• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.221-235
• /
• 2022

Manual inspection of steel box girders on long span bridges is time-consuming and labor-intensive. The quality of inspection relies on the subjective judgements of the inspectors. This study proposes an automated approach to detect and segment cracks in high-resolution images. An end-to-end cascaded framework is proposed to first detect the existence of cracks using a deep convolutional neural network (CNN) and then segment the crack using a modified U-Net encoder-decoder architecture. A Naïve Bayes data fusion scheme is proposed to reduce the false positives and false negatives effectively. To generate the binary crack mask, first, the original images are divided into 448 × 448 overlapping image patches where these image patches are classified as cracks versus non-cracks using a deep CNN. Next, a modified U-Net is trained from scratch using only the crack patches for segmentation. A customized loss function that consists of binary cross entropy loss and the Dice loss is introduced to enhance the segmentation performance. Additionally, a Naïve Bayes fusion strategy is employed to integrate the crack score maps from different overlapping crack patches and to decide whether a pixel is crack or not. Comprehensive experiments have demonstrated that the proposed approach achieves an 81.71% mean intersection over union (mIoU) score across 5 different training/test splits, which is 7.29% higher than the baseline reference implemented with the original U-Net.

• Proceedings of the Korea Information Processing Society Conference
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• pp.171-174
• /
• 2021

최근 사물인터넷(IoT) 환경에서 다양한 장비의 인터넷 통신이 가능하여 이에 적절한 경량 블록 암호 알고리즘에 대한 연구가 활발히 진행되고 있다. ICISC 2020에서 새로 발표된 국산 경량 블록 암호 알고리즘인 PIPO는 새로운 경량 S-Box를 조합한 unbalanced-Bridge 구조로 효율적인 비트슬라이싱 구현을 제공한다. IoT 환경에 PIPO가 적용되기 위해서는 부채널 분석에 대한 안전성이 보장되어야 한다. 따라서 본 논문에서는 PIPO가 1차 CPA 공격에 취약함을 확인한다. 그리고 부채널 공격에 대응하기 위해 1차 마스킹 기법을 제안한다. 제안한 마스킹 기법은 1차 CPA 공격에 안전하였으며, 마스킹 적용 전보다 -375%의 성능을 보였다. 그리고 기존 기법보다 1287% 속도가 빨라진 것을 확인하였다.

• Journal of the Korea Concrete Institute
• /
• 제26권3호
• /
• pp.277-285
• /
• 2014

The purpose of this study is to investigate experimentally the variation factors range having influence on the rheological properties of the combined self-compacting concrete according to materials quality, weighting error and site conditions. Two types cement (blast-furnace slag cement and belite cement), lime stone powder as binder and the optimum mix proportions in the preceded study are selected for this study. Also, variations for sensitivity test are as followings; (1) Concrete temperature 3 cases (2) Surface moisture of sand 5cases (3) Fineness modulus of sand 5cases (4) Specific surface of lime stone powder 3cases (5) Dosage of chemical admixture 5cases. Slump flow ($650{\pm}50mm$), 500 mm reaching time (($7{\pm}3sec$), V-type flowing time ($15{\pm}5sec$) and U-box height (min. 300 mm) are tested for sensitivity. As test results, the variations range for quality control are as followings. (1) Concrete temperature; $10{\sim}20^{\circ}C$(below $30^{\circ}C$) (2) Surface moisture of sand; $base{\pm}0.6%$ (3) Fineness modulus of sand; $2.6{\pm}0.2$ (4) Dosage of chemical admixture; $base{\pm}0.2%$ (5) Specific surface of lime stone powder $6000cm^2/g$. Compared with two types cement including based belite cement (binary type) and based slag cement (ternary type), the combined self-compacting concrete used belite cement type is most stable in the quality control because of high contents for lime stone powder and $C_2S$. It is to propose a control scheme of the combined self-compacting concrete in the actual construction work.

• The KIPS Transactions:PartC
• /
• 제11C권6호
• /
• pp.731-738
• /
• 2004

With the increase of digital content usages, the protection for digital content and intellectual property becomes more important. The DRM(digital rights management) technologies are applicable to protect not only any kind of digital contents but also intellectual property. Besides such techniques are required for recorded digital broadcasting contents due to introduction of digital broadcasting techniques and storage devices such as personal video recorder. The conventional protection scheme for broadcasting content is the CAS(conditional access system) by which the access of viewer is controlled on the specific channels or programs. The CAS prohibits the viewer from delivering the digital broadcasting content to other person, so it results in restriction of superdistribution on the digital broadcasting content. In this paper, for broadcast targeting unspecfic many people, we will design the service model of the protection and distribution of digital broadcasting content using encryption and license by employing the concept of DRM. The results of implementation are also shown to verify some functions of each component. An implemented system of this paper has some advantages that the recording of broadcast content is allowed on set-top-box and superdistribution is available by consumer. Hence it provides content providers and consumers with trustworthy environment for content protection and distribution.

• Korean Journal of Food Science and Technology
• /
• 제21권5호
• /
• pp.676-685
• /
• 1989

Optimal operating conditions of on-farm red pepper dryer were searched by using the simulation-optimization algorithm combining the drying and quality deterioration models of red pepper with Box's complex method. Determination of control variables such as air temperature, air recycle ratio and air flow rate was based on a criterion of minimizing energy consumption under the constrainst conditions that satisfied the specified color retention of carotenoids. As quality constraint was stricter, energy consumption increased and total drying time decreased with lower recycle ratio and higher air flow rate Product mixing during drying was found to be able to improve the energy efficiency and product quality. Currently used air flow rate was assessed to be increased for the optimal operation. Two stage drying at the fixed optimal air flow rate was proven to be useful means for further saying of energy consumption. In the optimal bistaged drying, the second stage began at about one third of the total drying time and low air temperature in the first stage Increased to a high value and air recycle ratio increased slightly in the second stage. Optimal control variable scheme could be explained by the dryer performance and the carotenoids destruction kinetics in red pepper drying.

