• Journal of Convergence for Information Technology
• /
• v.12 no.2
• /
• pp.47-53
• /
• 2022

Recently, with the change of the intelligent security paradigm, study to apply various information generated from various information security systems to AI-based anomaly detection is increasing. Therefore, in this study, in order to convert log-like time series data into a vector, which is a numerical feature, the CBOW and Skip-gram inference methods of deep learning-based Word2Vec model and statistical method based on the coincidence frequency were used to transform the published ADFA system call data. In relation to this, an experiment was carried out through conversion into various embedding vectors considering the dimension of vector, the length of sequence, and the window size. In addition, the performance of the embedding methods used as well as the detection performance were compared and evaluated through GRU-based anomaly detection model using vectors generated by the embedding model as an input. Compared to the statistical model, it was confirmed that the Skip-gram maintains more stable performance without biasing a specific window size or sequence length, and is more effective in making each event of sequence data into an embedding vector.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.44 no.1
• /
• pp.125-132
• /
• 2024

Optical satellite images are being used for national security and collection of information, and their utilization is increasing. However, it acquires low-quality images that are not suitable for the user's requirement due to weather conditions and time constraints. In this paper, a deep learning-based conversion of image and colorization model referring to high-resolution SAR images was created to simulate the occluded area with clouds of optical satellite images. The model was experimented according to the type of algorithm applied and input data, and each simulated images was compared and analyzed. In particular, the amount of pixel value information between the input black-and-white image and the SAR image was similarly constructed to overcome the problem caused by the relatively lack of color information. As a result of the experiment, the histogram distribution of the simulated image learned with the Gray-scale image and the high-resolution SAR image was relatively similar to the original image. In addition, the RMSE value was about 6.9827 and the PSNR value was about 31.3960 calculated for quantitative analysis.

• Journal of Internet Computing and Services
• /
• v.23 no.5
• /
• pp.145-154
• /
• 2022

Automatic Target Recognition (ATR) technology is emerging as a core technology of Future Combat Systems (FCS). Conventional ATR is performed based on IMINT (image information) collected from the SAR sensor, and various image-based deep learning models are used. However, with the development of IT and sensing technology, even though data/information related to ATR is expanding to HUMINT (human information) and SIGINT (signal information), ATR still contains image oriented IMINT data only is being used. In complex and diversified battlefield situations, it is difficult to guarantee high-level ATR accuracy and generalization performance with image data alone. Therefore, we propose a knowledge graph-based ATR method that can utilize image and text data simultaneously in this paper. The main idea of the knowledge graph and deep model-based ATR method is to convert the ATR image and text into graphs according to the characteristics of each data, align it to the knowledge graph, and connect the heterogeneous ATR data through the knowledge graph. In order to convert the ATR image into a graph, an object-tag graph consisting of object tags as nodes is generated from the image by using the pre-trained image object recognition model and the vocabulary of the knowledge graph. On the other hand, the ATR text uses the pre-trained language model, TF-IDF, co-occurrence word graph, and the vocabulary of knowledge graph to generate a word graph composed of nodes with key vocabulary for the ATR. The generated two types of graphs are connected to the knowledge graph using the entity alignment model for improvement of the ATR performance from images and texts. To prove the superiority of the proposed method, 227 documents from web documents and 61,714 RDF triples from dbpedia were collected, and comparison experiments were performed on precision, recall, and f1-score in a perspective of the entity alignment..

