• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.165-167
• /
• 2014

In this paper, an packet drop technique is proposed to outperform the previous RIO (RED In and Out) drop mechanism for DiffServ in ship area networks. the proposed packet drop technique does not manage the individual flows and divides them into several flow groups according to a criterion. And it guarantees the fairness between individual flows in the same QoS class through the group-based control. In simulation results of the proposed packet drop technique, the link utilization decreases than RIO. But it guarantees more data rates to DiffServ flows passing multiple bottleneck links.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.4
• /
• pp.145-151
• /
• 2021

In this paper, we propose a novel visualization technique to explain the predictions of deep neural networks. We use knowledge distillation (KD) to identify the interior of a black-box model for which we know only inputs and outputs. The information of the black box model will be transferred to a white box model that we aim to create through the KD. The white box model will learn the representation of the black-box model. Second, the white-box model generates attention maps for each of its layers using Grad-CAM. Then we combine the attention maps of different layers using the pixel-wise summation to generate a final saliency map that contains information from all layers of the model. The experiments show that the proposed technique found important layers and explained which part of the input is important. Saliency maps generated by the proposed technique performed better than those of Grad-CAM in deletion game.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.1
• /
• pp.65-70
• /
• 2023

In this paper, we used a deep learning model to detect and segment flame and smoke in real time from fires. To this end, well known U-NET was used to separate and divide the flame and smoke of the fire using multi-class. As a result of learning using the proposed technique, the values of loss error and accuracy are very good at 0.0486 and 0.97996, respectively. The IOU value used in object detection is also very good at 0.849. As a result of predicting fire images that were not used for learning using the learned model, the flame and smoke of fire are well detected and segmented, and smoke color were well distinguished. Proposed method can be used to build fire prediction and detection system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.641-643
• /
• 2023

Deep Learning(DL)을 사용한 호흡음의 자동 분석은 폐 질환의 조기 진단에 중추적인 역할을 한다. 그러나 현재의 DL 방법은 종종 호흡음의 공간적 및 시간적 특성을 분리하여 검사하기 때문에 한계가 있다. 본 연구는 컨볼루션 연산을 통해 공간적 특징을 캡처하고 시간 컨볼루션 네트워크를 사용하여 이러한 특징의 공간적-시간적 상관 관계를 활용하는 새로운 DL 프레임워크를 제한한다. 제안된 프레임워크는 앙상블 학습 접근법 내에 컨볼루션 네트워크를 통합하여 폐음 녹음에서 호흡 이상 및 질병을 검출하는 정확도를 크게 향상시킨다. 잘 알려진 ICBHI 2017 챌린지 데이터 세트에 대한 실험은 제안된 프레임워크가 호흡 이상 및 질병 검출을 위한 4-Class 작업에서 비교모델 성능보다 우수함을 보여준다. 특히 민감도와 특이도를 나타내는 점수 메트릭 측면에서 최대 45.91%와 14.1%의 개선이 이진 및 다중 클래스 호흡 이상 감지 작업에서 각각 보여준다. 이러한 결과는 기존 기술보다 우리 방법의 두드러진 이점을 강조하여 호흡기 의료 기술의 미래 혁신을 주도할 수 있는 잠재력을 보여준다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.49-54
• /
• 2024

Data imbalance is a common issue encountered in classification problems, stemming from a significant disparity in the number of samples between classes within the dataset. Such data imbalance typically leads to problems in classification models, including overfitting, underfitting, and misinterpretation of performance metrics. Methods to address this issue include resampling, augmentation, regularization techniques, and adjustment of loss functions. In this paper, we focus on loss function adjustment, particularly comparing the performance of various configurations of loss functions (Cross Entropy, Balanced Cross Entropy, two settings of Focal Loss: 𝛼 = 1 and 𝛼 = Balanced, Asymmetric Loss) on Multi-Class black-box video data with imbalance issues. The comparison is conducted using the I3D, and R3D_18 models.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.2
• /
• pp.135-140
• /
• 2009

