• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.21 no.2
• /
• pp.99-105
• /
• 2007

In this paper, diskless personal multimedia player(PMP) that can access and decode the remote large multimedia data is implemented via wireless network. To implement this, WLAN based NFS protocol is used to connect PMP to the remote server and text image and movie files are decoded and played using ARM9 cored PXA255 embedded processor and Linux OS. The fuction and performance of the PMP is evaluated and verified using variuos types of contents. Linux kernel elements are configured and built in according to the hardware and software on the target board to install on the target board. The confirming result shows the required functions and performances.

• Journal of Korea Multimedia Society
• /
• v.20 no.5
• /
• pp.769-781
• /
• 2017

Deep learning networks like Convolutional Neural Networks (CNNs) show successful performances in many computer vision applications such as image classification, object detection, and so on. For implementation of deep learning networks in embedded system with limited processing power and memory, deep learning network may need to be simplified. However, simplified deep learning network cannot learn every possible scene. One realistic strategy for embedded deep learning network is to construct a simplified deep learning network model optimized for the scene images of the installation place. Then, automatic training will be necessitated for commercialization. In this paper, as an intermediate step toward automatic training under fisheye camera environments, we study more precise human localization in fisheye images, and propose an accurate human localization method, Automatic Ground-Truth Labelling Method (AGTLM). AGTLM first localizes candidate human object bounding boxes by utilizing GoogLeNet-LSTM approach, and after reassurance process by GoogLeNet-based CNN network, finally refines them more correctly and precisely(tightly) by applying saliency object detection technique. The performance improvement of the proposed human localization method, AGTLM with respect to accuracy and tightness is shown through several experiments.

• Journal of Korea Multimedia Society
• /
• v.25 no.4
• /
• pp.583-599
• /
• 2022

Although the object detection accuracy with a single image has been significantly improved with the advance of deep learning techniques, the detection accuracy for pig monitoring is challenged by occlusion problems due to a complex structure of a pig room such as food facility. These detection difficulties with a single image can be mitigated by using a video data. In this research, we propose a method in pig detection for video monitoring environment with a static camera. That is, by using both image processing and deep learning techniques, we can recognize a complex structure of a pig room and this information of the pig room can be utilized for improving the detection accuracy of pigs in the monitored pig room. Furthermore, we reduce the execution time overhead by applying a pruning technique for real-time video monitoring on an embedded board. Based on the experiment results with a video data set obtained from a commercial pig farm, we confirmed that the pigs could be detected more accurately in real-time, even on an embedded board.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.592-597
• /
• 2003

In this paper, some differential geometric conditions for the observer using a recurrent neural network are provided in terms of a stochastic nonlinear system control. In the stochastic nonlinear system, it is necessary to make an additional condition for observation of stochastic nonlinear system, called perfect filtering condition. In addition, we provide a observer using a recurrent neural network for the observation of a stochastic nonlinear system with the proposed observation conditions. Computer simulation shows that the control performance of the stochastic nonlinear system with a observer using a recurrent neural network satisfying the proposed conditions is more efficient than the conventional observer as Kalman filter

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.945-948
• /
• 2005

In wireless and handheld platforms area, performance, power and cost are key metrics for product success. This is driving increasing levels of on-chip integration in state-of-the-art application processors. The purpose of this project is to optimize and design the energy-efficient embedded system that properly displays video and audio in real time. The requirements are for the media player to be capable of decoding real-time streaming video and audio with the least possible energy consumption for a variety of different clips at different resolutions. We implemented this Linux based multimedia player on Intel's PXA27x platform.

• Journal of Korea Multimedia Society
• /
• v.10 no.12
• /
• pp.1601-1611
• /
• 2007

In this paper, we presented geometric CAD watermarking scheme for Architectural design drawings using line, arc, and circle components to prevent infringement of copyright from unlawfulness reproductions and distribution. The conventional CAD watermarking scheme can be applied to both line and arc components. But the proposed scheme consists of line, arc and circle watermarking schemes for three basic components of CAD design. After extracting line, arc and circle components from designed drawing, the watermark is embedded into the length of Line component, the angle of arc component, and the radius of circle component considering the robustness against various geometric transformations. The embedding strengths in each component are determined to be preserving the transparency of the watermark. By experimental result, we confirmed the robustness and the invisibility of embedded watermarks in several conversions of architectural design drawing.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.11a
• /
• pp.91-94
• /
• 2001

본 논문에서는 멀티웨이브릿 필터뱅크를 이용한 임베디드 제로 트리 구조(embedded zero tree structure)의 영상 압축방법을 제안한다. 본 논문에서 사용한 멀티웨이브릿 함수는 근사화 차수가 2인 새로운 방법의 DGHM 멀티웨이브릿을 사용하였으며, 영상을 부호화하기 위하여 임베디드 제로 트리 구조의 코딩기법을 사용한다. 높은 근사화 차수를 갖는 멀티웨이브릿 필터는 보다 우수한 에너지 압축 효과를 가지게 된다. 따라서 본 논문에서는 우수한 에너지 압축 효과는 내기 위하여 수학적으로 정립이 된 근사화 차수가 2인 DGHM 멀티웨이브릿을 영상압축에 사용한다. 또한 기존의 멀티웨이브릿 필터들보다 높은 성능을 갖는 DGHM 멀티스케일링 및 멀티웨이브릿 필터를 구성하여 영상을 완전복원이 되도록 한다.

• Journal of Korea Multimedia Society
• /
• v.18 no.10
• /
• pp.1149-1156
• /
• 2015

This paper proposes the hand contour detector which is robust to noises. Existing methods reduce noises by applying morphology to extracted edges, detect finger tips by using the center of hands, or exploit the intersection of curves from hand area candidates based on J-value segmentation(JSEG). However, these approaches are so vulnerable to noises that are prone to detect non-hand parts. We propose the noise tolerant hand contour detection method in which non-skin area noises are removed by applying skin area detection, contour detection, and a threshold value. By using the implemented system, we observed that the system was successfully able to detect hand contours.

• Journal of Korea Multimedia Society
• /
• v.17 no.7
• /
• pp.829-837
• /
• 2014

This paper develops an automatic go recording system based on image processing. We use Random Sample Consensus to detect the circular shape of stone and propose a set of methods to improve the computational overhead of RANSAC. The proposed scheme is not affected by the changes of stone location, illumination, and camera distance, which is different from existing methods. We implemented the proposed scheme into a working system and confirmed that the recording is feasible and the problems have been improved.

• Convergence Security Journal
• /
• v.9 no.2
• /
• pp.85-89
• /
• 2009

This study to solve these problems a search system has been designed by combining the two methods. Also the search and manage image data by handheld devices such as portable PDA or smart phone, a system has been designed application to an embedded system. Once this is used, multimedia data can be efficiently searched and utilized by handheld devices.