• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.2
• /
• pp.89-96
• /
• 2016

The High Efficiency Video Coding (MPEG-H HEVC/ITU-T H.265) is the newest video coding standard which has the quadtree-structured coding unit (CU). The quadtree-structure splits a CU adaptively, and its optimum CU depth can be determined by rate-distortion optimization. Such HEVC encoding requires very high computational complexity for CU depth decision. Motivated that the blob detection, which is a well-known algorithm in computer vision, detects keypoints in pictures and decision of CU depth needs to consider high frequency energy distribution, in this paper, we propose to utilize these keypoints for fast CU depth decision. Experimental results show that 20% encoding time can be saved with only slightly increasing BDBR by 0.45% on all intra case.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.4
• /
• pp.699-706
• /
• 2008

H.264/AVC is a new international standard for the compression of video images, in which a deblocking filter has been adopted to remoye blocking artifacts. This paper proposes an efficient architecture of deblocking filter in H.264/AVC. By making good use of data dependence between neighboring $4{\times}4$ blocks, the memory sire is reduced and the throughput of the deblocking filter processing is increased. The designed deblocking filter further enhances the parallelism by simultaneously executing horizontal and vertical filtering within a macroblock in pipeline method and adopting overlap between macroblocks. The implementation result shows that the proposed architecture enhances the performance of deblocking filter processing from 1.75 to 4.23 times than that of the conventional deblocking filter. Hence the Proposed architecture of deblocking filter is able to perform real-time deblocking in high-resolution($2048{\times}1024$) video applications.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.6
• /
• pp.7-14
• /
• 2019

The automatic monitoring and detection of crowd behavior in the surveillance videos has obtained significant attention in the field of computer vision due to its vast applications such as security, safety and protection of assets etc. Also, the field of crowd analysis is growing upwards in the research community. For this purpose, it is very necessary to detect and analyze the crowd behavior. In this paper, we proposed a deep learning-based method which detects abnormal activities in surveillance cameras installed in a smart city. A fine-tuned VGG-16 model is trained on publicly available benchmark crowd dataset and is tested on real-time streaming. The CCTV camera captures the video stream, when abnormal activity is detected, an alert is generated and is sent to the nearest police station to take immediate action before further loss. We experimentally have proven that the proposed method outperforms over the existing state-of-the-art techniques.

• Cross-Cultural Studies
• /
• v.51
• /
• pp.103-131
• /
• 2018

This study examines motivations for international students to study abroad at Korean universities. Employing qualitative and mixed methods, this study used grounded theory to analyse data obtained from student interviews, essays, digital storytelling videos, and student video representations to explicate the nature of study of six subjects. All subjects were enrolled in English Education courses during years 2014-2017. The researcher was the course instructor. Results from this study revealed that major codes that emerged from data analyses were those of economics, culture, language study, and personal development, corroborating with findings of most research literature regarding international students' motivations (OUSO, 2015). However, survey of professional literature and study data showed that motivational codes presented in the literature and this study, were discursive in nature in that each code was not only connected to all other codes, but also mutually co-constructive. As such, this study suggests that motivational codes found in study abroad literature were discursive in nature, resembling Bourdieu's (1991) theory of economic, social, and cultural capitals. Results of this study suggest that various motivations for studying abroad are subsumed under economic logic of expense and career development.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.3
• /
• pp.201-210
• /
• 2019

This paper proposes the AMCS(Agricultural Machine Control System that distinguishes farms using satellite photos or drone photos of farms and controls the self-driving and operation of farm drones and tractors. The AMCS consists of the LSM(Local Server Module) which separates farm boundaries from sensor data and video image of drones and tractors, reads remote control commands from the main server, and then delivers remote control commands within the management area through the link with drones and tractor sprinklers and the PSM that sets a path for drones and tractors to move from the farm to the farm and to handle work at low cost and high efficiency inside the farm. As a result of AMCS performance analysis proposed in this paper, the PSM showed a performance improvement of about 100% over Dijkstra algorithm when setting the path from external starting point to the farm and a higher working efficiency about 13% than the existing path when setting the path inside the farm. Therefore, the PSM can control tractors and drones more efficiently than conventional methods.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.20-25
• /
• 2021

