• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.302-313
• /
• 2023

This paper aims to develop a framework that can fully automate the quality management of training data used in large-scale Artificial Intelligence (AI) models built by the Ministry of Science and ICT (MSIT) in the 'AI Hub Data Dam' project, which has invested more than 1 trillion won since 2017. Autonomous driving technology using AI has achieved excellent performance through many studies, but it requires a large amount of high-quality data to train the model. Moreover, it is still difficult for humans to directly inspect the processed data and prove it is valid, and a model trained with erroneous data can cause fatal problems in real life. This paper presents a dataset reconstruction framework that removes abnormal data from the constructed dataset and introduces strategies to improve the performance of AI models by reconstructing them into a reliable dataset to increase the efficiency of model training. The framework's validity was verified through an experiment on the autonomous driving dataset published through the AI Hub of the National Information Society Agency (NIA). As a result, it was confirmed that it could be rebuilt as a reliable dataset from which abnormal data has been removed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.4C
• /
• pp.312-320
• /
• 2008

The macroblock-based adaptive interpolation filter method has been considered to be able to achieve high coding efficiency in H.264/AVC. In this method, although the filter selection criterion considered in terms of rate and distortion have showed a good performance, it still leaves room for improvement. To improve high coding efficiency better than conventional method, we propose a new filter selection criterion which considers two bit rates, motion vector and prediction error, and reconstruction error. In addition, the algorithm for reducing the overhead of transmitting the selected filter information is presented. Experimental results show that the proposed method significantly improves the coding efficiency compared to ones using conventional criterion. It leads to about a 5.19% (1 reference frame) and 5.14% (5 reference frames) bit rate savings on average compared to H.264/AVC, respectively.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.295-307
• /
• 2022

In this paper, we propose a novel compression method for scalable Fisher vectors (SCFV) which is used as a global visual feature description of individual video frames in MPEG CDVA standard. CDVA standard has adopted a temporal descriptor redundancy removal technique that takes advantage of the correlation between global feature descriptors for adjacent keyframes. However, due to the variable length property of SCFV, the temporal redundancy removal scheme often results in inferior compression efficiency. It is even worse than the case when the SCFVs are not compressed at all. To enhance the compression efficiency, we propose an asymmetric SCFV difference computation method and a SCFV reconstruction method. Experiments on the FIVR dataset show that the proposed method significantly improves the compression efficiency compared to the original CDVA Experimental Model implementation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.295-302
• /
• 2018

Face recognition (FR) has been widely used in various applications, such as smart surveillance systems, immigration control in airports, user authentication in smart devices, and so on. FR in well-controlled conditions has been extensively studied and is relatively mature. However, in unconstrained conditions, FR performance could degrade due to undesired characteristics of the input face image (such as irregular facial pose variations). To overcome this problem, this paper proposes a new method for determining if an input image is suitable for FR. In the proposed method, for an input face image, reconstruction error is computed by using a predefined set of reference face images. Then, suitability can be determined by comparing the reconstruction error with a threshold value. In order to reduce the effect of illumination changes on the determination of suitability, a preprocessing algorithm is applied to the input and reference face images before the reconstruction. Experimental results show that the proposed method is able to accurately discriminate non-frontal and/or incorrectly aligned face images from correctly aligned frontal face images. In addition, only 3 ms is required to process a face image of $64{\times}64$ pixels, which further demonstrates the efficiency of the proposed method.

• Korean Journal of Construction Engineering and Management
• /
• v.12 no.1
• /
• pp.3-12
• /
• 2011

Technology based civilization has made a sustained advancement, resulting in economic growth and material prosperity. The outcome was the beginning of environmental destruction. Especially, in regard to construction, it has been recognized as a key culprit for pollution. As such, the construction industry has recently shown interest in the environment, as it has been applying environment friendly method of construction and reducing wastes. As the number of old common housing increases in our country, reconstruction and remodeling are implemented as a means of improvement. However, it can be said that remodeling is the inevitable option because of cost and environmental problems associated with reconstruction. As part of its feature, remodeling work is preceded by removal work. And, removal work inevitably creates construction wastes. Treatment of wastes that takes into consideration environmental aspects is important and for this, removal work that is based on environment friendly remodeling to ensure selective separation must be implemented. Yet, a removal work based on environment friendly remodeling has a lower level of output compared to existing methods. Because remodeling work by its nature has a post construction work which proceeds after it, securing work efficiency is important as the removal work is a critical path activity. Thus, the present study improves the work process of a removal work that is based on environment friendly remodeling so that it becomes a work process that includes work efficiency. For this, as a case study, old common housing has been selected, and a TACT process has been developed which is based on data acquired from environment friendly removal work. And, this study develops an optimal TACT process which is based on environment friendly remodeling by conducting a simulation for a 15 floor apartment.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.5
• /
• pp.199-202
• /
• 2013

