• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.191-203
• /
• 2022

In this paper, we developed a system that intelligently identifies skin image data from big data collected from social media Instagram and extracts standardized skin sample data for skin condition diagnosis and management. The system proposed in this paper consists of big data collection and analysis stage, skin image analysis stage, training data preparation stage, artificial neural network training stage, and skin image identification stage. In the big data collection and analysis stage, big data is collected from Instagram and image information for skin condition diagnosis and management is stored as an analysis result. In the skin image analysis stage, the evaluation and analysis results of the skin image are obtained using a traditional image processing technique. In the training data preparation stage, the training data were prepared by extracting the skin sample data from the skin image analysis result. And in the artificial neural network training stage, an artificial neural network AnnSampleSkin that intelligently predicts the skin image type using this training data was built up, and the model was completed through training. In the skin image identification step, skin samples are extracted from images collected from social media, and the image type prediction results of the trained artificial neural network AnnSampleSkin are integrated to intelligently identify the final skin image type. The skin image identification method proposed in this paper shows explain high skin image identification accuracy of about 92% or more, and can provide standardized skin sample image big data. The extracted skin sample set is expected to be used as standardized skin image data that is very efficient and useful for diagnosing and managing skin conditions.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.1
• /
• pp.9-18
• /
• 2024

The importance of active sonar systems is emerging due to the quietness of underwater targets and the increase in ambient noise due to the increase in maritime traffic. However, the low signal-to-noise ratio of the echo signal due to multipath propagation of the signal, various clutter, ambient noise and reverberation makes it difficult to identify underwater targets using active sonar. Attempts have been made to apply data-based methods such as machine learning or deep learning to improve the performance of underwater target recognition systems, but it is difficult to collect enough data for training due to the nature of sonar datasets. Methods based on mathematical modeling have been mainly used to compensate for insufficient active sonar data. However, methodologies based on mathematical modeling have limitations in accurately simulating complex underwater phenomena. Therefore, in this paper, we propose a sonar signal synthesis method based on a deep neural network. In order to apply the neural network model to the field of sonar signal synthesis, the proposed method appropriately corrects the attention-based encoder and decoder to the sonar signal, which is the main module of the Tacotron model mainly used in the field of speech synthesis. It is possible to synthesize a signal more similar to the actual signal by training the proposed model using the dataset collected by arranging a simulated target in an actual marine environment. In order to verify the performance of the proposed method, Perceptual evaluation of audio quality test was conducted and within score difference -2.3 was shown compared to actual signal in a total of four different environments. These results prove that the active sonar signal generated by the proposed method approximates the actual signal.

• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.95-108
• /
• 2017

Recently, AlphaGo which is Bakuk (Go) artificial intelligence program by Google DeepMind, had a huge victory against Lee Sedol. Many people thought that machines would not be able to win a man in Go games because the number of paths to make a one move is more than the number of atoms in the universe unlike chess, but the result was the opposite to what people predicted. After the match, artificial intelligence technology was focused as a core technology of the fourth industrial revolution and attracted attentions from various application domains. Especially, deep learning technique have been attracted as a core artificial intelligence technology used in the AlphaGo algorithm. The deep learning technique is already being applied to many problems. Especially, it shows good performance in image recognition field. In addition, it shows good performance in high dimensional data area such as voice, image and natural language, which was difficult to get good performance using existing machine learning techniques. However, in contrast, it is difficult to find deep leaning researches on traditional business data and structured data analysis. In this study, we tried to find out whether the deep learning techniques have been studied so far can be used not only for the recognition of high dimensional data but also for the binary classification problem of traditional business data analysis such as customer churn analysis, marketing response prediction, and default prediction. And we compare the performance of the deep learning techniques with that of traditional artificial neural network models. The experimental data in the paper is the telemarketing response data of a bank in Portugal. It has input variables such as age, occupation, loan status, and the number of previous telemarketing and has a binary target variable that records whether the customer intends to open an account or not. In this study, to evaluate the possibility of utilization of deep learning algorithms and techniques in binary classification problem, we compared the performance of various models using CNN, LSTM algorithm and dropout, which are widely used algorithms and techniques in deep learning, with that of MLP models which is a traditional artificial neural network model. However, since all the network design alternatives can not be tested due to the nature of the artificial neural network, the experiment was conducted based on restricted settings on the number of hidden layers, the number of neurons in the hidden layer, the number of output data (filters), and the application conditions of the dropout technique. The F1 Score was used to evaluate the performance of models to show how well the models work to classify the interesting class instead of the overall accuracy. The detail methods for applying each deep learning technique in the experiment is as follows. The CNN algorithm is a method that reads adjacent values from a specific value and recognizes the features, but it does not matter how close the distance of each business data field is because each field is usually independent. In this experiment, we set the filter size of the CNN algorithm as the number of fields to learn the whole characteristics of the data at once, and added a hidden layer to make decision based on the additional features. For the model having two LSTM layers, the input direction of the second layer is put in reversed position with first layer in order to reduce the influence from the position of each field. In the case of the dropout technique, we set the neurons to disappear with a probability of 0.5 for each hidden layer. The experimental results show that the predicted model with the highest F1 score was the CNN model using the dropout technique, and the next best model was the MLP model with two hidden layers using the dropout technique. In this study, we were able to get some findings as the experiment had proceeded. First, models using dropout techniques have a slightly more conservative prediction than those without dropout techniques, and it generally shows better performance in classification. Second, CNN models show better classification performance than MLP models. This is interesting because it has shown good performance in binary classification problems which it rarely have been applied to, as well as in the fields where it's effectiveness has been proven. Third, the LSTM algorithm seems to be unsuitable for binary classification problems because the training time is too long compared to the performance improvement. From these results, we can confirm that some of the deep learning algorithms can be applied to solve business binary classification problems.

• Horticultural Science & Technology
• /
• v.34 no.5
• /
• pp.665-676
• /
• 2016

As high temperature during citrus growing season has caused a serious problems including inferior coloration in production of mandarins in Korea, we were to investigate the expression pattern of several genes related with coloration during the ripening in high temperature condition of citrus fruits. The expression of genes related with sugar metabolism, cell wall degradation, and flavonoid synthesis in high temperature conditions was investigated in fruits of 'Haryejosaeng' (Citrus unshiu) and 'Shiranuhi' mandarin (C. reticulata). While the expression of beta-amylase (BMY), phenylalanine ammonia-lyase (PAL), chalcone synthase (CHS), and flavanone 3-hydroxylase (F3H) was differently induced, expression of polygalacturonase (PG) decreased dependently on temperature conditions. In 'Haryejosaeng' mandarin, while the expression of genes related to the skin coloration, such as CHS and F3H genes increased at $25^{\circ}C$, the expression of PAL and stilbene synthase (STS) genes were induced at $30-35^{\circ}C$ in all ripening stages. In 'Shiranuhi' mandarin, the expression of the BMY gene decreased at early time point in all temperature condition and then increased at $30-35^{\circ}C$ than at $25^{\circ}C$ in the ripening stage 2 to 3 of fruits. F3H and STS genes also showed the tendency to decrease at $30-35^{\circ}C$. Although the expression levels of genes in ripening stage 1 and stage 2 of fruits showed similar patterns in both 'Haryejosaeng' and 'Shiranuhi', the expression levels of genes were down-regulated in late ripening stage of 'Shiranuhi' fruits compared to 'Haryejosaeng'. In general, the mRNA levels of seven tested genes were higher in 'Haryejosaeng' than in 'Shiranuhi' mandarin, and expression of genes by high temperature was regulated sensitively in 'Haryejosaeng' compared to 'Shiranuhi' mandarin. Further investigations of expression of various genes based on transcriptome analysis in early ripening stage can provide valuable information about the responses to climatic changes in ripening citrus fruits.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.9B
• /
• pp.610-618
• /
• 2005

