast in today's technology. The velocity of gathering data might cause unlawful distribution of content. The major problem in content authorization is the robustness of methodology. Since frangible methodologies result unauthorized content access quicker, more robust solutions are essential for copyright protection. Watermarking technology is considered as a robust solution for copyright protection and authentication. In watermarking, quality of the image is a challenge. Applying a watermark on a medical image might cause corruption in original image, which leads to misleading content. The proposed copyright protection mechanism includes Quadtree algorithm which finds a region of non-interest to apply watermark on medical image in Discrete Wavelet Domain to provide authentication of the content without altering region of interest. Furthermore, in this paper, the visual quality of watermark implemented medical images and sample values are also discussed with the experimental results as well. © 2018 IEEE.
Citation - Scopus: 1
Distributed Query Processing and Reasoning over Linked Big Data
(Springer Science and Business Media Deutschland GmbH, 2022) Mohammed, H.H.; Doğdu, E.; Choupani, R.; Zarbega, T.S.A.; 21259
The enormous amount of structured and unstructured data on the web and the need to extract and derive useful knowledge from this big data make Semantic Web and Big Data Technology explorations of paramount importance. Open semantic web data created using standard protocols (RDF, RDFS, OWL) consists of billions of records in the form of data collections called “linked data”. With the ever-increasing linked big data on the Web, it is imperative to process this data with powerful and scalable techniques in distributed processing environments such as MapReduce. There are several distributed RDF processing systems, including SemaGrow, FedX, SPLENDID, PigSPARQL, SHARD, SPARQLGX, that are developed over the years. However, there is a need for computational and qualitative comparison of the differences and similarities among these systems. In this paper, we extend a previous comparative analysis to a diverse study with respect to qualitative and quantitative analysis views, through an experimental approach for these distributed RDF systems. We examine each of the selected RDF query systems with respect to the implementation setup, system architecture, underlying framework, and data storage. We use two widely used RDF benchmark datasets, FedBench and LUBM. Furthermore, we evaluate and examine their performances in terms of query execution time, thus, analyzing how those different types of large-scale distributed query engines, support long-running queries over federated data sources and the query processing times for different queries. The results of the experiments in this study show that SemaGrow distributed system performs more efficiently compared to FedX and Splendid, even though in smaller queries the former performs slower. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Citation - Scopus: 0
Hand gesture recognition in variable length sequences
(2005) Choupani, R.; Tolun, M.R.; 1863
Using hand gestures in human computer interaction has been a major challenge during the recent years. Many of the hand gesture recognition systems however, have been based on the recognition of hand postures and estimating the related gesture which is restricted to a few numbers of possible movements. However when dealing with applications such as understanding sign languages which include a large number of classes, an automatic learning method based on matching a sequence of postures with the characterizing feature sequence of each class is necessary. An important characteristic of this method is that each sample sequence of a class may have a variable length and different position of the key features. In this paper a syntactic method has been proposed for classifying the input sequences. An algorithm foe extracting the grammar of the method during training stage is also given.
Citation - Scopus: 1
Hierarchical SNR Scalable Video Coding with Adaptive Quantization for Reduced Drift Error
(SciTePress, 2015) Choupani, R.; Wong, S.; Tolun, M.
In video coding, dependencies between frames are being exploited to achieve compression by only coding the differences. This dependency can potentially lead to decoding inaccuracies when there is a communication error, or a deliberate quality reduction due to reduced network or receiver capabilities. The dependency can start at the reference frame and progress through a chain of dependent frames within a group of pictures (GOP) resulting in the so-called drift error. Scalable video coding schemes should deal with such drift errors while maximizing the delivered video quality. In this paper, we present a multi-layer hierarchical structure for scalable video coding capable of reducing the drift error. Moreover, we propose an optimization to adaptively determine the quantization step size for the base and enhancement layers. In addition, we address the trade-off between the drift error and the coding efficiency. The improvements in terms of average PSNR values when one frame in a GOP is lost are 3.70(dB) when only the base layer is delivered, and 4.78(dB) when both the base and the enhancement layers are delivered. The improvements in presence of burst errors are 3.52(dB) when only the base layer is delivered, and 4.50(dB) when both base and enhancement layers are delivered. Copyright © 2015 SCITEPRESS - Science and Technology Publications All rights reserved.