In the cluster mechanism of the wireless sensor network, the network lifetime is affected by how cluster heads are selected. One of the representative clustering mechanisms, the low-energy adaptive clustering hierarchy (LEACH), selects cluster heads periodically, resulting in high energy consumption in cluster reconstruction. On the other hand, the adaptive clustering algorithm via waiting timer (ACAWT) proposes a non-periodic re-clustering mechanism that reconstructs clusters if the remaining energy level of a cluster head reaches a given threshold. In this paper, we propose a re-clustering mechanism that uses multiple remaining node energy levels and does re-clustering when the remaining energy level of a cluster head reaches one level lower. Also, in determining cluster heads, both of the number of neighbor nodes and the remaining energy level are considered so that cluster heads can be more evenly placed. From the simulations based on the Qualnet simulator, we validate that our proposed mechanism outperforms ACAWT in terms of the network lifetime.

• Journal of IKEEE
• /
• v.2 no.2 s.3
• /
• pp.239-246
• /
• 1998

In this paper, a new morphological details extraction algorithm is proposed. It is known that separate transmission of details and background smoothed image is a powerful technique for very low bit-rate video data transmission. Several details extraction algorithms show relatively large variation of grayscale levels in details, smoothed image, which is the difference between original and details images, provides highly distortedand complicated result. In order to remedy those pitfails, and thus to improve the coding efficiency, we propose a new algorithm using the reconstruction top-hat result as the reference image in process of obtaining details, instead of top-hat result, which is used fur existing a1gorithms. Experimental results show that details, extracted using the proposed algorithm, are much similar to original image, and thus the smoothed image is not too complicated.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.3 s.345
• /
• pp.10-17
• /
• 2006

For orthogonal frequency division multiplexing (OFDM), cyclic prefix (CP) should be longer than the length of channel impulse response, resulting in a loss of bandwidth efficiency. In this letter, the CP reconstruction (CPR) technique is first applied to a multi-input multi-output (MIMO)-OFDM system with insufficient CP. The intercarrier interference (ICI) from multiple transmit antennas is so large for MIMO system that it can not be sufficiently suppressed with the conventional CPR procedure used in single-input single-output (SISO) system. A new minimum mean-square error (MMSE) equalization and ordering process is proposed for MIMO system to suppress the ICI during the CPR procedure. By applying the proposed CPR algerian to MIMO-OFDM system, we can obtain both the benefits of multiplexing gai and spectral efficiency gain.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.3
• /
• pp.305-312
• /
• 2012

This paper aims at improving the accuracy and computational efficiency in reconstructing building boundaries from airborne Lidar points. We proposed an adaptive convex hull algorithm, which is a modified version of local convex hull algorithm in three ways. The candidate points for boundary are first selected to improve efficiency depending on their local density. Second, a searching-space is adjusted adaptively, based on raw data structure, to extract boundary points more robustly. Third, distance between two points and their IDs are utilized in detecting the seed points of inner boundary to distinguish between inner yards and inner holes due to errors or occlusions. The practicability of the approach were evaluated on two urban areas where various buildings exist. The proposed method showed less shape-dissimilarity(8.5%) and proved to be two times more efficient than the other method.

• Korean Journal of Construction Engineering and Management
• /
• v.7 no.4 s.32
• /
• pp.69-77
• /
• 2006

In recent years, there has been more reconstruction and redevelopment of apartments rather than efforts to consider alternative to extend the life time of the apartments. It is essential to try to develop ways to adequately maintain and to use the apartment facilities efficiently in order to preserve earth environment and the limited resource. However, tacking research on the effects obtained through remodelling and maintenance have interfered with the vitalizing of the market. The objective of this study is to propose criterions and methods with which to evaluate adequacy of maintenance management. A survey was performed to investigate important maintenance methods in order to obtain advices that can smoothly progress improve apartment maintenance management. This study applied AHP(Analytical Hierarchy Process) methods for reasonable maintenance in apartment. Through this study, the flow of apartment maintenance market is elevated to reaching the level in advanced nation.