• Korean Journal of Poultry Science
• /
• v.37 no.1
• /
• pp.51-62
• /
• 2010

This study was conducted to investigate the effects of dietary supplementation of multiple probiotics on the performance, small intestinal microflora and immune response in laying hens and broilers. In Exp.1, a total of 800, 82 wk old Hy-line Brown$^{(R)}$ laying hens were assigned to one of the following five dietary treatment; Control, Antibiotics (avilamycin 6 ppm), Probiotics; PB-M (Micro-ferm$^{(R)}$ 0.2%), PB-L (Lacto-sacc$^{(R)}$ 0.1%), PB-Y (Y University probiotics 0.2%). Each treatment was replicated eight times with 20 birds in each replicate and two birds were housed in each cage. Twenty birds units were arranged according to completely randomized block design. Feeding trial lasted 6 wk under 16 h lighting regimen. The Exp. 2, was conducted with a total of 1,000 broilers chicks (Ross$^{(R)}$). They were divided into five treatments, same as those of Exp. 1. Birds were fed starter (0~3 wk) and grower (4~5 wk) diets. Each treatment was replicated four times with 50 birds per pen comprising of deep litter. In Exp. 1, egg production parameters, such as hen-day and hen-house egg production, egg weight, broken and soft shell egg production, feed intake and feed conversion were not significantly different among treatments. However, strength and thickness of eggshell were significantly (P<0.05) different. Among the probiotics, PB-Y showed the highest strength and thickness of eggshell. Eggshell color, egg yolk color and Haugh unit were not significantly influenced. In Exp. 2, overall weight gain (0~5 wk) and mortality were not significantly different among treatments. However, weight gain of birds from PB-Y treatment during starter (0~3 wk) was significantly lower than the birds from Control and Antibiotic treatment. During the whole period (0~5 wk), birds from Antibiotics treatment had higher feed intake and Production Index (PI) and lower feed conversion than birds from Control treatment. Probiotics treatments were not significantly different from the Control on feed intake and feed conversion. In Exp.1, there were significant (P<0.05) differences in leukocytes parameters, such as white blood cell (WBC), hetrophil (HE), lymphocytes (LY), monocyte (MO), eosinophil (EO) and stress index (SI; HE/LY) in the blood of layers. Birds from Antibiotics and probiotics treatments tended to increase these parameters. In Exp. 2, however, only SI was significantly (P<0.05) decreased in Antibiotics treatments. Concentration of serum immunoglobulin (IgG) were higher (P<0.05) in PB-M and PB-Y treatments when compared with Control treatment in Exp. 1. The population of E. coli significantly (P<0.05) decreased in birds from Antibiotics, PB-L and PB-Y treatments when compared with birds from Control treatment in Exp. 1. Metalbolizability of crude fat decreased significantly (P<0.05) in birds from probiotic treatments in Exp. 2. It was concluded that the response of probiotics on the productivity of layers and broilers were different. Probiotics increased strength and thickness of eggshell in layers, and decreased feed conversion and increased PI in broilers. Leukocytes and IgG tended to increase by supplementation of antibiotics and probiotics in layers. Intestinal E. coli tended to decrease in layers. Digestibility of crude fat of diet decreased in probiotics treatments broilers. Parameters of blood and microbial were more sensitive in layers than broilers.

• Ocean and Polar Research
• /
• v.34 no.4
• /
• pp.365-384
• /
• 2012

Microalgae are photosynthetic microorganisms that are highly productive in the presence of basic renewable natural sources (light, $CO_2$, water and nutrients). They can synthesize lipids, carbohydrates and proteins in a small number of days. Subsequently, these carbon-captured products can be processed into both biofuels and valuable co-products. Additionally, microalgae would be an ideal feedstock for replacing land-based food crops with cellular products as high energy density transportation fuels. These microscopic organisms could contribute a significant amount of renewable energy on a global scale. In Korea, microalgae biofuel research was common in the early 1990s. The research activities were unfortunately stopped due to limited governmental funds and low petroleum prices. Interest in algal biofuels in Korea has been growing recently due to an increased concern over oil prices, energy security, greenhouse gas emissions, and the potential for other biofuel feedstock to compete for limited agricultural resources. The high productivity of microalgae suggests that much of the Korean transportation fuel requirements can be met by biofuels at a production cost competitive with the increasing cost of petroleum seen in early 2008. At this time, the development of microlalgal biomass production technology remains in its infancy. This study reviewed microalgae culture systems and biomass production, harvesting, oil extraction, conversion, and technoeconomical bottlenecks. Many technical and economic barriers to using microalgal biofuels need to be overcome before mass production of microalgal-derived fuel substitutes is possible. However, serious efforts to overcome these barriers could become a large-scale commercial reality. Overall, this study provides a brief overview of the past few decades of global microalgal research.

