• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.04a
• /
• pp.781-784
• /
• 2014

Face recognition under controlled settings, such as limited viewpoint and illumination change, can achieve good performance nowadays. However, real world application for face recognition is still challenging. In this paper, we use Affine SIFT to detect affine invariant local descriptors for face recognition under large viewpoint change. Affine SIFT is an extension of SIFT algorithm. SIFT algorithm is scale and rotation invariant, which is powerful for small viewpoint changes in face recognition, but it fails when large viewpoint change exists. In our scheme, Affine SIFT is used for both gallery face and probe face, which generates a series of different viewpoints using affine transformation. Therefore, Affine SIFT allows viewpoint difference between gallery face and probe face. Experiment results show our framework achieves better recognition accuracy than SIFT algorithm on FERET database.

• Transactions of Materials Processing
• /
• v.3 no.1
• /
• pp.110-119
• /
• 1994

In this paper, a new forming process of large-size forgings of converged nozzle-shape is developed by the experimental study using the incremental forging method and combined forming method. The development of the forming process is focused on the manufacturing of large-size forgings by the press with medium load capacity. Various related processes are proposed and modelling experiments using plasticine are carried out. Thus, the incremental forging method for expanding is recommanded from the study of formability and forming load, etc. The selected process is then subjected to modelling experiments of lead and the design parameters such as preform for final process, die-width of the upper die and reduction amount of each stroke are determined. In order to verify the effectiveness of the selected process, 1/7 scale prototype experiment of the real material is carried out. Forgings of converged nozzle shape can be produced by the developed process within the limit loads and with the simple tools.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.9 no.5
• /
• pp.195-205
• /
• 2001

The mechanical properties of aluminum honeycomb on the direction of axial crushing under quasistatic loading test was investigated. The crushing process was simulated numerically by full-scale finite element models. Simulations reproduce the experimental results both qualitatively as well as quantitatively. From the investigation, we suggested the constitutive model of energy absorbing honeycomb structure for large scale impact analysis. Real impact test of the WB(Moving Deformable Barrier) was carried and compared with finite element simulation. Constitutive model used in the numerical simulation had a good correlation with experiment. By suggesting the optimizing method fur honeycomb cell configuration design, relationship between cell configuration and crush strength is studied.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.10a
• /
• pp.265-268
• /
• 1997

In the recent day, fatigue life prediction techniques play a major role in the design of components in the ground vehicle industry. Full scale durability testing in the laboratory is an essential of any fatigue life evaluation of components or structures of the automotwe vehicle. Component testing is part~cularly important in today's highly competitive industries where the design to reduce weight and production costs must be balanced with the necessity to avoid expensive service failure. Generally, multi-axial road simulator is used to carry out the fatigue test and the vibration test. In this paper, the algorithm and software to realize the real road profile are developed. The validity of the software are verified by applying the belgian road, the city road, the highway, and the gravel road. The results of the above experiment show that the real road profiles are realized well after loth iteration.

• Proceedings of the KSR Conference
• /
• 2001.10a
• /
• pp.540-545
• /
• 2001

When a testing body is being inserted to the ground by Front Jacking Method, the influence of friction forces around the body plays important role in the characteristics of whole ground behavior. In the study, the Front Jacking Method that most frequently used in Korea who applied with experimental way. And the characteristics of ground behavior that caused by the friction that acts to the upper slab was analysed for the basic data that can be utilized for real scale experiment.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.50-53
• /
• 2016

Image processing techniques play an increasingly important role in many aspects of our daily life. For example, it has been shown to improve agricultural productivity in a number of ways such as plant pest detecting or fruit grading. However, massive quantities of images generated in real-time through multi-devices such as remote sensors during monitoring plant growth lead to the challenges of big data. Meanwhile, most current image processing systems are designed for small-scale and local computation, and they do not scale well to handle big data problems with their large requirements for computational resources and storage. In this paper, we have proposed an IPABigData (Image Processing Algorithm BigData) platform which provides algorithms to support large-scale image processing in agriculture based on Hadoop framework. Hadoop provides a parallel computation model MapReduce and Hadoop distributed file system (HDFS) module. It can also handle parallel pipelines, which are frequently used in image processing. In our experiment, we show that our platform outperforms traditional system in a scenario of image segmentation.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.493-505
• /
• 2023

