• Journal of the Korean Society of Industry Convergence
• /
• v.3 no.2
• /
• pp.167-178
• /
• 2000

Software RAID(Redundant Arrays of Inexpensive Disks) is defined as a storage system that provides capabilities of hardware RAID, and guarantees high reliability as well as high performance. In this paper, we propose an enhanced disk scheduling algorithm and a scheme to guarantee reliability of data. We also design and implement software RAID by utilizing these mechanism to develop a storage system for multimedia applications. Because the proposed algorithm improves a defect of traditional GSS algorithm that disk I/O requests arc served in a fixed order, it minimizes buffer consumption and reduces the number of deadline miss through service group exchange. Software RAID also alleviates data copy overhead during disk services by sharing kernel memory. Even though the implemented software RAID uses the parity approach to guarantee reliability of data, it adopts different data allocation scheme. Therefore, we reduce disk accesses in logical XOR operations to compute the new parity data on all write operations. In the performance evaluation experiments, we found that if we apply the proposed schemes to implement the Software RAID, it can be used as a storage system for small-sized video-on-demand servers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1684-1699
• /
• 2017

The deblocking filter (DF) reduces blocking artifacts in encoded video sequences, and thereby significantly improves the subjective and objective quality of videos. Statistics show that the DF accounts for 5-18% of the total decoding time in high-efficiency video coding. Therefore, speeding up the DF will improve codec performance, especially for the decoder. In view of the rapid development of multicore technology, we propose a parallel DF scheme based on a modified order of accessing the coding tree units (CTUs) by analyzing the data dependencies between adjacent CTUs. This enables the DF to run in parallel, providing accelerated performance and more flexibility in the degree of parallelism, as well as finer parallel granularity. We additionally solve the problems of variable privatization and thread synchronization in the parallelization of the DF. Finally, the DF module is parallelized based on the HM16.1 reference software using OpenMP technology. The acceleration performance is experimentally tested under various numbers of cores, and the results show that the proposed scheme is very effective at speeding up the DF.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.6C
• /
• pp.368-375
• /
• 2011

In the distributed video coding system, the parity bits, which are generated in encoders, are used to reconstruct Wyner-Ziv frames. Since the original Wyner-Ziv frames are not known in decoders, the efficient correlation noise modeling for turbo or LDPC code is necessary. In this paper, an efficient correlation noise modeling method is proposed and the performance is analyzed. The method to estimate the quantization parameters for key frames, which are encoded using H.264 intraframe coding technique, is also proposed. The performance of the proposed system is evaluated by extensive computer simulations.

• International Journal of Internet, Broadcasting and Communication
• /
• v.11 no.4
• /
• pp.76-85
• /
• 2019

We introduce a vision-based object detection method for real-time video surveillance system in low-end edge computing environments. Recently, the accuracy of object detection has been improved due to the performance of approaches based on deep learning algorithm such as Region Convolutional Neural Network(R-CNN) which has two stage for inferencing. On the other hand, one stage detection algorithms such as single-shot detection (SSD) and you only look once (YOLO) have been developed at the expense of some accuracy and can be used for real-time systems. However, high-performance hardware such as General-Purpose computing on Graphics Processing Unit(GPGPU) is required to still achieve excellent object detection performance and speed. To address hardware requirement that is burdensome to low-end edge computing environments, We propose sub-frame analysis method for the object detection. In specific, We divide a whole image frame into smaller ones then inference them on Convolutional Neural Network (CNN) based image detection network, which is much faster than conventional network designed forfull frame image. We reduced its computationalrequirementsignificantly without losing throughput and object detection accuracy with the proposed method.

• Journal of Information Processing Systems
• /
• v.18 no.5
• /
• pp.677-687
• /
• 2022

We propose a detection algorithm based on tree-structured convolutional neural networks (TsCNNs) that finds pornography, propaganda, or other inappropriate content on a social media network. The algorithm sequentially applies the typical convolutional neural network (CNN) algorithm in a tree-like structure to minimize classification errors in similar classes, and thus improves accuracy. We implemented the detection system and conducted experiments on a data set comprised of 6 ordinary classes and 11 inappropriate classes collected from the Korean military social network. Each model of the proposed algorithm was trained, and the performance was then evaluated according to the images and videos identified. Experimental results with 20,005 new images showed that the overall accuracy in image identification achieved a high-performance level of 99.51%, and the effectiveness of the algorithm reduced identification errors by the typical CNN algorithm by 64.87 %. By reducing false alarms in video identification from the domain, the TsCNNs achieved optimal performance of 98.11% when using 10 minutes frame-sampling intervals. This indicates that classification through proper sampling contributes to the reduction of computational burden and false alarms.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.5
• /
• pp.252-259
• /
• 2007

