• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.1
• /
• pp.163-170
• /
• 2022

This paper introduces a method of extracting facial features due to unrefined inputs in real life and improving the problem of not guaranteeing the ideal performance and speed of the object recognition model through a storage algorithm through weight summation. Many facial recognition processes ensure accuracy in ideal situations, but the problem of not being able to cope with numerous biases that can occur in real life is drawing attention, which may soon lead to serious problems in the face recognition process closely related to security. This paper presents a method of quickly and accurately recognizing faces in real time by comparing feature points extracted as input with a small number of feature points that are not overfit to multiple biases, using that various variables such as picture composition eventually take an average form.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.161-161
• /
• 2023

Bias correction of values is a necessary step in downscaling coarse and systematically biased global climate models for use in local climate change impact studies. In addition to univariate bias correction methods, many multivariate methods which correct multiple variables jointly - each with their own mathematical designs - have been developed recently. While some literature have focused on the inter-comparison of these multivariate bias correction methods, none have focused extensively on the effect of diverse configurations (i.e., different combinations of input variables to be corrected) of climate variables, particularly high-dimensional ones, on the ability of the different methods to remove biases in uni- and multivariate statistics. This study evaluates the impact of three configurations (inter-variable, inter-spatial, and full dimensional dependence configurations) on four state-of-the-art multivariate bias correction methods in a national-scale domain over South Korea using a gridded approach. An inter-comparison framework evaluating the performance of the different combinations of configurations and bias correction methods in adjusting various climate variable statistics was created. Precipitation, maximum, and minimum temperatures were corrected across 306 high-resolution (0.2°) grid cells and were evaluated. Results show improvements in most methods in correcting various statistics when implementing high-dimensional configurations. However, some instabilities were observed, likely tied to the mathematical designs of the methods, informing that some multivariate bias correction methods are incompatible with high-dimensional configurations highlighting the potential for further improvements in the field, as well as the importance of proper selection of the correction method specific to the needs of the user.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.276-278
• /
• 2022

All drugs have a specific action in the body, and in many cases, drugs are combinated due to complications or new symptoms during existing drug treatment. In this case, unexpected interactions may occur within the body. Therefore, predicting drug-drug interactions is a very important task for safe drug use. In this study, we propose a deep learning-based predictive model that learns using drug information documents to predict drug interactions that may occur when using multiple drugs. The drug information document was created by combining several properties such as the drug's mechanism of action, toxicity, and target using DrugBank data. And drug information document is pair with another drug documents and used as an input to a deep learning-based predictive model, and the model outputs the interaction between the two drugs. This study can be used to predict future interactions between new drug pairs by analyzing the differences in experimental results according to changes in various conditions.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1759-1768
• /
• 2022

This paper proposes a model to detect depression-related emotions in a user's speech using wellness dialogue scripts provided by AI Hub, topic-specific daily conversation datasets, and chatbot datasets published on Github. There are 18 emotions, including depression and lethargy, in depression-related emotions, and emotion classification tasks are performed using KoBERT and KOELECTRA models that show high performance in language models. For model-specific performance comparisons, we build diverse datasets and compare classification results while adjusting batch sizes and learning rates for models that perform well. Furthermore, a person performs a multi-classification task by selecting all labels whose output values are higher than a specific threshold as the correct answer, in order to reflect feeling multiple emotions at the same time. The model with the best performance derived through this process is called the Depression model, and the model is then used to classify depression-related emotions for user utterances.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.3
• /
• pp.339-349
• /
• 2017

This paper proposes a four-stage algorithm for detecting lanes on a driving car. In the first stage, it extracts region of interests in an image. In the second stage, it employs a median filter to remove noise. In the third stage, a binary algorithm is used to classify two classes of backgrond and foreground of an input image. Finally, an image erosion algorithm is utilized to obtain clear lanes by removing noises and edges remained after the binary process. However, the proposed lane detection algorithm requires high computational time. To address this issue, this paper presents a parallel implementation of a real-time line detection algorithm on a multi-core architecture. In addition, we implement and simulate 8 different processing element (PE) architectures to select an optimal PE architecture for the target application. Experimental results indicate that 40×40 PE architecture show the best performance, energy efficiency and area efficiency.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.149-154
• /
• 2022

Due to high power capabilities and high linearity of GaN devices, GaN Low-Noise Amplifiers (LNAs) without a limiter can be implemented in order to improve noise figure and reduce chip area in radar receivers. In this paper, a GaN LNA is presented for Ka-band radar receivers. The designed LNA was realized in a 150-nm GaN HEMT process and measurement results show that the voltage gain of >23 dB and the noise figure of <6.5 dB including packaging loss in the target frequency range. Under the high-power stress test, measured gain and noise figure of the GaN LNA is degraded after the first stress test, but no more degradation is observed under multiple stress tests. Through post-stress noise and s-parameter measurements, we verified that the GaN LNA is resilient to pulsed input power of ~40 dBm.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.647-654
• /
• 2023

