• Korean Journal of Remote Sensing
• /
• v.39 no.6_3
• /
• pp.1771-1777
• /
• 2023

With the advancements of deep learning, many semantic segmentation-based methods for land cover classification have been proposed. However, existing deep learning-based models only use image information and cannot guarantee spatiotemporal consistency. In this study, we propose a land cover classification model using geographical coordinates. First, the coordinate features are extracted through the Coordinate Hash Encoder, which is an extension of the Multi-resolution Hash Encoder, an implicit neural representation technique, to the longitude-latitude coordinate system. Next, we propose an architecture that combines the extracted coordinate features with different levels of U-net decoder. Experimental results show that the proposed method improves the mean intersection over union by about 32% and improves the spatiotemporal consistency.

• Proceedings of the IEEK Conference
• /
• 2001.06b
• /
• pp.313-316
• /
• 2001

In this paper we present the design of a JPEG Encoder using SystemC Methodology Our methodology supports the efficient mapping of C/C++ functional descriptions directly into hardware. The use of C/C++ to model al1 parts of the system provides great flexibility and enables faster simulation compared to existing methodologies. The designer can estimate system performance and verify functional correctness of the designs using commonly available software compilers. A design flow in SystemC begins with an untimed description in C++, using a library of new data types useful for modeling hardware. The description can be compiled and simulated for functional correctness. Then, the design may be refined by adding interface specification and timing information, and again the timed description can be compiled with a standard C++ compiler, simulated, and debugged.

• ETRI Journal
• /
• v.43 no.6
• /
• pp.1049-1057
• /
• 2021

Automatic spacing in Korean is used to correct spacing units in a given input sentence. The demand for automatic spacing has been increasing owing to frequent incorrect spacing in recent media, such as the Internet and mobile networks. Therefore, herein, we propose a transformer encoder that reads a sentence bidirectionally and can be pretrained using an out-of-task corpus. Notably, our model exhibited the highest character accuracy (98.42%) among the existing automatic spacing models for Korean. We experimentally validated the effectiveness of bidirectional encoding and pretraining for automatic spacing in Korean. Moreover, we conclude that pretraining is more important than fine-tuning and data size.

• Journal of Information Processing Systems
• /
• v.20 no.1
• /
• pp.13-23
• /
• 2024

Music brings pleasure and relaxation to people. Therefore, it is necessary to classify musical genres based on scenes. Identifying favorite musical genres from massive music data is a time-consuming and laborious task. Recent studies have suggested that machine learning algorithms are effective in distinguishing between various musical genres. However, meeting the actual requirements in terms of accuracy or timeliness is challenging. In this study, a hybrid machine learning model that combines a deep residual auto-encoder (DRAE) and support vector machine (SVM) for musical genre recognition was proposed. Eight manually extracted features from the Mel-frequency cepstral coefficients (MFCC) were employed in the preprocessing stage as the hybrid music data source. During the training stage, DRAE was employed to extract feature maps, which were then used as input for the SVM classifier. The experimental results indicated that this method achieved a 91.54% F1-score and 91.58% top-1 accuracy, outperforming existing approaches. This novel approach leverages deep architecture and conventional machine learning algorithms and provides a new horizon for musical genre classification tasks.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.41 no.4
• /
• pp.75-82
• /
• 2004

The channel transmission ratio is speeded up by the combination of the Huffman algorithm, the model scheme of the lossy transform having minimum average code lengths for the image information and good instantaneous decoding capability, with the Lempel-Ziv algorithm showing the fast processing performance during the compression process. In order to increase the processing speed during the compression process, ICMEP algorithm is proposed and the entropy encoder of HDTV is designed and inspected. The ICMEP entropy encoder have been designed by choosing the top-down method and consisted of the source codes and the test benches by the behavior expression with VHDL. As a simulation results, implemented ICMEP entropy encoder confirmed that whole system efficiency by memory saturation prevention and compressibility increase improves.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.1
• /
• pp.1-14
• /
• 2024