• Asian-Australasian Journal of Animal Sciences
• /
• v.16 no.10
• /
• pp.1501-1509
• /
• 2003

The simultaneous addition of xylanase (5,600 EXU/kg) and phytase (500 FTU/kg) feed enzymes to wheat-based broiler diets was investigated. Starter, grower and finisher diets, with three tiers of nutrient specifications, were fed to 1,440 broiler chicks kept on deep litter from 1-42 days post-hatch, without and with xylanase plus phytase, to determine the effects of diet type and enzyme supplementation on growth performance. The nutrient specifications of type A diets were standard; energy density and protein/amino acid levels were reduced on a least-cost basis to formulate type B diets and further reduced to type C diets. Phosphorus (P) and calcium (Ca) levels were adjusted in supplemented diets. From 1-42 days post-hatch, diet type significantly influenced growth performance. Birds on type C diets had lower growth rates (2,429 vs. 2,631 g/bird; p<0.001), higher feed intakes (4,753 vs. 4,534 g/bird; p<0.005) and less efficient feed conversion (1.96 vs. 1.72; p<0.001) than birds offered type A diets. Enzyme supplementation increased growth rates by 3.2% (2,580 vs. 2,501 g/bird; p<0.005) and improved feed efficiency by 2.7% (1.80 vs. 1.85; p<0.05) over the entire feeding period. There were no interactions between diet type and enzyme supplementation. At 21 days, 5 out of 30 birds per pen were transferred to cages to ascertain treatment effect on apparent metabolisable energy (AME) and nitrogen (N) retention. Xylanase plus phytase enhanced AME (13.48 to 13.91 MJ/kg DM; p<0.001) and N retention (56.3 to 59.7%; p<0.005). Carcass and breast weights of the caged birds were determined following commercial processing. Diet type significantly influenced breast weight, carcass weight and yield. Birds offered Type A diets, in comparison to Type C diets, supported heavier breast (467 vs. 424 g; p<0.001) and carcass weights (1,868 vs. 1,699 g; p<0.001) with superior carcass yields (71.8 vs. 70.6%; p<0.005). Enzyme addition increased carcass weight by 3.9% (1,752 vs. 1,821 g; p<0.005) and breast weight by 5.8% (431 vs. 456 g; p<0.01) without influencing yields. Feed ingredient costs per kg live weight gain and per kg carcass weight indicated that enzyme addition was economically feasible, where supplementation of Type A diets generated the most effective results. Importantly, soluble and total non-starch polysaccharide and phytate contents of the wheat used were typical by local standards. This study confirms the potential of supplementing wheat-based broiler diets with xylanase plus phytase but further investigations are required to define the most appropriate inclusion rates and dietary nutrient specifications in this context.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2016.02a
• /
• pp.154.2-154.2
• /
• 2016

In this work, we introduce a solution-processed CdS interlayer for use in inverted bulk heterojunction (BHJ) solar cells, and compare this material to a series of standard organic and inorganic cathode interlayers. Different combinations of solution-processed CdS, ZnO and conjugated polyelectrolyte (CPE) layers were compared as cathode interlayers on ITO substrates to construct inverted solar cells based on $PTB7:PC_{71}BM$ and a $P3HT:PC_{61}BM$ as photoactive layers. Introduction of a CdS interlayer significantly improved the power conversion efficiency (PCE) of inverted $PTB7:PC_{71}BM$ devices from 2.0% to 4.9%, however, this efficiency was still fairly low compared to benchmark ZnO or CPE interlayers due to a low open circuit voltage ($V_{OC}$), stemming from the deep conduction band energy of CdS. The $V_{OC}$ was greatly improved by introducing an interfacial dipole (CPE) layer on top of the CdS layer, yielding outstanding diode characteristics and a PCE of 6.8%. The best performing interlayer, however, was a single CPE layer alone, which yielded a $V_{OC}$ of 0.727 V, a FF of 63.2%, and a PCE of 7.89%. Using $P3HT:PC_{61}BM$ as an active layer, similar trends were observed. Solar cells without the cathode interlayer yielded a PCE of 0.46% with a poor $V_{OC}$ of 0.197 V and FF of 34.3%. In contrast, the use of hybrid ZnO/CPE layer as the cathode interlayer considerably improved the $V_{OC}$ of 0.599 V and FF of 53.3%, resulting the PCE of 2.99%. Our results indicate that the CdS layer yields excellent diode characteristics, however, performs slightly worse than benchmark ZnO and CPE layers in solar cell devices due to parasitic absorption below 550 nm. These results suggest that the hybrid inorganic/organic interlayer materials are promising candidates as cathode interlayers for high efficiency inverted solar cells through the modification of interface contacts.

