• International Journal of Computer Science & Network Security
• /
• v.22 no.10
• /
• pp.11-16
• /
• 2022

Applying predictive analytics to predict software defects has improved the overall quality and decreased maintenance costs. Many supervised and unsupervised learning algorithms have been used for defect prediction on publicly available datasets. Most of these datasets suffer from an imbalance in the output classes. We study the impact of class imbalance in the defect datasets on the efficiency of the defect prediction model and propose a CPP method for handling imbalances in the dataset. The performance of the methods is evaluated using measures like Matthew's Correlation Coefficient (MCC), Recall, and Accuracy measures. The proposed sampling technique shows significant improvement in the efficiency of the classifier in predicting defects.

• Journal of Information Processing Systems
• /
• v.19 no.1
• /
• pp.98-108
• /
• 2023

This paper is an attempt to design segmentation method based on fully convolutional networks (FCN) and attention mechanism. The first five layers of the Visual Geometry Group (VGG) 16 network serve as the coding part in the semantic segmentation network structure with the convolutional layer used to replace pooling to reduce loss of image feature extraction information. The up-sampling and deconvolution unit of the FCN is then used as the decoding part in the semantic segmentation network. In the deconvolution process, the skip structure is used to fuse different levels of information and the attention mechanism is incorporated to reduce accuracy loss. Finally, the segmentation results are obtained through pixel layer classification. The results show that our method outperforms the comparison methods in mean pixel accuracy (MPA) and mean intersection over union (MIOU).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.861-880
• /
• 2023

Semantic segmentation of road scene is the key technology of autonomous driving, and the improvement of convolutional neural network architecture promotes the improvement of model segmentation performance. The existing convolutional neural network has the simplification of learning knowledge and the complexity of the model. To address this issue, we proposed a road scene semantic segmentation algorithm based on multi-task collaborative learning. Firstly, a depthwise separable convolution atrous spatial pyramid pooling is proposed to reduce model complexity. Secondly, a collaborative learning framework is proposed involved with saliency detection, and the joint loss function is defined using homoscedastic uncertainty to meet the new learning model. Experiments are conducted on the road and nature scenes datasets. The proposed method achieves 70.94% and 64.90% mIoU on Cityscapes and PASCAL VOC 2012 datasets, respectively. Qualitatively, Compared to methods with excellent performance, the method proposed in this paper has significant advantages in the segmentation of fine targets and boundaries.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.30-33
• /
• 2020

최근 엣지 컴퓨팅과 같은 임베디드 디바이스에서 CNN과 같은 딥러닝 모듈을 수행하기 위해서 하드웨어 설계 및 구현이 많이 진행되고 있다. 이러한 임베디드 시스템에 필요한 CNN모듈을 위한 하드웨어 설계를 위해서 먼저 모델링을 통해서 시뮬레이션이 필요하다. 본 논문에서는 오픈 라이센스를 이용한 RISC-V로 딥러닝 시뮬레이터를 제작하였다. SystemC로 구현된 RISC-V를 Virtual Platform로 시뮬레이터의 제작을 하여 시뮬레이팅을 하였고, SystemC의 특징인 모듈화와 모듈간 통신에 유의하여 시스템을 구성하였다. CNN 알고리즘을 참조하여 Convolution, Activation, Pooling 연산의 기능을 하는 시스템을 구성하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.561-562
• /
• 2021

CNN (Convolutional Neural Network) is widely used to recognize various images. In this presentation, a single digit handwritten by humans was recognized by applying the CNN technique of deep learning. The deep learning network consists of a convolutional layer, a pooling layer, and a platen layer, and finally, we set an optimization method, learning rate and loss functions.

• ETRI Journal
• /
• v.46 no.3
• /
• pp.526-537
• /
• 2024

Monitoring fetal growth in utero is crucial to anomaly diagnosis. However, current computer-vision models struggle to accurately assess the key metrics (i.e., head circumference and occipitofrontal and biparietal diameters) from ultrasound images, largely owing to a lack of training data. Mitigation usually entails image augmentation (e.g., flipping, rotating, scaling, and translating). Nevertheless, the accuracy of our task remains insufficient. Hence, we offer a U-Net fetal head measurement tool that leverages a hybrid Dice and binary cross-entropy loss to compute the similarity between actual and predicted segmented regions. Ellipse-fitted two-dimensional ultrasound images acquired from the HC18 dataset are input, and their lower feature layers are reused for efficiency. During regression, a novel region of interest pooling layer extracts elliptical feature maps, and during segmentation, feature pyramids fuse field-layer data with a new scale attention method to reduce noise. Performance is measured by Dice similarity, mean pixel accuracy, and mean intersection-over-union, giving 97.90%, 99.18%, and 97.81% scores, respectively, which match or outperform the best U-Net models.

• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.13-28
• /
• 2018

The ministry of Environment of Korea initiated two follow-up surveys in 2005 and 2006 to investigate environmental effect on children's health. These two cohorts, referred to as the 2005 Cohort and 2006 Cohort, were followed up three times every two years. This data set was referred to as the Children's Health and Environmental Research (CHEER) data set. This paper reproduces the existing research results of Kim et al. (Journal of the Korean Data and Information Science Society, 25, 987-998, 2014) and Lee et al. (The Korean Journal of Applied Statistics, 29, 1295-1310, 2016) and derive a benchmark dose lower limit (BMDL) for blood lead level for attention deficit hyperactivity disorder (ADHD) after pooling two cohort data sets. The different ADHD rating scales were unified by applying the conversion formula proposed by Lee et al. (2016). The random effect model and AR(1) model were built to reflect the longitudinal characteristics and regression to the mean phenomenon. Based on these models the BMDLs for blood lead levels were derived using the BMDL formula and the simulation. We obtained a hight level of BMDLs when we pooled two independent cohort data sets.

• The Korean Journal of Physiology
• /
• v.24 no.1
• /
• pp.51-65
• /
• 1990

The purpose of the present study was to examine the hemodynamic responses, especially in arterial and skin blood flows, in conjunction with the changes of plasma catecholamine levels as an indirect marker of adrenergic tone during the early stage of head-down tilt (HDT), and to evaluate the early physiological regulatory mechanism in simulated weightlessness. Ten mongrel dogs, weighing8\;{\sim}\;14\;kg, were intravenously anesthetized with nembutal, and postural changes were performed by using the tilting table. The postural changes were performed in the following order: supine, prone, HDT $(-6^{\circ}C)$ and lastly recovery prone position. The duration of each position was 30 minutes. The measurements were made before, during and after each postural change. The arterial blood flow $({\.{Q}})$ at the left common carotid and right brachial arteries was measured by the electromagnetic flowmeter. Blood pressure (BP) was directly measured by pressure transducer in the left brachial artery. To evaluate the peripheral blood flow, skin blood flow $({\.{Q}})$ was calculated by the percent changes of photoelectric pulse amplitude on the forepaw, and skin temperature was recorded. The peripheral vascular resistance (PR) was calculated by dividing respective mean BP values by ${\.{Q}}$ of both sides of common carotid and brachial arteries. Heart rate (HR), respiratory rate (f) and PH, $Po_{2},\;Pco_{2}$ and hematocrit of arterial and venous blood were also measured. The concentration of plasma epinephrine and norepinephrine was measured by radioenzymatic method. The results are summarized as follows: Tilting to head-down position from prone position, HR was initially increased (p<0.05) and BP was not significantly changed. While ${\.{Q}}$ of the common carotid artery was decreased (p<0.05) and PR through the head was increased, ${\.{Q}}$ of the brachial artery was increased (p<0.05) and PR through forelimbs was decreased. ${\.{Q}}$ of the forepaw was initially increased (p<0.05) and then slightly decreased, on the whole revealing an increasing trend. Plasma norepinephrine was slightly decreased and the epinephrine was slightly increased. f was increased and arterial pH was increased (p<0.05). In conclusion, the central blood pooling during HDT shows an increased HR via Bainbridge reflex and an increased ${\.{Q}}$ of the forepaw and brachial ${\.{Q}}$, due to decreased PR which may be originated from the depressor reflex of cardiopulmonary baroreceptors. It is suggested that the blood flow to the brain was adequately regulated throughout HDT $(-6^{\circ}C)$ in spite of central blood pooling. And it is apparent that the changes of plasma norepinephrine level are inversely proportional to those of ${\.{Q}}$ of the forepaw, and the changes of epinephrine level are paralleled with those of the brachial ${\.{Q}}$.

• Journal of Navigation and Port Research
• /
• v.36 no.10
• /
• pp.895-904
• /
• 2012

Today, the competition for hub-port is getting fierce and the shipping liners have enjoyed the increased bargaining power over the terminal operators through the mergers & acquisitions (M&A) and strategic alliances. This result leads the competition among terminal operators to attract liner companies and cargoes in their terminals. In demand side, however, there is a limited container cargo volume to handle because of a steady growth of cargo traffic. While, in supply side, continuous development of port terminals increased more competition among ports or terminals for cargoes. In particular the terminal operating market of Busan port is distorted because of the cargo competition between Busan North-port and Newport. The main purpose of this study is to suggest the stabilization measures of container terminal operating market in Busan port through analysis of the terminal operation market structures and market survey analysis method. For stabilizing the container terminal market, this study suggests the improvement of the legal and institutional system such as improvement in determining and reporting system of stevedoring tariff, establishment of fair competition rules etc., the introduction of port pooling system and adoption of volume-linked terminal lease system with cargo volume ceiling system for each terminal operator.

• Journal of the Korean Society of Radiology
• /
• v.14 no.1
• /
• pp.39-44
• /
• 2020

The purpose of this study was learning about chest X-ray image classification and accuracy research through Deep Learning using big data technology with Convolution Neural Network. Normal 1,583 and Pneumonia 4,289 were used in chest X-ray images. The data were classified as train (88.8%), validation (0.2%) and test (11%). Constructed as Convolution Layer, Max pooling layer size 2×2, Flatten layer, and Image Data Generator. The number of filters, filter size, drop out, epoch, batch size, and loss function values were set when the Convolution layer were 3 and 4 respectively. The test data verification results showed that the predicted accuracy was 94.67% when the number of filters was 64-128-128-128, filter size 3×3, drop out 0.25, epoch 5, batch size 15, and loss function RMSprop was 4. In this study, the classification of chest X-ray Normal and Pneumonia was predictable with high accuracy, and it is believed to be of great help not only to chest X-ray images but also to other medical images.