• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.5
• /
• pp.64-74
• /
• 2019

Although a non-rigid registration has high demands in clinical practice, it has a high computational complexity and it is very difficult for ensuring the accuracy and robustness of registration. This study proposes a method of applying a non-rigid registration to 3D magnetic resonance images of brain in an unsupervised learning environment by using a deep-learning network. A feature vector between two images is produced through the network by receiving both images from two different patients as inputs and it transforms the target image to match the source image by creating a displacement vector field. The network is designed based on a U-Net shape so that feature vectors that consider all global and local differences between two images can be constructed when performing the registration. As a regularization term is added to a loss function, a transformation result similar to that of a real brain movement can be obtained after the application of trilinear interpolation. This method enables a non-rigid registration with a single-pass deformation by only receiving two arbitrary images as inputs through an unsupervised learning. Therefore, it can perform faster than other non-learning-based registration methods that require iterative optimization processes. Our experiment was performed with 3D magnetic resonance images of 50 human brains, and the measurement result of the dice similarity coefficient confirmed an approximately 16% similarity improvement by using our method after the registration. It also showed a similar performance compared with the non-learning-based method, with about 10,000 times speed increase. The proposed method can be used for non-rigid registration of various kinds of medical image data.

• Journal of Convergence for Information Technology
• /
• v.12 no.4
• /
• pp.260-269
• /
• 2022

In this paper, We propose an anomaly detection model using deep neural network to automate the identification of outliers of the national air pollution measurement network data that is previously performed by experts. We generated training data by analyzing missing values and outliers of weather data provided by the Institute of Environmental Research and based on the BeatGAN model of the unsupervised learning method, we propose a new model by changing the kernel structure, adding the convolutional filter layer and the transposed convolutional filter layer to improve anomaly detection performance. In addition, by utilizing the generative features of the proposed model to implement and apply a retraining algorithm that generates new data and uses it for training, it was confirmed that the proposed model had the highest performance compared to the original BeatGAN models and other unsupervised learning model like Iforest and One Class SVM. Through this study, it was possible to suggest a method to improve the anomaly detection performance of proposed model while avoiding overfitting without additional cost in situations where training data are insufficient due to various factors such as sensor abnormalities and inspections in actual industrial sites.

• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.257-274
• /
• 2021

Although image segmentation is a critical part of object-based analysis of high resolution imagery, there has been lack of studies to evaluate the quality of image segmentation. In this study, we aimed to find practical and effective methods to obtain optimal parameters for image segmentation. Evaluations of image segmentation are divided into unsupervised, supervised, and qualitative visual interpretation methods. Using the multispectral UAV images, sampled from urban and forest over the Incheon Metropolitan City Park, three evaluation methods were compared. In overall, three methods showed very similar results regardless of the computational costs and applicability, although the optimal parameters determined by the evaluations were different between the urban and forest images. There is no single measure that outperforms in the unsupervised evaluation. Any combinations of intra-segment measures (V, COV, WV) and inter-segment measures (MI, BSH, DTNP) provided almost the same results. Although supervised method may be biased by subjective selection of reference data, it can be easily applied to detect object of interest. The qualitative visual interpretation on the segmentation results corresponded with the unsupervised and supervised evaluations.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.8
• /
• pp.355-364
• /
• 2023

As advanced cyber threats continue to increase in recent years, it is difficult to detect new types of cyber attacks with existing pattern or signature-based intrusion detection method. Therefore, research on anomaly detection methods using data learning-based artificial intelligence technology is increasing. In addition, supervised learning-based anomaly detection methods are difficult to use in real environments because they require sufficient labeled data for learning. Research on an unsupervised learning-based method that learns from normal data and detects an anomaly by finding a pattern in the data itself has been actively conducted. Therefore, this study aims to extract a latent vector that preserves useful sequence information from sequence log data and develop an anomaly detection learning model using the extracted latent vector. Word2Vec was used to create a dense vector representation corresponding to the characteristics of each sequence, and an unsupervised autoencoder was developed to extract latent vectors from sequence data expressed as dense vectors. The developed autoencoder model is a recurrent neural network GRU (Gated Recurrent Unit) based denoising autoencoder suitable for sequence data, a one-dimensional convolutional neural network-based autoencoder to solve the limited short-term memory problem that GRU can have, and an autoencoder combining GRU and one-dimensional convolution was used. The data used in the experiment is time-series-based NGIDS (Next Generation IDS Dataset) data, and as a result of the experiment, an autoencoder that combines GRU and one-dimensional convolution is better than a model using a GRU-based autoencoder or a one-dimensional convolution-based autoencoder. It was efficient in terms of learning time for extracting useful latent patterns from training data, and showed stable performance with smaller fluctuations in anomaly detection performance.

