International Journal of Computer Networks and Communications Security

Volume 4, Issue 8, August 2015

 

 

 

Error Concealment by Tracking and Velocity Estimation of Macro-Block in Video Communication at the Decoder of H.264/AVC
Pages: 199-203 (5) | [Full Text] PDF (572 KB)
MA Marjan, NB Sadia, MD Haque, MP Uddin, MI Afjal, MR Islam
Telecommunication & Electronics Engineering, Hajee Mohammad Danesh Science and Technology University (HSTU), Dinajpur-5200, Bangladesh
Computer Science & Information Technology, Hajee Mohammad Danesh Science and Technology University (HSTU), Dinajpur-5200, Bangladesh
Computer Engineering, Hajee Mohammad Danesh Science and Technology University (HSTU), Dinajpur-5200, Bangladesh

Abstract - Nowadays it is an important concern passing large amount of data over the wireless or wired media using comparatively low bandwidth. Video is one of them. To reduce data size of video, many compression techniques is used like H.261, H.263, H.26L, H.264/AVC, MPEG-2 and MPEG-4 etc. When compressed data pass throw comparatively low bandwidth media or for the characteristics of media there may be some data loss in the transmission line. This loss affects the original video and it occur some error. In this paper introduce a new technique for error concealment by post processing using an artificial intelligence at the receiver of H.264/AVC decoder. The artificial intelligence of the decoder based on image/video characteristics and human visual system properties, previous video frame of current error affected frame and frame rate, velocity of the macroblock and the physical relationship of the parameter, position and velocity of an object and time. In this methodology; at normal situation the system learn about two parameter which is the velocity and the rate of change of intensity of macroblock .When error occur, this technique gives an optimum solution for error concealment of error affected frame in the video communication. For live video streaming this methodology gives better performance for smooth object moving videos like video conference, video lecture, stage performance etc and it also give better performance for all downloaded video.
 
Index Terms - Video Error Concealment, Artificial Intelligence, Velocity of Macroblock, Estimation Macroblock Position, Video Compression, Video Communication

Citation - MA Marjan, NB Sadia, MD Haque, MP Uddin, MI Afjal, MR Islam. "Error Concealment by Tracking and Velocity Estimation of Macro-Block in Video Communication at the Decoder of H.264/AVC." International Journal of Computer Science and Software Engineering 4, no. 8 (2015): 199-203.
 

Towards Protocols for Vehicular Ad Hoc Networks (VANETs)
Pages: 204-208 (5) | [Full Text] PDF (245 KB)
J Afzal
Global University Defense Housing Authority Campus, Lahore 54000, Pakistan

Abstract - Vehicular Ad Hoc Networks are becoming popular all over the world. Like Vehicle to Vehicle and Vehicle to Infrastructure is major types of communications in VANETs. The design of routing protocols is one of the important research areas that deal with the problems of frequent topology change and quick movements of vehicles. One of the major challenges of VANETs is routing the packets in effective and efficient manner. In this paper we will studies different security aspects and communication protocols.
 
Index Terms - Video Error Concealment, Artificial Intelligence, Velocity of Macroblock, Estimation Macroblock Position, Video Compression, Video Communication

Citation - J Afzal . "Towards Protocols for Vehicular Ad Hoc Networks (VANETs)." International Journal of Computer Science and Software Engineering 4, no. 8 (2015): 204-208.
 

Features Extraction for Pattern Recognition Based on Local ZERNIKE Moments
Pages: 209-217 (9) | [Full Text] PDF (693 KB)
MH Saad, HI Saleh
Radiation Engineering Department, NCRRT, AEA, P. O. Box. 29, 8th District, Nasr City, Cairo, Egypt

Abstract - Pattern recognition is one of the most common problems faced in scientific disciplines and engineering, which contains developing prediction classification models from historic data , training samples or Content-based image retrieval (CBIR). This paper proposes a new feature extraction technique for pattern recognition based on local Zernike moments. Moreover, the proposed technique segments an image to the most salient regions in order to provide efficient segmentation and treats problems occurred in earlier region-based segmentation algorithms. This technique extracts interest salient regions then calculates local Zernike moment for each region which works as local descriptors to be robust against scaling, rotation, illumination changes and background clutter. A spatial graph is constructed from the descripted salient regions then, the final image rank is calculated using greedy graph matching algorithm. The proposed technique is tested on Brodatz, Caltech101 and Wang databases. The results show that proposed technique is appropriate for accurately retrieving images than previous techniques even in distortion cases such as noise and geometric deformations.
 
Index Terms - Local Zernike Moments, Graph Matching, Pattern Recognition, CBIR, YCbCr Quantization

Citation - MH Saad, HI Saleh. "Features Extraction for Pattern Recognition Based on Local ZERNIKE Moments." International Journal of Computer Science and Software Engineering 4, no. 8 (2015): 209-217.
 

Micro-analytics for Student Performance Prediction
Pages: 218-223 (6) | [Full Text] PDF (416 KB)
D Azcona, K Casey
School of Computing, Dublin City University, Glasnevin, Dublin 9, Ireland

Abstract - Analytics have a major role in the future of higher education. Virtual Learning Environments are an excellent source of information for enhancing the learning process. The availability of real time insight into the performance of learners can be a significant help for educators in their planning of teaching activities. For students, getting feedback about their progress can be both motivating and encouraging.
 
Index Terms - Learning Analytics; Data Mining; Virtual Learning Environments; Student Behaviour; Early Intervention

Citation - D Azcona, K Casey. "Micro-analytics for Student Performance Prediction." International Journal of Computer Science and Software Engineering 4, no. 8 (2015): 218-223.