• Tunnel and Underground Space
• /
• 제34권3호
• /
• pp.218-230
• /
• 2024

Recently, many studies have been conducted for safety management in construction sites by incorporating computer vision. Anchor box parameters are used in state-of-the-art deep learning-based object detection and segmentation, and the optimized parameters are critical in the training process to ensure consistent accuracy. Those parameters are generally tuned by fixing the shape and size by the user's heuristic method, and a single parameter controls the training rate in the model. However, the anchor box parameters are sensitive depending on the type of object and the size of the object, and as the number of training data increases. There is a limit to reflecting all the characteristics of the training data with a single parameter. Therefore, this paper suggests a method of applying multiple parameters optimized through data split to solve the above-mentioned problem. Criteria for efficiently segmenting integrated training data according to object size, number of objects, and shape of objects were established, and the effectiveness of the proposed data split method was verified through a comparative study of conventional scheme and proposed methods.

• Smart Structures and Systems
• /
• 제20권3호
• /
• pp.273-283
• /
• 2017

This paper presents and compares three feedback control strategies for active control of noise inside a 3-D vibro-acoustic cavity. These are a) control strategy based on direct output feedback (DOFB) b) control strategy based on linear quadratic regulator (LQR) to reduce structural vibrations and c) LQR control strategy with a weighting scheme based on structural-acoustic coupling coefficients. The first two strategies are indirect control strategies in which noise reduction is achieved through active vibration control (AVC), termed as AVC-DOFB and AVC-LQR respectively. The third direct strategy is based on active structural-acoustic control (ASAC). This strategy is an LQR based optimal control strategy in which the coupling between the various structural and the acoustic modes is used to design the controller. The strategy is termed as ASAC-LQR. A numerical model of a 3-D rectangular box cavity with a flexible plate (glued with piezoelectric patches) and with other five surfaces treated rigid is developed using finite element (FE) method. A single pair of collocated piezoelectric patches is used for sensing the vibrations and applying control forces on the structure. A comparison of frequency response function (FRF) of structural nodal acceleration, acoustic nodal pressure, and piezoelectric actuation voltage is carried out. It is found that the AVC-DOFB control strategy gives equal importance to all the modes. The AVC-LQR control strategy tries to consume the control effort to damp all the structural modes. It is seen that the ASAC-LQR control strategy utilizes the control effort more intelligently by adding higher damping to those structural modes that matter more for reducing the interior noise.

• Journal of Advanced Marine Engineering and Technology
• /
• 제31권1호
• /
• pp.26-33
• /
• 2007

The intake swirl motion, as one of dominant effects for an engine combustion. is very effective for turbulence enhancement during the compression process in the cylinder of 2-valve engine. Because the combustion flame speed is determined by the turbulence that is mainly generated from the mean flow of the charge air motion in intake port system. This paper describes the experimental results of swirl flow and combustion characteristics by using the oil spot method and back-scattering Laser Doppler velocimeter (LDV) in 2-valve single cylinder transparent LPG engine using the liquid phase LPG injection. For this. various intake port configurations were developed by using the flow box system and swirl ratios for different intake port configurations were determined by impulse swirl meter in a steady flow rig test. And the effects of intake swirl ratio on combustion characteristics in an LPG engine were analyzed with some analysis parameters that is swirl ratio. mean flow coefficient, swirl mean velocity fuel conversion efficiency. combustion duration and cyclic variations of indicated mean effective pressure(IMEP). As these research results, we found that the intake port configuration with swirl ratio of 2.0 that has a reasonable lean combustion stability is very suitable to an $11{\ell}$ heavy-duty LPG engine with liquid phase fuel injection system. It also has a better mean flow coefficient of 0.34 to develope a stable flame kernel and to produce high performance. This research expects to clarify major factor that effects on the design of intake port efficiently with the optimized swirl ratio for the heavy duty LPG engine.

• Journal of the Korean Society for Library and Information Science
• /
• 제36권2호
• /
• pp.55-72
• /
• 2002

This study is a part of the serial studies on the historical developments of cataloging rules in Korea and focuses on the rules related with KORMARC. As for KORMARC formats, the experimental format based on LCMARC and UKMARC was developed in 1980, after that, KORMARC Pyojunyong Format for monographs in 1981, KORMARC Pyojun Format for monographs in 1984. At last KORMARC KS Format for monographs has been published in 1993 after discussing carefully the revised scheme of KORMARC format by the Korean Society for Information Management. The pre-note edition of KORMARC Description Rules for monographs, based on ISBD, was published in 1983, the revised pre-note edition in 1985, and at last A study on KORMARC Description Rules for Monographs by the Korean Society for Information Management in 1991.

• KIISE Transactions on Computing Practices
• /
• 제20권10호
• /
• pp.534-542
• /
• 2014

Multiscale modeling is a new simulation approach which can manage different spatial and temporal scales of system. In this study, as part of multiscale modeling research, we propose the way of combining two different simulation methods, molecular dynamics(MD) and stochastic rotation dynamics(SRD). Our conceptual implementations are based on LAMMPS, one of the well-known molecular dynamics programs. Our prototype of multiscale modeling follows the form of the third party implementation of LAMMPS. It added MD to SRD in order to simulate the boundary area of the simulation box. Because it is important to guarantee the seamless simulation, we also designed the overlap zones and communication zones. The preliminary experimental results showed that our proposed scheme is properly worked out and the execution time is also reduced.