• Korean Journal of Clinical Laboratory Science
• /
• v.50 no.1
• /
• pp.29-36
• /
• 2018

Dimorphic fungi are characterized by the production of mycelia in the environment or when incubated at lower temperatures ($25{\sim}30^{\circ}C$) on media. The conversion to yeast occurs when invading a human body or when incubated at elevated temperatures ($35{\sim}37^{\circ}C$). Dimorphic fungi cause systemic infection and occur only in endemic regions; they can cause deep mycosis, which invades tissues and organs. Imported thermally dimorphic fungal pathogen has five genera, such as Coccidioides immitis, Histoplasma capsulatum, Blastomyces dermatitidis, Talaromyces (Penicillium) marneffei, and Paracoccidioides brasiliensis. Thus far, 30 cases of dimorphic fungi have been reported in Korea, including C. immitis in 18, H. capsulatum in 5, B. dermatitidis in 3, and T. marneffei in 4 cases. Three 3 cases of C. immitis, 2 cases of H. capsulatum, 1 case of B. dermatitidis, and 1 case of Talaromyces (Penicillium) marneffei were reported in Korea, even though no experience of an epidemic was present. None the dimorphic fungi except for Sporothrix schenkii (Dimorphic fungi) occur in Korea but are caused by endemic fungi that can be found only in specific regions and limited areas of foreign countries. Considering the trends of migratory locals and the increasing number of immigrants living in and returning home, the incidence of imported infections is expected to increase.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.24 no.6
• /
• pp.810-817
• /
• 2018

In Korea, tuna purse seine fishery is the industry with the highest production volume in domestic deep-sea fishery. Research has been continuously carried out since the late 1980s, and research on modernization of technology has been actively conducted. However, while auxiliary works boats on the fishing vessels play an important role in the purse seine fishing, related research is not sufficient. In this study, we tried to rebuild the propulsion system of the Net-boat from the propeller to the waterjet type and to perform the hull form conversion to improve the resistance performance of the hull part and improve the motion performance. For this, we calculated the change of the fluid performance of the ship through computational numerical analysis. In addition, towing tank tests were carried out to verify the performance of the existing ship and the ship which changed the propulsion system. As a result, resistance performance of the waterjet type compared to the propeller type was improved by approximately 45 %. It was confirmed that this was due to reduction of the wetted surface area by removing net and improvements in hull form.

• Journal of Radiation Protection and Research
• /
• v.24 no.4
• /
• pp.215-223
• /
• 1999

Characteristics of radiation field in the steam generator(S/G) water chamber of a PWR were investigated and the anticipated effective dose rates to the worker in the S/G chamber were evaluated by Monte Carlo simulation. The results of crud analysis in the S/G of the Kori nuclear power plant unit 1 were adopted for the source term. The MCNP4A code was used with the MIRD type anthropomorphic sex-specific mathematical phantoms for the calculation of effective doses. The radiation field intensity is dominated by downward rays, from the U-tube region, having approximate cosine distribution with respect to the polar angle. The effective dose rates to adults of nominal body size and of small body size(The phantom for a 15 year-old person was applied for this purpose) appeared to be 36.22 and 37.06 $mSvh^{-1}$) respectively, which implies that the body size effect is negligible. Meanwhile, the equivalent dose rates at three representative positions corresponding to head, chest and lower abdomen of the phantom, calculated using the estimated exposure rates, the energy spectrum and the conversion coefficients given in ICRU47, were 118, 71 and 57 $mSvh^{-1}$, respectively. This implies that the deep dose equivalent or the effective dose obtained from the personal dosimeter reading would be the over-estimate the effective dose by about two times. This justifies, with possible under- or over- response of the dosimeters to radiation of slant incidence, necessity of very careful planning and interpretation for the dosimetry of workers exposed to a non-regular radiation field of high intensity.