In this paper, we synthesized active sonar target signals based on highlights model, and then carried out target classification using the synthesized signals. If the target aspect angle is changed, the different signals are synthesized. To know the result, two different experiments are done. First, The classification results with respect to each aspect angle are shown. Second, the results in two group in aspect angle are acquired. Time domain feature extraction is done using matched filter and envelope detection. It shows the pattern of each highlights. Artificial neural networks and multi-class SVM are used for classifying target signals.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_3
• /
• pp.1975-1984
• /
• 2021

Urban expansion results in raising the temperature in the city, which can cause social, economic and physical damage. In order to prevent the urban heat island and reduce the urban land surface temperature, it is important to quantify the cooling effect of the features of the urban space. Therefore, in order to understand the relationship between each object of land cover and the land surface temperature in Seoul, the land cover map was classified into 6 classes. And the correlation and multiple regression analysis between land surface temperature and the area of objects, perimeter/area, and normalized difference vegetation index was analyzed. As a result of the analysis, the normalized difference vegetation index showed a high correlation with the land surface temperature. Also, in multiple regression analysis, the normalized difference vegetation index exerted a higher influence on the land surface temperature prediction than other coefficients. However, the explanatory power of the derived models as a result of multiple regression analysis was low. In the future, if continuous monitoring is performed using high-resolution MIR Image from KOMPSAT-3A, it will be possible to improve the explanatory power of the model. By utilizing the relationship between such various land cover types considering vegetation vitality of green areas with that of land surface temperature within urban spaces for urban planning, it is expected to contribute in reducing the land surface temperature in urban spaces.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.27-35
• /
• 2023

In this paper, we propose a process of increasing productivity by applying a deep learning-based defect detection and classification system to the prepreg fiber manufacturing process, which is in high demand in the field of producing composite materials. In order to apply it to toe prepreg manufacturing equipment that requires a solution due to the occurrence of a large amount of defects in various conditions, the optimal environment was first established by selecting cameras and lights necessary for defect detection and classification model production. In addition, data necessary for the production of multiple classification models were collected and labeled according to normal and defective conditions. The multi-classification model is made based on CNN and applies pre-learning models such as VGGNet, MobileNet, ResNet, etc. to compare performance and identify improvement directions with accuracy and loss graphs. Data augmentation and dropout techniques were applied to identify and improve overfitting problems as major problems. In order to evaluate the performance of the model, a performance evaluation was conducted using the confusion matrix as a performance indicator, and the performance of more than 99% was confirmed. In addition, it checks the classification results for images acquired in real time by applying them to the actual process to check whether the discrimination values are accurately derived.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.1
• /
• pp.7-14
• /
• 2010

In this paper, we introduce a hierarchical internet application traffic classification system based on SVM as an alternative overcoming the uppermost limit of the conventional methodology which is using the port number or payload information. After selecting an optimal attribute subset of the bidirectional traffic flow data collected from the campus, the proposed system classifies the internet application traffic hierarchically. The system is composed of three layers: the first layer quickly determines P2P traffic and non-P2P traffic using a SVM, the second layer classifies P2P traffics into file-sharing, messenger, and TV, based on three SVDDs. The third layer makes specific classification of the entire 16 application traffics. By classifying the internet application traffic finely or coarsely, the proposed system can guarantee an efficient system resource management, a stable network environment, a seamless bandwidth, and an appropriate QoS. Also, even a new application traffic is added, it is possible to have a system incremental updating and scalability by training only a new SVDD without retraining the whole system. We validate the performance of our approach with computer experiments.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.7
• /
• pp.209-220
• /
• 2014

In this paper, we analyze the discriminability of features for the classification of transient signals with an ambient noise in a shallow water. For the classification of the transient signals, robust features for the variance of a noise are required due to a low SNR under a marine environment. In the modelling the ambient noise in shallow water, theoretical noise model, Wenz's observation data from the shallow water, and Yule-walker filter are used. Discrimination of each feature of the transient signals with an additive ambient noise is analyzed by utilizing a Fisher score. As the analysis of a classification accuracy about the transient signals of 24 classes using the selected features with a high discriminability, the features selected in the environment without a noise relatively have a good classification accuracy. From the analyzed results, we finally select a total 16 features out of 28 features. The recognition using the selected features results in the classification accuracy of 92% in SNR 20dB using Multi-class SVM.