• Journal of the Semiconductor & Display Technology
• /
• v.19 no.2
• /
• pp.11-15
• /
• 2020

In a typical initial setup of a PCB component inspection system, operators should manually input various information such as category, position, and inspection area for each component to be inspected, thus causing much inconvenience and longer setup time. Although there are many deep learning based object detectors, RetinaNet is regarded as one of best object detectors currently available. In this paper, a method using an extended RetinaNet is proposed that automatically detects its component category and position for each component mounted on PCBs from a high-resolution color input image. We extended the basic RetinaNet feature pyramid network by adding a feature pyramid layer having higher spatial resolution to the basic feature pyramid. It was demonstrated by experiments that the extended RetinaNet can detect successfully very small components that could be missed by the basic RetinaNet. Using the proposed method could enable automatic generation of inspection areas, thus considerably reducing the setup time of PCB component inspection systems.

• ETRI Journal
• /
• v.45 no.4
• /
• pp.666-677
• /
• 2023

In the last few years, convolutional neural networks (CNNs) have demonstrated good performance while solving various computer vision problems. However, since CNNs exhibit high computational complexity, signal processing is performed on the server side. To reduce the computational complexity of CNNs for edge computing, a lightweight algorithm, such as a MobileNet, is proposed. Although MobileNet is lighter than other CNN models, it commonly achieves lower classification accuracy. Hence, to find a balance between complexity and accuracy, additional hyperparameters for adjusting the size of the model have recently been proposed. However, significantly increasing the number of parameters makes models dense and unsuitable for devices with limited computational resources. In this study, we propose a novel MobileNet architecture, in which the number of parameters is adaptively increased according to the importance of feature maps. We show that our proposed network achieves better classification accuracy with fewer parameters than the conventional MobileNet.

• Communications for Statistical Applications and Methods
• /
• v.21 no.4
• /
• pp.349-361
• /
• 2014

This paper compares of lasso type estimators in various high-dimensional data situations with sparse parameters. Lasso, adaptive lasso, fused lasso and elastic net as lasso type estimators and ridge estimator are compared via simulation in linear models with correlated and uncorrelated covariates and binary regression models with correlated covariates and discrete covariates. Each method is shown to have advantages with different penalty conditions according to sparsity patterns of regression parameters. We applied the lasso type methods to Arabidopsis microarray gene expression data to find the strongly significant genes to distinguish two groups.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.28 no.1
• /
• pp.54-61
• /
• 2000

Covering rate, visual rating and sod development were evaluated under three natural fiber materials with Phragmites spp. when over a plastic sheet. The results were as follows. (1) The last covering rate was high on jute net, coir mat and on Miscanthus sacchariflorus, respectively while the early covering rate was high on coir mat and on Miscanthus sinensis+perennial ryegrass. (2) The early growth was good on perennial ryegrass but the covering rate gradually turned poor because of summer drought. (3) Sod was highly developed on Phragmites japonica, Miscanthus sacchariflorus and Miscanthus sinensis compared with other species and mixtures. (4) The covering rate and visual rating were high on natural fiber materials such as coir mat and jute net when compared with on natural fiber materials such as none treatment plots. (5) The natural fibers materials on Phragmites spp. and Miscanthus spp. were effect on sod establishment. Sod coir mat was highly established. (6) The carpet-type sod was best developed on the coir mat.

• The KSFM Journal of Fluid Machinery
• /
• v.13 no.6
• /
• pp.19-24
• /
• 2010

The demands for small, high performance and high loaded aircraft compressor are increased in the world. But the design requirements become increasingly complex to design these high technical engines, the requirement of the design optimization become increased. The optimal design result of several disciplines show different tendencies and nonlinear characteristics of the compressor design, the multidisciplinary design optimization method must be considered in compressor design. Therefore, the artificial Neural Net method is adapted to make the approximation model of 3-stage axial compressor design optimization for considering the nonlinear characteristic. At last, the optimal result of this study is compared to that of previous study.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.2
• /
• pp.584-589
• /
• 2010

In this paper we suggest multimedia synchronization model that is based on the Petri-net and services desirable quality of service requirement. Proposed model applies variable buffer which can be allowed, and then it presents high quality of service and real time characteristics. In this paper we expend Petri-net and propose new multimedia synchronization specification model in PBMSM and apply two analysis method of Petri-net to prove our model suggested. We compared with other models and showed high QoS.