With the rapid development of multimedia applications and wireless communication networks, consumer demand for video-over-wireless capability on mobile computing systems is growing rapidly. In this regard, this paper introduces YUV-aware instructions that enhance the performance and efficiency in the processing of color image and video. Traditional multimedia extensions (e.g., MMX, SSE, VIS, and AltiVec) depend solely on generic subword parallelism whereas the proposed YUV-aware instructions support parallel operations on two-packed 16-bit YUV (6-bit Y, 5-bits U, V) values in a 32-bit datapath architecture, providing greater concurrency and efficiency for color image and video processing. Moreover, the ability to reduce data format size reduces system cost. Experiment results on a representative dynamically scheduled embedded superscalar processor show that YUV-aware instructions achieve an average speedup of 3.9x over the baseline superscalar performance. This is in contrast to MMX (a representative Intel#s multimedia extension), which achieves a speedup of only 2.1x over the same baseline superscalar processor. In addition, YUV-aware instructions outperform MMX instructions in energy reduction (75.8% reduction with YUV-aware instructions, but only 54.8% reduction with MMX instructions over the baseline).

• Journal of Korea Multimedia Society
• /
• v.7 no.3
• /
• pp.357-367
• /
• 2004

In VOD systems the number of channels available to a video server is limited since the number of channels, the video server can support, is determined by the communication bandwidth of the video server. Multicast techniques such as batching, patching and batch-patching have been proposed to reduce I/O demand on the video server by sharing multicast data. In this paper, we propose a channel reservation batch-patching technique that first applies the batching technique to hot video requests and then applies the patching technique to the batches of the hot video requests. And the proposed technique reserves a part of the channel capacity of the video server for many hot video requests, so that the hot video requests can be served without defection. The performance of the proposed technique is compared with those of Patching and Batch-Patching techniques in terms of average service latency, defection rate, fairness and frame reduction rate by simulations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.4B
• /
• pp.349-359
• /
• 2009

In this paper, we propose a stereo-video transmission method for reduction of delay and maximization of 3D effect. Conventional multimedia synchronization algorithms were designed to achieve minimum delay and synchronize multiple video and audio streams, however, they could not be effective for 3D video transmission. In this paper, we proposed a synchronization algorithm by considering the minimum error of time difference between streams for 3D effect. The minimum error of time difference for 3D effect was derived based on a 3D subjective quality test. We compute display time of the delivered videos within the allowed time-difference and the video are displayed according to the display time. To evaluate the performance of the proposed algorithm, we implemented a real-time video communication system and subjective quality test has been conducted with the proposed system. We found that video quality displayed by the proposed system. We found that video quality displayed by the proposed algorithm ranks 'good' and 'excellent' in the DMOS (Differential Mean Opinion Score) scale, based on the MOS (Mean Opinion Score) test.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.4
• /
• pp.31-43
• /
• 1999

The main goal of this paper is to design a high performance SVP(Stack based Video Processor) for network applications. The SVP is a comprehensive scheme; 'better' in the sense that it is an optimal selection of previously proposed enhancements of a stack machine and a video processor. This can process effectively object-based video data using a S-RISC(Stack-based Reduced Instruction Set Computer) with a semi -general-purpose architecture having a stack buffer for OOP(Object-Oriented Programming) with many small procedures at running programs. And it includes a vector processor that can improve the MPEG coding speed. The vector processor in the SVP can execute advanced mode motion compensation, motion prediction by half pixel and SA-DCT(Shape Adaptive-Discrete Cosine Transform) of MPEG-4. Absolutors and halfers in the vector processor make this architecture extensive to a encoder. We also designed a VLSI stack-oriented video processor using the proposed architecture of stack-oriented video decoding. It was designed with O.5$\mu\textrm{m}$ 3LM standard-cell technology, and has 110K logic gates and 12 Kbits SRAM internal buffer. The operating frequency is 50MHz. This executes algorithms of video decoding for QCIF 15fps(frame per second), maximum rate of VLBV(Very Low Bitrate Video) in MPEG-4.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.4
• /
• pp.381-391
• /
• 2013

Wireless transmission of uncompressed video guarantees higher quality with lower latency than compressed video transmission. Although current wireless technologies cannot fully cover required data rates of about a few Gb/s for full high definition resolution, some wireless technologies such as ultra-wideband (UWB) provide 1 Gb/s data rate which is adequate for uncompressed video transmission in portable devices. In this paper, we propose an uncompressed video transmission system for wireless mirroring services in portable devices. We firstly simulated the performance of uncompressed video transmission using single or multiple 1 Gb/s UWB technology. Then we implemented hardware-based uncompressed video processing block and Gb/s wireless MAC accelerator. Finally, we show the implementation result and the demonstration of uncompressed HD video transmission using multiple 1 Gb/s UWB PHYs.