• Journal of Korea Multimedia Society
• /
• v.23 no.7
• /
• pp.803-811
• /
• 2020

This paper proposes an image classification algorithm that transforms the number of convolution layers in the residual block of ResNet, CNN's representative method. The proposed method modified the structure of 34/50 layer of ResNet structure. First, we analyzed the performance of small and many convolution layers for the structure consisting of only shortcut and 3 × 3 convolution layers for 34 and 50 layers. And then the performance was analyzed in the case of small and many cases of convolutional layers for the bottleneck structure of 50 layers. By applying the results, the best classification method in the residual block was applied to construct a 34-layer simple structure and a 50-layer bottleneck image classification model. To evaluate the performance of the proposed image classification model, the results were analyzed by applying to the cifar10 dataset. The proposed 34-layer simple structure and 50-layer bottleneck showed improved performance over the ResNet-110 and Densnet-40 models.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.798-801
• /
• 2002

This paper proposes unsupervised classification of multiple attributes via five-layer autoassociative neural network with bottleneck layer. In the conventional methods, high dimensional data are compressed into low dimensional data at bottleneck layer and then feature extraction is performed (Fig.1). In contrast, in the proposed method, analog data is compressed into digital data. Furthermore bottleneck layer is divided into two segments so that each attribute, which is a discrete value, is extracted in corresponding segment (Fig.2).

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.493-500
• /
• 2021

In this paper, we propose the development of deep learning structure to improve quality of polygonal containers. The deep learning structure consists of a convolution layer, a bottleneck layer, a fully connect layer, and a softmax layer. The convolution layer is a layer that obtains a feature image by performing a convolution 3x3 operation on the input image or the feature image of the previous layer with several feature filters. The bottleneck layer selects only the optimal features among the features on the feature image extracted through the convolution layer, reduces the channel to a convolution 1x1 ReLU, and performs a convolution 3x3 ReLU. The global average pooling operation performed after going through the bottleneck layer reduces the size of the feature image by selecting only the optimal features among the features of the feature image extracted through the convolution layer. The fully connect layer outputs the output data through 6 fully connect layers. The softmax layer multiplies and multiplies the value between the value of the input layer node and the target node to be calculated, and converts it into a value between 0 and 1 through an activation function. After the learning is completed, the recognition process classifies non-circular glass bottles by performing image acquisition using a camera, measuring position detection, and non-circular glass bottle classification using deep learning as in the learning process. In order to evaluate the performance of the deep learning structure to improve quality of polygonal containers, as a result of an experiment at an authorized testing institute, it was calculated to be at the same level as the world's highest level with 99% good/defective discrimination accuracy. Inspection time averaged 1.7 seconds, which was calculated within the operating time standards of production processes using non-circular machine vision systems. Therefore, the effectiveness of the performance of the deep learning structure to improve quality of polygonal containers proposed in this paper was proven.

• Journal of Korean Vacuum Science & Technology
• /
• v.4 no.1
• /
• pp.27-32
• /
• 2000

We have studied the carrier relaxation of InAs/GaAs modulation-doped quantum dots depending on the excitation wavelength and modulation-doping concentration by using the time-ressolved spectroscopy. At the excitation below GaAs barrier band gap, the relaxation processes become very slow, implying to observe the phonon bottleneck effects. On the other hand, at the excitation far above GaAs band gap, phonon bottleneck effects are broken down due to Auger processes. Increasing modulation-doping concentration, the relaxation times, by virtue of Coulomb scattering between electrons in GaAs doped layer and carriers in InAs quantum dots, are observed to become fast.

