• Journal of Bio-Environment Control
• /
• v.18 no.3
• /
• pp.280-284
• /
• 2009

One of the big obstacles to cultivate asparagus was long days taking before first harvesting. This study was carried out to hasten the first harvesting of summer planted asparagus in Jeju. Seedlings were raised for three months and planted June 20th in green house. Harvesting of Spring were separated into non-harvested (control) and harvested (partly-harvesting, completely-harvesting). The first year we could harvest $399kg{\sim}400kg/10a$ in harvesting treatment. Second year's yield was 834kg/10a in partly-harvesting, 825kg/10a in completely-harvesting treatment and 908kg/10a in control. There is no significant difference in second years yield regardless of first year's harvesting methods. The accumulated total yield was increased by 35% (l,229kg/10a) in harvesting treatment from the first spring compared with the control. Marketable yield was increased by 33% (1,116kg/10a) compared with non harvesting in first year (839kg/10a). The result of this study shows that doing harvest of the first year's spring in summer planting asparagus would be desirable for yield and possible to harvest after 8 months planting.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.6
• /
• pp.133-145
• /
• 2023

This paper examines whether investors have adapted to the venture capital(VC) investment style. VC firms invest in privately held companies and generate returns by selling them after the lock-up period expires. We analyze the impact on stock prices before and after the lock-up period expiration, and compare the Cumulative Abnormal Return(CAR) between the past period(2015-2017) and the recent period(2020-2022) to investigate the effect of the second venture boom. The main findings are as follows. First, unlike in the past, stock price returns around the lock-up period expiration have been lower than the KOSDAQ index in recent years. Second, the impact on stock prices is significant for both 1-month and 12-month lock-up periods. Specifically, it is confirmed that stocks held by venture capital and professional investors with a 1-month lock-up period respond in advance to their information after the second venture boom. Finally, we find that there is a difference in CAR depending on whether or not the company received VC investment after the second venture boom. Based on our findings, we suggest that VC firms need to revise their exit strategies to improve performance. This includes finding ways to reduce information asymmetry and fees, as well as developing strategies to mitigate market volatility. Additionally, the current lock-up period for VCs should be reconsidered as it may increase the risk of stock price decline. We recommend that the government revise the scope and duration of lock-up periods to protect investors after IPO.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.4
• /
• pp.789-795
• /
• 2020

Many deaths have been reported due to the worldwide pandemic of COVID-19. In order to prevent the further spread of COVID-19, it is necessary to quickly and accurately read images of suspected patients and take appropriate measures. To this end, this paper introduces a deep learning-based COVID-19 chest X-ray reading technique that can assist in image reading by providing medical staff whether a patient is infected. First of all, in order to learn the reading model, a sufficient dataset must be secured, but the currently provided COVID-19 open dataset does not have enough image data to ensure the accuracy of learning. Therefore, we solved the image data number imbalance problem that degrades AI learning performance by using a Stacked Generative Adversarial Network(StackGAN++). Next, the DenseNet-based classification model was trained using the augmented data set to develop the reading model. This classification model is a model for binary classification of normal chest X-ray and COVID-19 chest X-ray, and the performance of the model was evaluated using part of the actual image data as test data. Finally, the reliability of the model was secured by presenting the basis for judging the presence or absence of disease in the input image using Grad-CAM, one of the explainable artificial intelligence called XAI.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.9
• /
• pp.11-19
• /
• 2009

A RAM-based Neural Net(RBNN) which has multi-discriminators is more effective than RBNN with a discriminator. Experience Sensitive Cumulative Neural Network and 3-D Neuro System(3DNS) that accumulate the features point improved the performance of BNN, which were enabled to train additional and repeated patterns and extract a generalized pattern. In recognition process of Neural Net with multi-discriminator, the selection of class was decided by the value of MRD which calculates the accumulated sum of each class. But they had a saturation problem of its memory cells caused by learning volume increment. Therefore, the decision of MRD has a low performance because recognition rate is decreased by saturation. In this paper, we propose the method which improve the MRD ability. The method consists of the optimum MRD and the matching ratio prototype to generalized image, the cumulative filter ratio, the gap of prototype response MRD. We experimented the performance using NIST database of NIST without preprocessor, and compared this model with 3DNS. The proposed MRD method has more performance of recognition rate and more stable system for distortion of input pattern than 3DNS.

