• Journal of Broadcast Engineering
• /
• v.5 no.1
• /
• pp.82-93
• /
• 2000

Recently, the capability of efficient access to the desired video content is of growing importance because more digital video data are available at an increasing rate. A video summary abstracting the gist from the entirety enables the efficient browsing as well as the fast skimming of the video contents. In this paper, we discuss a novel dynamic summarization method based on the detection of highlights which represent semantically significant content and the description scheme (DS) proposed to MPEG-7 aiming to provide summary description. The summary DS proposed to MPEG-7 allows for efficient navigation and browsing to the contents of interest through the functionalities of multi-level highlights, hierarchical browsing and user-customized summarization. In this paper, we also show the validation and the usefulness of the methodology for dynamic summarization and the summary DS in real applications with soccer video sequences.

• Journal of Digital Contents Society
• /
• v.5 no.1
• /
• pp.87-94
• /
• 2004

In this paper, the digital watermarking algorithm is proposed using CDMA technique for protection of the mobile contents in the mobile environment. The digital watermarking was designed to robust the errors in the mobile environment where pathloss, multipath fading, interference, and noise exist. In case of the multimedia content service in the mobile environment, the construction method of the watermark, the algorithm of insertion and detection are also proposed. The watermark consists of the information of the mobile user. Invisibility and robustness required in watermarking are etimated. It is observed that PSNR of the mobile content inserted the watermark is 90.31 dB, and the signal processing and noise attack are also robust. Especially, because random noise occurs in wireless transmission can overcome, the proposed watermarking algorithm is adequate for protection of the multimedia contents in the mobile environment.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.37 no.7
• /
• pp.1-8
• /
• 2000

In this paper, we introduce a modified interference cancellation scheme for multiuser detection in CDMA(Code Division Multiple Access). This detector uses SIC(Successive Interference Canceller) scheme and divides the received signals to reduce the delay time. In this proposed structure, the active users are divided into a number of groups. Within each group, parallel detection is performed to estimate the output signal of that group. The estimated output signal due to that group is then subtracted from the received signal and the resulting residual signal is used for the parallel detection of the next group. This parallel and serial cancellation process is repeated until the last group in the stage is completed. The estimated output signals due to all groups except -th group are MAI for the user signals in -th group. Therefore, the estimated output signals due to all groups except th group are subtracted from the received signal, and then the obtained signal becomes the input signal of -th SIC. The proposed RDSIC (Reduced Delay time of Successive Interference Canceller) has performance and complexity close to the SIC, but with much less detection delay.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.18 no.2
• /
• pp.143-155
• /
• 2000

With the increasing needs for the integrated use of topographic and cadastral data in order to build an efficient geo-spatial information system. it is urgently necessary to research into its solution. The intention of this study is to detect error types of data and to propose adjustment methods for solving the problems caused by integrating topographic and cadastral data. For this purpose a primary integrated data model is created to link attribute data(land management system) and graphic data within cadastral information in the first step. In next, a secondary integrated data model based on the improved method is formed to coincide the graphic data of cadastral map with that of topographic map. At the first, because a numerous error types md sources caused by separate management of graphic and attribute data are easily checked, it is possible to suggest an improved method to correct these errors using the primary integrated data model. In addition, the accuracy in position and area with coordinate transformation method based on multi-block adjustment is more efficient than rubber-sheeting method. As a result, the secondary integrated data model could be built by harmonizing cadastral map with topographic map using the improved solution.

• Journal of Intelligence and Information Systems
• /
• v.16 no.1
• /
• pp.45-56
• /
• 2010

Sensor Networks are composed of many sensor nodes, which are capable of sensing, computing, and communicating with each other, and one or more sink node(s). Sensor networks collect information of various objects' identification and surrounding environment. Due to the limited resources of sensor nodes, use of wireless channel, and the lack of infrastructure, sensor networks are vulnerable to security threats. Most research of sensor networks have focused on how to detect and counter one type of attack. However, in real sensor networks, it is impractical to predict the attack to occur. Additionally, it is possible for multiple attacks to occur in sensor networks. In this paper, we propose the Secure Routing Mechanism to Defend Multiple Attacks in Sensor Networks. The proposed mechanism improves and combines existing security mechanisms, and achieves higher detection rates for single and multiple attacks.