Deep Learning is a useful method for classifying and recognizing complex data such as images and text, and the accuracy of the deep learning method is the basis for making artificial intelligence-based services on the Internet useful. However, the vast amount of user da vita used for training in deep learning has led to privacy violation problems, and it is worried that companies that have collected personal and sensitive data of users, such as photographs and voices, own the data indefinitely. Users cannot delete their data and cannot limit the purpose of use. For example, data owners such as medical institutions that want to apply deep learning technology to patients' medical records cannot share patient data because of privacy and confidentiality issues, making it difficult to benefit from deep learning technology. In this paper, we have designed a privacy preservation technique-applied deep learning technique that allows multiple workers to use a neural network model jointly, without sharing input datasets, in multi-party system. We proposed a method that can selectively share small subsets using an optimization algorithm based on modified stochastic gradient descent, confirming that it could facilitate training with increased learning accuracy while protecting private information.

• International Journal of Computer Science & Network Security
• /
• v.22 no.1
• /
• pp.241-247
• /
• 2022

This study examines teachers' implementation of strategies to modify the aggressive behavior of students with intellectual disabilities in Saudi Arabia, to determine the obstacles to their real-world execution. In addition, it presents potential approaches to overcome the obstacles to implementing strategies with this group of students. The research employed a qualitative design using semi-structured interviews as a data collection tool and applied a thematic analysis. The study population comprised 35 teachers of students with intellectual disabilities and the study sample numbered six teachers. The interviews were conducted via different methods: three by phone, two face-to face, and one using the Zoom platform. The results revealed inadequate understanding among teachers of intellectual disability and behaviour modification strategies, and this affected their capacity to develop plans that were compatible with the needs of students with intellectual disability. The findings also identified multiple obstacles that impede teachers' implementation of strategies to modify aggressive behaviour among students with intellectual disabilities; the most important being the lack of input from a psychological specialist when developing programs to modify aggressive behaviour. In general, it is apparent that programs for modifying aggressive behaviour are neither structured nor complementary, due to the scarcity of administrators with sufficient knowledge and familiarity with the characteristics and personalities of students with intellectual disabilities. This study presents several recommendations, the most important of which is that teachers of students with intellectual disability should develop themselves through training courses to enable them to deal with these students and create treatment plans that include strategies and clear steps to modify the aggressive behaviour of students with intellectual disabilities. To support teachers, it is also necessary to remove the obstacles facing education centres by providing financial support to create an environment in which they can access the required devices and equipment in their classes.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.416-418
• /
• 2022

Glioma is a type of brain tumor that occurs in glial cells and is classified into two types: high hrade hlioma with a poor prognosis and low grade glioma. Magnetic resonance imaging (MRI) as a non-invasive method is widely used in glioma diagnosis research. Studies to obtain complementary information by combining multiple modalities to overcome the incomplete information limitation of single modality are being conducted. In this study, we developed a 3D CNN-based model that applied input-level fusion to MRI of four modalities (T1, T1Gd, T2, T2-FLAIR). The trained model showed classification performance of 0.8926 accuracy, 0.9688 sensitivity, 0.6400 specificity, and 0.9467 AUC on the validation data. Through this, it was confirmed that the grade of glioma was effectively classified by learning the internal relationship between various modalities.

• The Journal of the Korea Contents Association
• /
• v.22 no.8
• /
• pp.274-287
• /
• 2022

Competition in the broadcasting market is intensifying as OTT services are spreading. And Korea is positioned as a competent international contents supply base. This can be helpful for the domestic contents production industry. However, it can result in being incorporated as a subcontractor in the global video industry. Therefore, it is necessary for Korean OTT operators to expand their market upto overseas and maintain competitiveness by linking content competitiveness to the sales expansion. This study was conducted to reduce the risk and encourage implementation through feasibility analysis of overseas business of domestic OTT operators. The North American market was selected as a region with high potential through in-depth interviews with experts and literatures review. And it was confirmed that the partnership with local platform is effective. Then, the sales and input costs were estimated, and business was evaluated using the net present value method. Totally 18 scenarios were created using multiple estimates for copyright cost, subscribers, and rate, which are highly uncertain. From the analyses, 8 scenarios were found to be acceptable. And copyright cost has the greatest impact on business success, followed by rates and subscribers.