• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2769-2792
• /
• 2013

In this paper, we propose an authenticated key agreement scheme, TinyIBAK, based on the identity-based cryptography and bilinear paring, for large scale sensor networks. We prove the security of our proposal in the random oracle model. According to the formal security validation using AVISPA, the proposed scheme is strongly secure against the passive and active attacks, such as replay, man-in-the middle and node compromise attacks, etc. We implemented our proposal for TinyOS-2.1, analyzed the memory occupation, and evaluated the time and energy performance on the MICAz motes using the Avrora toolkits. Moreover, we deployed our proposal within the TOSSIM simulation framework, and investigated the effect of node density on the performance of our scheme. Experimental results indicate that our proposal consumes an acceptable amount of resources, and is feasible for infrequent key distribution and rekeying in large scale sensor networks. Compared with other ID-based key agreement approaches, TinyIBAK is much more efficient or comparable in performance but provides rekeying. Compared with the traditional key pre-distribution schemes, TinyIBAK achieves significant improvements in terms of security strength, key connectivity, scalability, communication and storage overhead, and enables efficient secure rekeying.

• 국제학술발표논문집
• /
• The 1th International Conference on Construction Engineering and Project Management
• /
• pp.841-844
• /
• 2005

Computer vision technology can be used to interpret the images captured by web cameras installed on construction sites to automatically quantify the results. This information can be used for quality control, productivity measurement and to direct construction. Steel frame construction is particularly well suited for automatic monitoring as all structural members can be viewed from a small number of camera locations, and three-dimensional computer models of steel structures are frequently available in a standard electronic format. A system is being developed that interprets the 3-D model and directs a camera to look for individual members as regular intervals to determine when each is in place and report the results. Results from a simple lab-scale system are presented along with preliminary full-scale development.

• 소성∙가공
• /
• 제30권4호
• /
• pp.186-194
• /
• 2021

Finite element simulation is a widely applied method for practical purpose in various metal forming process. However, in the simulation of elasto-plastic behavior of porous material or in crystal plasticity coupled multi-scale simulation, it requires much calculation time, which is a limitation in its application in practical situations. A machine learning model that directly outputs the constitutive equation without iterative calculations would greatly reduce the calculation time of the simulation. In this study, we examined the possibility of artificial intelligence based constitutive equation with the input of existing state variables and current velocity filed. To introduce the methodology, we described the process of obtaining the training data, machine learning process and the coupling of machine learning model with commercial software DEFROM^TM, as a preliminary study, via rigid plastic finite element simulation.

• Structural Engineering and Mechanics
• /
• 제80권6호
• /
• pp.727-736
• /
• 2021

In this work, mathematical modeling of the passive vibration controls of a three-layered sandwich beam under hard excitation is developed. Kelvin-Voigt Viscoelastic model is considered in the core. The formulation is based on the higher-order zig-zag theories where the normal and shear deformations are taken into account only in the viscoelastic core. The dynamic behaviour of the beam is represented by a complex highly nonlinear ordinary differential equation. The method of multiple scales is adopted to solve the analytical frequency-amplitude relationships in the super-harmonic resonance case. Parametric studies are carried out by using HSDT and first-order deformation theory by considering different geometric and material parameters.

• 한국연소학회지
• /
• 제15권1호
• /
• pp.22-29
• /
• 2010

In this study, carbon conversion was measured using an electronic mass balance. In a lab scale furnace, each coal sample was pyrolyzed in a nitrogen environment and became coal char, which was then gasified with steam under isothermal conditions. The reactivity of coal char was investigated at various temperatures and steam concentrations. The VRM(volume reaction model), SCM(shrinking core model), and RPM(random pore model) were used to interpret experimental data. For each model the activation energy(Ea), pre-exponential factor (A), and reaction order(n) of the coal char-steam reaction were determined by applying the Arrhenius equation into the data obtained with thermo-gravimetric analysis(TGA). According to this study, it was found that experimental data agreed better with the VRM and SCM for 1,000 and $1,100^{\circ}C$, and the RPM for 1,200 and $1,300^{\circ}C$. The reactivity of chars increased with the increase of gasification temperature. The structure parameter(${\psi}$) of the surface area for the RPM was obtained.