As developed Internet and Computer Capability, Many Users take the many information through the network. So requirement of User that use to network was rapidly increased and become various. But it spend much time to accept user requirement on current network, so studied such as Active network for solved it. This Active node on Active network have the capability that stored and processed execution code aside from capability of forwarding packet on current network. So required execution code for executed packet arrived in active node, if execution code should not be in active node, have to take by request previous Action node and Code Server to it. But if this execution code take from previous active node and Code Server, bring to time delay by transport execution code and increased traffic of network and execution time. So, As used execution code stored in cache on active node, it need to increase execution time and decreased number of request. So, our paper suggest ANC caching technique that able to decrease number of execution code request and time of execution code by efficiently store execution code to active node. ANC caching technique may decrease the network traffic and execution time of code, to decrease request of execution code from previous active node.

• Korean Journal of Remote Sensing
• /
• v.36 no.2_2
• /
• pp.351-363
• /
• 2020

It is well known that alluvial sediment located in coastal region has been easily affected by geohazard like ground subsidence, marine or meteorological disasters which threaten invaluable lives and properties. The subsidence is a sinking of the ground due to underground material movement that mostly related to soil compaction by water extraction. Thus, continuous monitoring is essential to protect possible damage from the ground subsidence in the coastal region. Radar interferometric application has been widely used to estimate surface displacement from phase information of synthetic aperture radar (SAR). Thanks to advanced SAR technique like the Small BAseline Subset (SBAS), a time-series of surface displacement could be successfully calculated with a large amount of SAR observations (>20). Because the ALOS-2 PALSAR-2 L-band observations maintain higher coherence compared with other shorter wavelength like X- or C-band, it has been regarded as one of the best resources for Earth science. However, the number of ALOS-2 PALSAR-2 observations might be not enough for the SBAS application due to its global monitoring observation scenario. Unfortunately, the number of the ALOS-2 PALSAR-2 Stripmap images in area of our interest, Busan which located in the Southeastern Korea, is only 11 which is insufficient to apply the SBAS time-series analysis. Although it is common that the radar interferometry utilizes multiple SAR images collected from same acquisition mode, it has been reported that the ALOS-2 PALSAR-2 Stripmap-ScanSAR interferometric application could be possible under specific acquisition mode. In case that we can apply the Stripmap-ScanSAR interferometry with the other 18 ScanSAR observations over Busan, an enhanced time-series surface displacement with better temporal resolution could be estimated. In this study, we evaluated feasibility of the ALOS-2 PALSAR-2 Stripmap-ScanSAR interferometric application using Gamma software considering differences of chirp bandwidth and pulse repetition frequency (PRF) between two acquisition modes. In addition, we analyzed the interferograms with respect to spectral shift of radar carrier frequency and common band filtering. Even though it shows similar level of coherence regardless of spectral shift in the radar carrier frequency, we found periodic spectral noises in azimuth direction and significant degradation of coherence in azimuth direction after common band filtering. Therefore, the characteristics of spectral bandwidth in the range and azimuth direction should be considered cautiously for the ALOS-2 PALSAR-2 Stripmap-ScanSAR interferometry.

• Journal of Navigation and Port Research
• /
• v.34 no.10
• /
• pp.781-786
• /
• 2010