• Nuclear Engineering and Technology
• /
• v.54 no.12
• /
• pp.4751-4758
• /
• 2022

Two-phase flow may almost exist in every branch of the energy industry. For the corresponding engineering design, it is very essential and crucial to monitor flow patterns and their transitions accurately. With the high-speed development and success of deep learning based on convolutional neural network (CNN), the study of flow pattern identification recently almost focused on this methodology. Additionally, the photographing technique has attractive implementation features as well, since it is normally considerably less expensive than other techniques. The development of such a two-phase flow pattern online monitoring system is the objective of this work, which seldom studied before. The ongoing preliminary engineering design (including hardware and software) of the system are introduced. The flow pattern identification method based on CNNs and transfer learning was discussed in detail. Several potential CNN candidates such as ALexNet, VggNet16 and ResNets were introduced and compared with each other based on a flow pattern dataset. According to the results, ResNet50 is the most promising CNN network for the system owing to its high precision, fast classification and strong robustness. This work can be a reference for the online monitoring system design in the energy system.

• Journal of the Korean Society of Safety
• /
• v.27 no.5
• /
• pp.95-98
• /
• 2012

There are many high-rise apartment building construction in Korea. There was an accident to pass through worker head by rebar dropped from height place. Therefore, low cost-high effectiveness method to prevent this type of accident should be revised and applied into the construction site. This study is to find out which method could be effectively applied to a site with low cost. Practical field test at 4th floor, 10th floor of apartment building site using re-bar diameter D10, D13, D16, D19, D22 with a length of 1 m, 1.5 m, 2 m, 2.5 m, 3 m which are common by used in a site. The test has also been done with a cover of styrofoam thickness 4.5 cm and thickness 9cm on field drop preventing net. One sheet of styrofoam thickness 45 mm has approximately two times stronger than only prevention net, It is found. Also, Two sheets have approximately two times stronger than one sheet of it.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.16 no.2
• /
• pp.146-154
• /
• 2014

In the result of effect on freezing damage reduction by the shade net colors and the shade rate to tea trees during wintering period, the high shade ratio decreased tea trees growth and increased freezing damage and 55% of shade ratio based on non treatment developed new leafs and green leaf productivity. By the shade net colors, colorless shade net (55% of shade type) treatment and green shade net treatment increased green leaf productivity and decreased damaged area compare to non treatment shade net and black shade net treatment. Colorless net shade treatment reduced over 50% of freezing damage and increased (10a) 68 kg for green leaf productivity compare to non treatment shade net. The colors of shade net treatments for reduction freezing damaged in order of Black < Green < Colorless but reduction of freezing damage was so high. In the relative microclimates, treatment shade nets were $0.7^{\circ}C$ lower, average relatively humidity 14.9% higher, soil temperature $0.6^{\circ}C$ lower and soil moisture 4.6% higher than non treatment shade net. And treatment shade nets decreased average wind speed 0.7 m/s and it showed us treatment net shades effected to excessive evapotranspiration and soil dry by wind and considered one of good solution for freezing damages.

• Tunnel and Underground Space
• /
• v.21 no.6
• /
• pp.460-470
• /
• 2011

Wave velocity was measured to define the weathering characteristics of rock and the strength evaluation of weathered rock on a target of the Aso gravestones with various sizes under the natural environment. As a result, the size correction method which was changed sample of the different size to one of the same size for evaluating wave velocity was proposed, and also suggested the NET (Normalized Elapsed Time) as a new weathering index of rock. In addition, the strength of the weathered rock was estimated from the weathering classification of rock using the NET. Wave velocity of welded tuff was high and didn't show velocity degradation, on the other hand, one of andesite was low and showed velocity degradation. The degree of weathering between rocks of the different size is considered to be comparable, applying the NET based on the on the $V_p/V_o$-NET curve. Furthermore, the classification of rock weathering stages using the NET based on the $S_c/S_o$-NET curve was available, and the estimation of strength for the weathered rock was also possible.