• Journal of Advanced Navigation Technology
• /
• v.17 no.6
• /
• pp.804-809
• /
• 2013

The current state monitoring system at a semiconductor manufacturing process is based on the manually collected sensor data, which involves limitations when it comes to complex malfunction detection and real time monitoring. This study aims to design a situation recognition algorithm to form a network over time by creating a domain ontology and to suggest a system to provide users with services by generating events upon finding risk factors in the semiconductor process. To this end, a multiple sensor node for situational inference was designed and tested. As a result of the experiment, events to which the rule of time inference was applied occurred for the contents formed over time with regard to a quantity of collected data while the events that occurred with regard to malfunction and external time factors provided log data only.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.9 no.11
• /
• pp.1133-1144
• /
• 2019

Recently, the amount of image on the Internet has rapidly increased, due to the advancement of smart devices and various approaches to effective image retrieval have been researched under these situation. Existing image retrieval methods simply detect the objects in a image and carry out image retrieval based on the label of each object. Therefore, the semantic gap occurs between the image desired by a user and the image obtained from the retrieval result. To reduce the semantic gap in image retrievals, we connect the module for multiple objects classification based on deep learning with the module for human behavior classification. And we combine the connected modules with a behavior ontology. That is to say, we propose an image retrieval system considering the relationship between objects by using the combination of deep learning and behavior ontology. We analyzed the experiment results using walking and running data to take into account dynamic behaviors in images. The proposed method can be extended to the study of automatic annotation generation of images that can improve the accuracy of image retrieval results.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.11 no.1
• /
• pp.81-89
• /
• 2017

In this paper, we propose a gait pattern recognition method for intelligent prosthesis that enables walking in various environments of femoral amputees. The proposed gait mode changing method is a single sensor based algorithm which can discriminate gait surface and gait phase using only strain gauges sensor, and it is designed to simplify the algorithm based on multiple sensors of existing intelligent prosthesis and to reduce cost of prosthesis system. For the recognition algorithm, we analyzed characteristics of the ground reaction force generated during gait of normal person and defined gait step segmentation and gait detection condition, A gait analyzer was constructed for the gait experiment in the environment similar to the femoral amputee. The validity of the paper was verified through the defined detection conditions and fabricated instruments. The accuracy of the algorithm based on the single sensor was 95%. Based on the proposed single sensor-based algorithm, it is considered that the intelligent prosthesis system can be made inexpensive, and the user can directly grasp the state of the walking surface and shift the walking mode. It is confirmed that it is possible to change the automatic walking mode to switch the walking mode that is suitable for the walking mode.

• Progress in Medical Physics
• /
• v.20 no.2
• /
• pp.51-61
• /
• 2009

A Compton camera, which is based on the geometrical interpretation of Compton scattering, is a very promising gamma-ray imaging device considering its several advantages over the conventional gamma-ray imaging devices: high imaging sensitivity, 3-D imaging capability from a fixed position, multi-tracing functionality, and almost no limitation in photon energy. In the present study, a Monte Carlo-based, user-friendly Compton imaging simulator was developed in the form of a graphical user interface (GUI) based on Geant4 and $MATLAB^{TM}$. The simulator was tested against the experimental result of the double-scattering Compton camera, which is under development at Hanyang University in Korea. The imaging resolution of the simulated Compton image well agreed with that of the measured image. The imaging sensitivity of the measured data was 2~3 times higher than that of the simulated data, which is due to the fact that the measured data contains the random coincidence events. The performance of a stacking-structure type Compton camera was evaluated by using the simulator. The result shows that the Compton camera shows its highest performance when it uses 4 layers of scatterer detectors.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.2
• /
• pp.119-125
• /
• 2023

In this study, we photograph driving obstacle objects such as crosswalks, side spheres, manholes, braille blocks, partial ramps, temporary safety barriers, stairs, and inclined curb that hinder or cause inconvenience to the movement of the vulnerable using electric mobility aids. We develop an optimal AI model that classifies photographed objects and automatically recognizes them, and implement an algorithm that can efficiently determine obstacles in front of electric mobility aids. In order to enable object detection to be AI learning with high probability, the labeling form is labeled as a polygon form when building a dataset. It was developed using a Mask R-CNN model in Detectron2 framework that can detect objects labeled in the form of polygons. Image acquisition was conducted by dividing it into two groups: the general public and the transportation weak, and image information obtained in two areas of the test bed was secured. As for the parameter setting of the Mask R-CNN learning result, it was confirmed that the model learned with IMAGES_PER_BATCH: 2, BASE_LEARNING_RATE 0.001, MAX_ITERATION: 10,000 showed the highest performance at 68.532, so that the user can quickly and accurately recognize driving risks and obstacles.