In this study, a new concept of ocean transport system, called the mobile harbor serving for a short distance transport of containers with cargo handling cranes between mother containerships and coastal ports, is introduced. Instead of direct berthing a very large containership at the coastal port, Mobile Harbor is moving to the offshore mooring basin with enough water depth condition. Therefore, investigation of the coastal environment, technical condition and limitation of the domestic trade ports for the application of Mobile Harbor, is essential process. To figure out the accessibility of mobile harbor, the environmental conditions, the cargo handling capacity and marine traffic volume and flow pattern has been analyzed with the tools for marine traffic simulation and virtual navigation aids system. The most proper Mobile Harbor mooring areas among trade ports of the south and east coast are selected by analyzing the obtained information and evaluating its application: (1) Under natural environmental conditions such as air and sea weather, three candidate areas are selected such as Masan port, Ulsan port, and Busan(New port) port. (2) Under marine traffic and appropriateness of water facilities, three candidate areas are selected as Mokpo port, Busan(New port) port, and Donghae & Mookho port (3) For a region-based analysis considering handling capacity and the local managed trade ports in vicinity, three candidate areas are selected as Busan region, Yosu & KwangYang region, and Mokpo region. Through this study, the basic guideline for selection of optimum trade port and offshore mooring basin for mothership and Mobile Harbor is recommended. In order to apply the Mobile Harbor to the real water, navigaton aids as the virtual route identification with AIS must be introduced for maritime safety in the vicinity of Mobile Harbor area which berthing and cargo handling is being conducted.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.31 no.3
• /
• pp.261-268
• /
• 2011

This study was carried out to determine the forage production and quality of Italian ryegrass (IRG) and forage barley developed by Korea in Suwon, 2009~'10. The nine treatments were two IRG varieties (Kowinearly with early maturity and Kowinmaster with medium maturity), five barley varieties (Youngyang, Wooho, Yuyeon, Dami and Youho), and two mixtures (Kowinearly + Yuyeon and Kowinmaster + Yuyeon). The heading dates of Kowinearly and Kowinmaster were 14 May and 18 May, respectively. The growth stage of barley investigated at 22 May were late milk in Youngyang and Wooho, early dough in Dami and early to medium dough in Yuyeon and Youho. Plant length of IRG in IRG + barley mixtures was 117~118 cm, which was longer than those of IRG monoculture of 98~101 cm, and no lodging was found in mixtures. The dry matter (DM) percentage at harvest was 20.7~25.4% in all treatments. The botanical composition of IRG in mixtures was 43.1%. The percentage of spike per barley plant was become high according to progressed maturity, as a 35.7%, 44.1%, 54.8% and 57.2% in late milk, dough, yellowish and full ripeness stage, respectively, and the spike percentages of Youngyang and Wooho were tends to high. The crude protein (CP) content of IRG as 9.0~10.0% was higher than that of barley (7.0~8.5%), and the contents of NDF and ADF of barley were lower than those of IRG, and in vitro DM digestibility were 64.4% in Kowinearly, 64.1% in Kowinmaster, 64.5% in mixture, and 60.2% (Youho) to 66.4% (Wooho) in barley. The yields of DM, CP and in vitro digestible DM were high in Kowinmaster+barley mixture as a 11,508 kg, 1,046 kg and 7,422 kg per ha, respectively (p<0.05). However, no significant differences in forage yield were observed among cultivar of IRG, and barley, although Wooho was tends to high in digestibility and forage yield among five barley varieties. In conclusion, the mixture cultivation of IRG Kowinmaster + forage barley was recommended, because of preventing of IRG lodging, higher plant length of IRG, increasing of forage yield, and stable production. Selection of suitable winter forage species and variety for district, climate environment, and utilization type of farm was also important.