The target-object classification method was implemented using a deep-learning-based detection model in real-time images. The object detection model was a deep-learning-based detection model that allowed extensive data collection and machine learning processes to classify similar target-objects. The recognition model was implemented by changing the processing structure of the detection model and combining developed the vision-processing module. To classify the target-objects, the identity and similarity were defined and applied to the detection model. The use of the recognition model in industry was also considered by verifying the effectiveness of the recognition model using the real-time images of an actual soccer game. The detection model and the newly constructed recognition model were compared and verified using real-time images. Furthermore, research was conducted to optimize the recognition model in a real-time environment.

• Journal of Broadcast Engineering
• /
• v.27 no.6
• /
• pp.914-922
• /
• 2022

In this paper, we construct a commercial 4K UHD(Ultra High Definition) streaming device that utilizes a 5G mobile communication network as a transport channel and conduct a streaming performance test. It uses RTP(Realtime Transport Protocol) which has transmission quality monitoring capability as a transmission protocol to apply adaptive streaming. In addition, it provides the function to adjust the encoding rate of the video signal so that encoding can be optimized for the change in the bandwidth of the transmission channel. Through the performance test, it is confirmed that the H.265 encoding rate for 4K UHD signal is 48.69Mbps, the average glass-to-glass delay time is 293.60ms, and the average time difference between video and audio for lip sync is 120ms. With the result of performance test, it is shown that the streaming device is applied to 4K UHD Streaming device over 5G mobile communication network.

• Journal of the Korean Society of Earth Science Education
• /
• v.15 no.2
• /
• pp.171-191
• /
• 2022

The purpose of this study is to explore inductively learning effects of virtual field trips(VFTs) programs developed by elementary school students under the theme of minerals and rocks, focusing on learning in virtual geological components. Ten students attending 'H' elementary school in the metropolitan area voluntarily participated. In order to develop a virtual field trips programs, pre-actual outdoor geological field trips were conducted and virtual field trips programs were developed. In this process, written data of students observing, all video recording and voice recording materials of the course in which students participated, VR development data, and post-interview data were collected. Data were inductively analyzed focusing on four areas(cognitive, psychological, geography, and technical components) of learning in virtual geological field trips. As a result, there were positive learning effects for students in four areas. This study revealed that the study participants were not just participants in virtual learning, but rather developed classes for virtual field trips programs, which had significant results in terms of authentic inquiry.

• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.251-257
• /
• 2023

In this paper, we propose a dual-structured self-attention method that improves the lack of regional features of the vision transformer's self-attention. Vision Transformers, which are more computationally efficient than convolutional neural networks in object classification, object segmentation, and video image recognition, lack the ability to extract regional features relatively. To solve this problem, many studies are conducted based on Windows or Shift Windows, but these methods weaken the advantages of self-attention-based transformers by increasing computational complexity using multiple levels of encoders. This paper proposes a dual-structure self-attention using self-attention and neighborhood network to improve locality inductive bias compared to the existing method. The neighborhood network for extracting local context information provides a much simpler computational complexity than the window structure. CIFAR-10 and CIFAR-100 were used to compare the performance of the proposed dual-structure self-attention transformer and the existing transformer, and the experiment showed improvements of 0.63% and 1.57% in Top-1 accuracy, respectively.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.145-147
• /
• 2020

As deep learning technology advances, there have been many attempts to improve video codecs such as High-Efficiency-Video-Coding (HEVC) using deep learning technology. One of the most researched approaches is improving filters inside codecs through image restoration researches. In this paper, we propose a method 01 replacing the sample adaptive offset (SAO) filtering with a deep neural network. The proposed method uses the deep neural network to find the optimal offset value. The proposed network consists of two subnetworks to find the offset value and its type of the signal, which can restore nonlinear and complex type of error. Experimental results show that the performance is better than the conventional HEVC in low delay P and random access mode.