• 대한환경공학회지
• /
• 제38권8호
• /
• pp.444-451
• /
• 2016

기존에는 lab-scale 연구에서 이론식을 기초로 막오염 모델식을 구성하였지만, 이러한 모델식은 여과, 역세, 배출이 연속적으로 이루어지는 실규모 현장에 적용하기에는 적합하지 않았다. 본 연구는 실제로 인천시 G-정수사업소에서 발생되는 배출수 처리를 위해 연속자동 운전되고 있는 침지막 공정을 대상으로 진행되었다. 정유량 조건에서 막오염 관리지표를 막간차압(Trans-Membrane Pressure, TMP)으로 결정하고 침지막 공정의 주요 운전변수인 총 투과유량과 조 내 SS농도를 독립변수로 하여 TMP의 시계열 예측을 시도하고 예측 가능성 및 적용성을 평가하였다. 유전자알고리즘을 이용한 시계열 예측모형을 구성한 결과, TMP 예측값이 펄스주기 형태와 경시적인 증가 추세 두 가지를 모두 반영하고 있어서 만족할 만한 결과가 나왔다. 두 번의 검증 결과, 선형회귀 방식으로 TMP 실측치와 예측치의 상관성(유의성)을 나타내면 각각 $r^2=0.721$, $r^2=0.928$ 수준이다. 본 연구에서는 하절기 자료를 활용하여 모델링 작업을 수행하였지만 추후에 연속자료가 더 쌓이면 같은 절차로 모델링 작업을 반복해서 더 높은 신뢰도의 예측모형을 구성할 수 있고 이를 실제 현장에 적용하여 2~3일 정도의 단기예측을 수행한다면 실제로 막공정을 에너지 효율적으로 운영하는데 도움이 될 것으로 사료된다.

• ETRI Journal
• /
• 제34권1호
• /
• pp.142-145
• /
• 2012

The main challenge at present in constructing hierarchical identity-based encryption (HIBE) is to solve the trade-off between private-key size and ciphertext size. At least one private-key size or ciphertext size in the existing schemes must rely on the hierarchy depth. In this letter, a new hierarchical computing technique is introduced to HIBE. Unlike others, the proposed scheme, which consists of only two group elements, achieves constant-size private keys. In addition, the ciphertext consists of just three group elements, regardless of the hierarchy depth. To the best of our knowledge, it is the first efficient scheme where both ciphertexts and private keys achieve O(1)-size, which is the best trade-off between private-key size and ciphertext size at present. We also give the security proof in the selective-identity model.

• 한국연소학회:학술대회논문집
• /
• 한국연소학회 2012년도 제44회 KOSCO SYMPOSIUM 초록집
• /
• pp.217-219
• /
• 2012

This study describes 3D RANS simulation of a 2.1 MW swirling pulverized coal flame in a semi-industrial scale furnace. The simulation of pulverized coal combustion involves various models for complex physical processes and needs information of pyrolysis rate, the yields and compositions of volatiles and char especially in coal conversion. The coal conversion information can be acquired by the experiment or the pre-processor code. The empirical model based on the experiment of the IFRF and the structural model based on the pre-processor code of the PC-COAL-LAB were evaluated against the measurement data.

• 한국분무공학회지
• /
• 제23권1호
• /
• pp.16-21
• /
• 2018

The current study has developed an in-house 3D FEM code in order to model thermoacoustic problems in a gas turbine combustion system and compared calculation results of main instability characteristics with measured ones from a lab-scale partially-premixed combustor. From the comparison of calculation results with the measured data, the current model could successfully capture the harmonic longitudinal instability frequencies and their spatial distributions of the acoustic field as well as the growth rate of self-excited modes.