• International Journal of Computer Science & Network Security
• /
• v.22 no.3
• /
• pp.292-302
• /
• 2022

The radar tomographic imaging is based on the Radar Cross-Section "RCS" of the materials of a shape under examination and investigation. The RCS varies as the conductivity and permittivity of a target, where the target has a different material profile than other background objects in a scene. In this research paper, we use Hierarchical Performance Modeling "HPM" and a framework developed earlier to determine/spot bottleneck(s) for pattern recognition of materials using a combination of the Single Layer Perceptron (SLP) technique and tomographic images in radar systems. HPM provides mathematical equations which create Objective Functions "OFs" to find an average performance metric such as throughput or response time. Herein, response time is used as the performance metric and during the estimation of it, bottlenecks are found with the help of OFs. The obtained results indicate that processing images consumes around 90% of the execution time.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1835-1845
• /
• 2021

To accurately identify black ice and warn the drivers of information in advance so they can control speed and take preventive measures. In this paper, we propose a lightweight black ice detection network based on infrared road images. A black ice recognition network model based on CNN transfer learning has been developed. Additionally, to further improve the accuracy of black ice recognition, an enhanced lightweight network based on MobileNetV2 has been developed. To reduce the amount of calculation, linear bottlenecks and inverse residuals was used, and four bottleneck groups were used. At the same time, to improve the recognition rate of the model, each bottleneck group was connected to a 3×3 convolutional layer to enhance regional feature extraction and increase the number of feature maps. Finally, a black ice recognition experiment was performed on the constructed infrared road black ice dataset. The network model proposed in this paper had an accurate recognition rate of 99.07% for black ice.

• ETRI Journal
• /
• v.34 no.1
• /
• pp.102-105
• /
• 2012

To serve the growing demand of the bidirectional information exchange, we propose a single relay selection (RS) scheme for physical-layer network coding (PNC) in a bidirectional cooperative network consisting of two sources and multiple relays. This RS scheme selects a single best relay by maximizing the bottleneck of the capacity region of both information flows in the bidirectional network. We show that the proposed RS rule minimizes the outage probability and that it can be used as a performance benchmark for any RS rules with PNC. We derive a closed-form exact expression of the outage probability for the proposed RS rule and show that it achieves full diversity gain. Finally, numerical results demonstrate the validity of our analysis.

• Journal of the Korean Ceramic Society
• /
• v.46 no.2
• /
• pp.161-169
• /
• 2009

MLCC (Multi-layer Ceramic Capacitor) is the most important passive component in electronic devices such as HHP, PC and digital display. The development trend of MLCC is a miniaturization with increasing the capacitance. In this paper, a development history of the high capacitance MLCC in Korea was introduced, and the necessity of the finer $BaTiO_3$ was explained in the viewpoint of the issued electrical and dielectric properties of high capacitance MLCC. The bottleneck technologies to realize the high capacitance was shortly introduced, followed by the prediction of the development trend of MLCC in near future.

• IE interfaces
• /
• v.10 no.1
• /
• pp.1-13
• /
• 1997

반도체 Wafer Fabrication 공정은 무수한 공정과 복잡한 Lot의 흐름 등으로 다른 제조 형태에 비해 효율적인 관리가 대단히 어려운 부문이다. 본 연구는 반도체 Fab을 대상으로 주어진 생산 소요량과 목표 공기를 효율적으로 달성하기 위한 Shift 단위의 생산 일정계획을 대상으로 하였다. 특히, 전 공정 및 장비를 고려하기보다는 Bottleneck인 Photo 공정의 Stepper를 중심으로, 공정을 Layer단위로 묶어, 한 Shift에서 어떻게 Stepper를 할당하고 생산계획을 할 것인가를 결정하기 위한 2단계 방법론을 제시하고, Stepper 할당 및 계획에 필요한 3가지 알고리즘들을 제시하였다. 이 기법들을 소규모의 예제들에 대해 적용한 결과와 최적해와의 비교를 통하여 그 성능을 평가하였다.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.8
• /
• pp.11-19
• /
• 2019

Against advanced image-related crimes, a high level of digital forensic methods is required. However, feature-based methods are difficult to respond to new device features by utilizing human-designed features, and deep learning-based methods should improve accuracy. This paper proposes a deep learning model to identify camera models based on DenseNet, the recent technology in the deep learning model field. To extract camera sensor features, a HPF feature extraction filter was applied. For camera model identification, we modified the number of hierarchical iterations and eliminated the Bottleneck layer and compression processing used to reduce computation. The proposed model was analyzed using the Dresden database and achieved an accuracy of 99.65% for 14 camera models. We achieved higher accuracy than previous studies and overcome their disadvantages with low accuracy for the same manufacturer.