Citation - Scopus: 3
Link Prediction in Knowledge Graphs with Numeric Triples Using Clustering
(Institute of Electrical and Electronics Engineers Inc., 2020) Bayrak, B.; Doğdu, Erdoğan; Choupani, R.; Dogdu, E.; Bilgisayar Mühendisliği
Knowledge graphs (KG) include large amounts of structured data in many different domains. Knowledge or information is captured by entities and relationships between them in KG. One of the open problems in knowledge graphs area is "link prediction", that is predicting new relationships or links between the given existing entities in KG. A recent approach in graph-based learning problems is "graph embedding", in which graphs are represented as low-dimensional vectors. Then, it is easier to make link predictions using these vector representations. We also use graph embedding for graph representations. A sub-problem of link prediction in KG is the link prediction in the presence of literal values, and specifically numeric values, on the receiving end of links. This is a harder problem because of the numeric literal values taking arbitrary values. For such entries link prediction models cannot work, because numeric entities are not embedded in the vector space. There are several studies in this area, but they are all complex approaches. In this study, we propose a novel approach for link prediction in KG in the presence of numerical values. To overcome the embedding problem of numeric values, we used a clustering approach for clustering these numerical values in a knowledge graph and then used the clusters for performing link prediction. Then we clustered the numerical values to enhance the prediction rates and evaluated our method on a part of Freebase knowledge graph, which includes entities, relations, and numerical literals. Test results show that a considerable increase in link prediction rate can be achieved in comparison to previous studies. © 2020 IEEE.
Citation - Scopus: 1
Lung Inflammatory Classification of Diseases using X-ray Images
(Institute of Electrical and Electronics Engineers Inc., 2021) Mohanned, H.H.; Sürücü, S.; Choupani, R.
Recently, studies in inflammatory diseases categorization become of interest in the research community, especially with the sudden outbreak of the Covid-19 virus. Transfer learning proved to be the state-of-the-art when it comes to image classification problems, or related tasks. These methods achieve good results in this type of applications. Lately, this pre-trained embedding became even popular due to X-ray related studies for early Covid-19 diagnosis. In this study, we investigate the X-ray image classification problem using the transfer learning method. We fine-tuned and trained our model using pre-trained models such as AlexNet, VGG16, DenseNet etc, and a baseline deep neural network. We then evaluated this model in terms of classification evaluation metrics. The study shows that DenseNet achieves high accuracy compared to the other pre-trained and baseline CNN models. © 2021 IEEE
Citation - Scopus: 0
Main issues in scalable video coding: A review
(2007) Choupani, R.; Tolun, M.R.
Video streaming over the Internet has gained popularity during the recent years which is mainly the result of the introduction of videoconferencing and videotele-phony. These in turn have made it possible to bring to life many applications such as transmitting video over the Internet and over telephone lines, surveillance and monitoring, telemedicine (medical consultation and diagnosis at a distance), and computer based training and education. The heterogeneous, dynamic and best-effort structure of the Internet however, can not guarantee any specific bandwidth for a connection. Many video coding standards have tried to deal with this problem by introducing the scalability feature as adapting video streams to the fluctuations in the available bandwidths. In this review, we have discussed the main technical features of more common scalable video coding techniques. The main problems of these methods and their applicability together with the available motion compensated video coding methods are discussed as well.
Citation - Scopus: 0
Optimized Multiple Description Coding for Temporal Video Scalability
(Springer Verlag, 2013) Choupani, R.; Wong, S.; Tolun, M.
The vast application of video streaming over the Internet requires video adaptation to the fluctuations of the available bandwidth, and the rendering capabilities of the receiver device. On the other hand, the available video coding standards are designed for optimum bit rate which makes them susceptible to packet losses. A combination of video adaptation methods and error resilient methods can make the video stream more robust against networking problems. In this paper, an optimization for combining scalable video coding with multiple description coding schemes have been proposed. Our proposed method is capable of creating balanced descriptions with optimum coding efficiency.
Citation - Scopus: 0
Recent challenges in video coding and streaming
(2006) Choupani, R.; Tolun, M.R.; 1863
Video streaming over the Internet has gained popularity during the recent years which is mainly the result of the introduction of video-conferencing and videotelephony. These in turn have made it possible to bring to life many applications such as transmitting video over the Internet and telephone lines, surveillance and monitoring, telemedicine (medical consultation and diagnosis at a distance), and computer based training and education. These applications need a large bandwidth which is not available in all cases. Many video encoding standards have been introduced to deal with video compression and transmission problems. In this study, we have discussed the main technical features of the most important video coding standards in a comparative approach. The appropriateness of these features is application and transmission environment dependant. Manipulating video stream features or video transcoding methods are discussed as well.
Citation - Scopus: 0
Weighted embedded zero tree for scalable video compression
(2008) Choupani, R.; Wong, S.; Tolun, M.R.
Video streaming over the Internet has gained popularity during recent years mainly due to the revival of videoconferencing and videotelephony applications and the proliferation of (video) content providers. However, the heterogeneous, dynamic, and best-effort nature of the Internet cannot always guarantee a certain bandwidth for an application utilizing the Internet. Scalability has been introduced to deal with such issues (up to a certain point) by intelligently separating any information stream into multiple streams. The reception of one, several, or all stream influences the perceived quality of the information as basic, improved, or best, respectively. In addition, wavelet-based scalability combined with representation methods such as embedded zero trees (EZWs) improves the decode-ability of the stream even when only the initial part of the streams have been received. In this paper, we propose a method to improve on the compression rate of the EZW for scalability purposes by reducing the number of levels used in the tree. Therefore, the proposed method should be able to deal more efficiently with the mentioned scalability issues in low bandwidth network. Initial experimental show that the first two layers of the generated EZW are about 22.6% more concise.