• Horticultural Science & Technology
• /
• v.34 no.1
• /
• pp.172-182
• /
• 2016

This study was aimed to determine the changes in vitamin $B_5$ and $B_6$ contents compared to fresh materials after parboiling treatment of the main vegetables consumed in Korea. The specificity of accuracy and precision for vitamin $B_5$ and $B_6$ analysis method were validated using high-performance liquid chromatography (HPLC). The recovery rate of standard reference material (SRM) was excellent, and all analysis was under the control line based on the quality control chart for vitamin $B_5$ and $B_6$. The Z-score for vitamin $B_6$ in food analysis performance assessment scheme (FAPAS) proficiency test was -1.0, confirming reliability of analytical performance. The vitamin $B_5$ and $B_6$ contents in a total of 39 fresh materials and parboiled samples were analyzed. The contents of vitamin $B_5$ and $B_6$ ranged from 0.000 to 2.462 and from 0.000 to $0.127mg{\cdot}100g^{-1}$, respectively. The highest contents of vitamin $B_5$ and $B_6$ were $2.462mg{\cdot}100g^{-1}$ in fresh fatsia shoots (stem vegetables), and $0.127mg{\cdot}100g^{-1}$ in fresh spinach beet (leafy vegetables), respectively. Moreover, the vitamin $B_5$ and $B_6$ contents for parboiling treatment in most vegetables were reduced or not detected. In particular, the contents of vitamin $B_5$ in parboiled fatsia shoots and vitamin $B_6$ in parboiled yellow potato and spinach beet were decreased 20- and 4-fold compared with fresh material, respectively. These results can be used as important basic data for utilization and processing of various vegetable crops, information for dietary life, management of school meals, and national health for Koreans.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.107-118
• /
• 2016

The advent of 5G mobile communications, which is expected in 2020, will provide many services such as Internet of Things (IoT) and vehicle-to-infra/vehicle/nomadic (V2X) communication. There are many requirements to realizing these services: reduced latency, high data rate and reliability, and real-time service. In particular, a high level of reliability and delay sensitivity with an increased data rate are very important for M2M, IoT, and Factory 4.0. Around the world, 5G standardization organizations have considered these services and grouped them to finally derive the technical requirements and service scenarios. The first scenario is broadcast services that use a high data rate for multiple cases of sporting events or emergencies. The second scenario is as support for e-Health, car reliability, etc.; the third scenario is related to VR games with delay sensitivity and real-time techniques. Recently, these groups have been forming agreements on the requirements for such scenarios and the target level. Various techniques are being studied to satisfy such requirements and are being discussed in the context of software-defined networking (SDN) as the next-generation network architecture. SDN is being used to standardize ONF and basically refers to a structure that separates signals for the control plane from the packets for the data plane. One of the best examples for low latency and high reliability is an intelligent traffic system (ITS) using V2X. Because a car passes a small cell of the 5G network very rapidly, the messages to be delivered in the event of an emergency have to be transported in a very short time. This is a typical example requiring high delay sensitivity. 5G has to support a high reliability and delay sensitivity requirements for V2X in the field of traffic control. For these reasons, V2X is a major application of critical delay. V2X (vehicle-to-infra/vehicle/nomadic) represents all types of communication methods applicable to road and vehicles. It refers to a connected or networked vehicle. V2X can be divided into three kinds of communications. First is the communication between a vehicle and infrastructure (vehicle-to-infrastructure; V2I). Second is the communication between a vehicle and another vehicle (vehicle-to-vehicle; V2V). Third is the communication between a vehicle and mobile equipment (vehicle-to-nomadic devices; V2N). This will be added in the future in various fields. Because the SDN structure is under consideration as the next-generation network architecture, the SDN architecture is significant. However, the centralized architecture of SDN can be considered as an unfavorable structure for delay-sensitive services because a centralized architecture is needed to communicate with many nodes and provide processing power. Therefore, in the case of emergency V2X communications, delay-related control functions require a tree supporting structure. For such a scenario, the architecture of the network processing the vehicle information is a major variable affecting delay. Because it is difficult to meet the desired level of delay sensitivity with a typical fully centralized SDN structure, research on the optimal size of an SDN for processing information is needed. This study examined the SDN architecture considering the V2X emergency delay requirements of a 5G network in the worst-case scenario and performed a system-level simulation on the speed of the car, radius, and cell tier to derive a range of cells for information transfer in SDN network. In the simulation, because 5G provides a sufficiently high data rate, the information for neighboring vehicle support to the car was assumed to be without errors. Furthermore, the 5G small cell was assumed to have a cell radius of 50-100 m, and the maximum speed of the vehicle was considered to be 30-200 km/h in order to examine the network architecture to minimize the delay.