Comprehensive condition monitoring of large industry systems such as nuclear power plants (NPPs) is essential for safety and maintenance. In this study, we developed novel system-scale diagnostic technology based on deep-learning and IR thermography that can efficiently and cost-effectively classify system conditions using compact Raspberry Pi and IR sensors. This diagnostic technology can identify the presence of an abnormality or accident in whole system, and when an accident occurs, the type of accident and the location of the abnormality can be identified in real-time. For technology development, the experiment for the thermal image measurement and performance validation of major components at each accident condition of NPPs was conducted using a thermal-hydraulic integral effect test facility with compact infrared sensor modules. These thermal images were used for training of deep-learning model, convolutional neural networks (CNN), which is effective for image processing. As a result, a proposed novel diagnostic was developed that can perform diagnosis of components, whole system and accident classification using thermal images. The optimal model was derived based on the modern CNN model and performed prompt and accurate condition monitoring of component and whole system diagnosis, and accident classification. This diagnostic technology is expected to be applied to comprehensive condition monitoring of nuclear power plants for safety.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1494-1507
• /
• 2018

Intelligent human identification using face information has been the research hotspot ranging from Internet of Things (IoT) application, intelligent self-service bank, intelligent surveillance to public safety and intelligent access control. Since 2D face images are usually captured from a long distance in an unconstrained environment, to fully exploit this advantage and make human recognition appropriate for wider intelligent applications with higher security and convenience, the key difficulties here include gray scale change caused by illumination variance, occlusion caused by glasses, hair or scarf, self-occlusion and deformation caused by pose or expression variation. To conquer these, many solutions have been proposed. However, most of them only improve recognition performance under one influence factor, which still cannot meet the real face recognition scenario. In this paper we propose a multi-scale parallel convolutional neural network architecture to extract deep robust facial features with high discriminative ability. Abundant experiments are conducted on CMU-PIE, extended FERET and AR database. And the experiment results show that the proposed algorithm exhibits excellent discriminative ability compared with other existing algorithms.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.34 no.3
• /
• pp.309-317
• /
• 2016

Debris flows are rapidly flowing masses of water mixed with soil and gravel from landslides which are caused by typhoons or rainstorms. The combination of Korea’s mountain dominated topography (70%) and seasonal heavy rains and typhoons causes landslides and large-scale debris flows from June to August. These phenomena often cause property damage and casualties that amount up to 20% of total annual disaster fatalities. The key point to predicting debris flow is to understand its movement mechanism, erosion, and deposition. In order to achieve a more accurate estimation of debris flow path and damage, this study incorporates quantitative analysis of high resolution LiDAR DEM (GSD 10cm) to delineate geomorphic and topographic changes induced by Jinbu real scale debris flow test.

• Fire Science and Engineering
• /
• v.31 no.6
• /
• pp.16-22
• /
• 2017

In Korea, performance-based fire safety designs are being discussed to deal with the various risks of fire in complex and diverse types of structure. However, performance-based fire safety designs are not actively employed because it is difficult to estimate the fire characteristics related to the various factors in buildings. In this study, real scale fire tests were conducted based on fire severity levels and fire loads provided in He New Zealand Building Code, in order to use the results as guidelines and fundamental data for performance-based designs. In the real scale fire tests conducted in a 10MW full-scale calorimeter, wood cribs were placed in a $2.4(L){\times}3.6(W){\times}2.4(H)m$ mock-up of a compartment which had one $0.8(L){\times}2.0(H)$ opening for different fire loads and heating was continued until all of the wood cribs were burned down. The heat release rate started to increase rapidly 90 seconds after the wood cribs caught fire. In the test with a fire load level 1, the maximum heat release rate of 4743.4 kW was reached at 244 second. In the test with fire load level 2, a maximum heat release rate of 5050.9 kW was reached at 497 second. In the test with fire load level 3, a maximum heat release rate of 4446.9 kW was reached at 677 second.