• 전기의세계
• /
• v.39 no.12
• /
• pp.41-56
• /
• 1990

Fuzzy set의 또 다른 장점은 "학습"을 함으로써 시간에 따라 변하는 환경에 대해서도 적용이 가능하다는데 있다. 학습하는 Fuzzy 인공지능의 필요성은 목표하는 바가 복잡할수록, 그리고 신뢰도가 떨어질수록 그 중요성을 더해간다. 이 Fuzzy System의 학습과정은 특히 "Unsupervised Learning"이라는 목표하에 그 연구열이 점점 더해가고 있다. 수년안에 사람들간에 이러한 대화가 오고갈 것은 명백해진 것 같다.화가 오고갈 것은 명백해진 것 같다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.1017-1021
• /
• 1991

This paper is to explore the potential use of the modeling capacity of neural networks for control applications. The tasks are carried out by two neural networks which act as a plant identifier and a system controller, respectively. Using information stored in the identification network control action has been developed. Without supervising control signals are generated by a gradient type iterative algorithm.

• Proceedings of the CALSEC Conference
• /
• 2004.02a
• /
• pp.294-301
• /
• 2004

·Text-mining approach to e-documents organization based on topic hierarchy - Machine-Learning ＆ information Theory-based ㆍ 'Category(topic) discovery' problem → document bundle-based user-constraint document clustering ㆍ 'Automatic categorization' problem → Accelerated EM with CU-based active learning → 'Hierarchy Construction' problem → Unsupervised learning of category subsumption relation

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2001.08a
• /
• pp.103-127
• /
• 2001

Gene expression studies require statistical experimental designs and validation before laboratory confirmation. Various clustering approaches, such as hierarchical, Kmeans, SOM are commonly used for unsupervised learning in gene expression data. Several classification methods, such as gene voting, SVM, or discriminant analysis are used for supervised lerning, where well-defined response classification is possible. Estimating gene-condition interaction effects require advanced, computationally-intensive statistical approaches.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1056-1059
• /
• 2003

Self-Oranizing Map(SOM) is an unsupervised neural network providing cluster analysis of high dimensional input data. The output from the SOM is represented in map that help us to explore data. The weak point of conventional SOM is when the map is large, it take a long time to train the data. The computing time is known to be O(MN) for trainning to find the winning node (M,N are the number of nodes in width and height of the map). This paper presents a new method to reduce the computing time by creating new map. Each node in a new map is the centroid of nodes' group that are in the original map. After create a new map, we find the winning node of this map, then find the winning node in original map only in nodes that are represented by the winning node from the new map. This new method is called "High Speed Self-Oranizing Map"(HS-SOM). Our experiment use HS-SOM to cluster documents and compare with SOM. The results from the experiment shows that HS-SOM can reduce computing time by 30%-50% over conventional SOM.

• Journal of the Korean Home Economics Association
• /
• v.46 no.2
• /
• pp.85-96
• /
• 2008

Given the evolving nature of the family unit, a large number of children are being left unsupervised after school. The purpose of this study is to understand the adaptation ability and emotional capacity of these children. To achieve this objective, we investigated the different characteristics of 708 middle-school students in Seoul, dividing them into two categories adult-care children, for whom adults provide care after school, and self-care children for whom no adult supervision was present. In particular, we examined children's adaptation to the school environment and possible self-consciousness difference between self-care and adult-care children, in consideration of their family characteristic; divorced, separated, widowed parent, remarried parents, ordinary families. The results showed that self-care children tend to have a higher rate of shame-proneness and guilt-proneness compared to adult-care children. Furthermore, self-care children exhibited lower school adaptation rate than adult-care children. There was no significant difference in schoolmate relationships between the two groups. In relation to specific family structures, children from reorganized families showed no significant differences in school adaptation and self-conscious, while self-care children from ordinary families revealed low school adaptation and high self-conscious characteristics. The results of this study are critical in the effective analysis and understanding of children's adaptive and emotional behaviors arising from changes in their family structure.