Time-series forecasting is extensively used in the actual world. Recent research has shown that Transformers with a self-attention mechanism at their core exhibit better performance when dealing with such problems. However, most of the existing Transformer models used for time series prediction use the traditional encoder-decoder architecture, which is complex and leads to low model processing efficiency, thus limiting the ability to mine deep time dependencies by increasing model depth. Secondly, the secondary computational complexity of the self-attention mechanism also increases computational overhead and reduces processing efficiency. To address these issues, the paper designs an efficient multi-layer attention-based time-series forecasting model. This model has the following characteristics: (i) It abandons the traditional encoder-decoder based Transformer architecture and constructs a time series prediction model based on multi-layer attention mechanism, improving the model's ability to mine deep time dependencies. (ii) A cross attention module based on cross attention mechanism was designed to enhance information exchange between historical and predictive sequences. (iii) Applying a recently proposed sparse attention mechanism to our model reduces computational overhead and improves processing efficiency. Experiments on multiple datasets have shown that our model can significantly increase the performance of current advanced Transformer methods in time series forecasting, including LogTrans, Reformer, and Informer.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.2
• /
• pp.39-44
• /
• 2020

Malicious codes cause damage to equipments while avoiding detection programs(vaccines). The reason why it is difficult to detect such these new malwares using the existing vaccines is that they use "signature-based" detection techniques. these techniques effectively detect already known malicious codes, however, they have problems about detecting new malicious codes. Therefore, most of vaccines have recognized these drawbacks and additionally make use of "heuristic" techniques. This paper proposes a technology to detecting unknown malicious code using deep learning. In addition, detecting malware skill using Supervisor Learning approach has a clear limitation. This is because, there are countless files that can be run on the devices. Thus, this paper utilizes Stacked Convolution AutoEncoder(SCAE) known as Semi-Supervisor Learning. To be specific, byte information of file was extracted, imaging was carried out, and these images were learned to model. Finally, Accuracy of 98.84% was achieved as a result of inferring unlearned malicious and non-malicious codes to the model.

• Journal of Internet Computing and Services
• /
• v.22 no.1
• /
• pp.13-22
• /
• 2021

Recently network based attack technologies are rapidly advanced and intelligent, the limitations of existing signature-based intrusion detection systems are becoming clear. The reason is that signature-based detection methods lack generalization capabilities for new attacks such as APT attacks. To solve these problems, research on machine learning-based intrusion detection systems is being actively conducted. However, in the actual network environment, attack samples are collected very little compared to normal samples, resulting in class imbalance problems. When a supervised learning-based anomaly detection model is trained with such data, the result is biased to the normal sample. In this paper, we propose to overcome this imbalance problem through One-Class Anomaly Detection using an auto encoder. The experiment was conducted through the NSL-KDD data set and compares the performance with the supervised learning models for the performance evaluation of the proposed method.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.91-98
• /
• 2022

Lifelog data continuously collected through a wearable device may contain many outliers, so in order to improve data quality, it is necessary to find and remove outliers. In general, since the number of outliers is less than the number of normal data, a class imbalance problem occurs. To solve this imbalance problem, we propose a method that applies Variational AutoEncoder to outliers. After preprocessing the outlier data with proposed method, it is verified through a number of machine learning models(classification). As a result of verification using body weight data, it was confirmed that the performance was improved in all classification models. Based on the experimental results, when analyzing lifelog body weight data, we propose to apply the LightGBM model with the best performance after preprocessing the data using the outlier processing method proposed in this study.

• Journal of the Korean Society of Visualization
• /
• v.20 no.3
• /
• pp.94-101
• /
• 2022

The evaluation of the drag and lift as the aerodynamic performance of airfoils is essential. In addition, the analysis of the velocity and pressure fields is needed to support the physical mechanism of the force coefficients of the airfoil. Thus, the present study aims at establishing two different deep learning models to predict force coefficients and flow fields of the airfoil. One is the convolutional neural network (CNN) model to predict drag and lift coefficients of airfoil. Another is the Encoder-Decoder (ED) model to predict pressure distribution and velocity vector field. The images of airfoil section are applied as the input data of both models. Thus, the computational fluid dynamics (CFD) is adopted to form the dataset to training and test of both CNN models. The models are established by the convergence performance for the various hyperparameters. The prediction capability of the established CNN model and ED model is evaluated for the various NACA sections by comparing the true results obtained by the CFD, resulting in the high accurate prediction. It is noted that the predicted results near the leading edge, where the velocity has sharp gradient, reveal relatively lower accuracies. Therefore, the more and high resolved dataset are required to improve the highly nonlinear flow fields.