• The Journal of Engineering Geology
• /
• v.24 no.3
• /
• pp.343-351
• /
• 2014

Tunnel face mapping plays an important role in risk analysis and infrastructure support decisions during tunnel construction. In this study, a digital mapping system using a mobile device is employed instead of existing face-mapping methods that rely upon face mapping sheets. The mobile device is then connected to the main server in the field, where a tunnel-specific database is compiled automatically. This information provides real-time feedback on the tunnel face to construction personnel and engineers, thus allowing for rapid assessment of tunnel face stability and infrastructure needs. The Douglas-Peucker algorithm, among others, is employed to resolve problems arising from the detailed mapping and speed problem by data accumulation. This system is expected to raise program optimization through field verification and additional functional improvements.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.3
• /
• pp.163-170
• /
• 2016

Regulations for the exhaust gas of diesel vehicles are being strengthened every year. Recently, diesel emission regulations for HC, CO, NOx, and particulate matter (PM) have been subject to very strict standards. In the future, the regulation of PM is expected to become stricter. Accordingly, diesel particulate filters (DPFs) have been applied to most diesel vehicles for PM reduction. With increasing engine mileage, ash and soot from the engine exhaust gas accumulate inside the DPF. This accumulation can damage the DPF or degrade engine performance. Therefore, efficient cleaning of the DPF is critical for the maintenance of the engine. If the DPF is well managed through regular cleaning, it can improve the power and fuel economy of the engine and reduce maintenance costs. Therefore, this study was performed to develop a high-efficiency cleaning method for DPFs and an apparatus that can more effectively clean out the accumulated ash and soot.

• Journal of the Korean Society for Railway
• /
• v.18 no.3
• /
• pp.241-251
• /
• 2015

Compared with ballasted track (BT), ballasted asphalt track (BAT) has been increasingly adopted in many countries due to its more greatly reduced reinforced roadbed thickness and smaller cumulative plastic deformation, and its advantages in terms of maintenance. In this respect, the authors' previous research includes analysis of BAT sections that show performance similar to that of BT sections of the present specifications; reliability verification of the analysis results through real-sized static and dynamic train-load tests were performed. Based on previous research, this paper estimates the track lifetime using the strain of the lower roadbed according to reinforced roadbed thickness; using probabilistic LCC analysis, this paper presents a BAT section that satisfies the design lifetime and that has performance similar to or higher than that of BT.

• Journal of Korea Water Resources Association
• /
• v.47 no.9
• /
• pp.839-852
• /
• 2014

This study analyzed the applicability of a rainfall downscaling algorithm in space-time multifractal framework (RDSTMF) in Korean Peninsula. To achieve this purpose, the 8 heavy rainfall events that occurred in Korea during the period between 2008 and 2012 were analyzed using the radar rainfall imagery. The result of the analysis indicated that there is a strong tendency of the multifractality for all 8 heavy rainfall events. Based on the multifractal exponents obtained from the analysis, the parameters of the RDSTMF were obtained and the relationship between the average intensity of the rainfall events and the parameters of the RDSTMF was developed. Based on this relationship, the synthetic space-time rainfall fields were generated using the RDSTMF. Then, the generated synthetic space-time rainfall fields were compared to the observation. The result of the comparison indicated that the RDSTMF can accurately reproduce the multifractal exponents of the observed rainfall field up to 3rd order and the cumulative density function of the observed space-time rainfall field with a reasoable accuracy.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.28 no.11
• /
• pp.853-864
• /
• 2017

This paper proposes a false alarm rate reduction technique for detection of small targets in a terrestrial environment. CFAR algorithm is useful in homogeneous background, but it is not easy to detect targets in non-homogeneous background. In particular, when the clutter power is not significantly different from the target signal, it is difficult to detect the target due to high false alarm rate. To solve these difficulties, this study presents the false alarm rate reduction technique based on CFAR algorithm, matched filter and binary integration technique. The parameters are studied through the theoretical analysis and the validity of the proposed study is examined by the test results.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.51-54
• /
• 2006

본 논문에서는 각 사용자들의 취향에 맞는 음악을 추천하는 개인화된 음악 추천 시스템을 소개한다. 추천 시스템이란 사용자의 선호도를 분석하고 아이템들에 대한 사용자의 선호도를 예측하여 영화, 음악, 기사, 책, 웹 페이지 등과 같은 아이템들을 추천하는 시스템을 말한다. 추천 시스템들에서 가장 많이 사용하고 있는 협동적 추천 방식은 선호도 데이터를 기반으로 유사한 사용자들을 찾고, 유사 사용자들의 선호도를 기반으로 예측을 수행하는 것으로서, 여러 장점들이 있으나 희소성(sparsity) 문제와 확장성(scalability) 문제에 대해 취약점을 가지고 있다. 아이템들의 전체 수에 비해 매우 적은 수의 아이템 선호도 데이터만 존재한다면 사용자들의 유사도를 계산하기가 어려우며, 또한 사용자의 수가 늘어날수록 유사도 계산에 걸리는 시간이 급격하게 늘어남으로써 수백만 사용자가 있는 웹 사이트 등에서 실시간 추천을 수행하기 어렵다. 본 논문에서 소개하는 음악 추천 시스템은 이러한 문제점들을 해결하기 위해 그래프 기반 협동적 여과 기법을 사용한다. 그래프 기반 협동적 여과 기법은 기존의 협동적 여과 기법들과 달리 아이템들 사이의 연관관계를 그래프 모델로 표현하고 저장함으로써 묵시적인 선호도 정보들을 누적하여 희소성 문제를 해결하고, 추천 아이템을 선정하는데 필요한 계산 시간을 크게 단축하여 대규모 데이터에서 실시간 추천을 가능하게 한다는 장점이 있다.