• Journal of Wind Energy
• /
• v.13 no.3
• /
• pp.33-42
• /
• 2022

In this paper, a semi-supervised machine learning technique applied to actual field vibration data acquired from Jeju-do wind turbines for predictive diagnosis of abnormal conditions of offshore wind turbines is introduced. Semi-supervised machine learning, which combines un-supervised learning with supervised learning, can be used to perform anomaly detection in situations where sufficient fault data cannot be obtained. The signal processing results using the spectrogram of the original signal were shown, and external data were used to overcome the problem that disturbance reactions easily occurred due to the imbalance between the number of normal and abnormal data. Out of distribution (OOD), which uses external data, is a technology that is regarded as abnormal data that is unlikely to occur in reality, but we were able to use it by expanding it. By rearranging the distribution of data in this way, classification can be performed more robustly. Specifically, by observing the trends of the abnormal score and the change in the feature of the representation layer, continuous learning was performed through a mixture of existing and new data.

• 기호학연구
• /
• no.56
• /
• pp.45-70
• /
• 2018

Criticism of the postmodern philosophy of the system of representation, which has continued since the Renaissance, is based on a critique of the dichotomy that separates the subjects and objects and the environment from the human being. Interactivity, highlighted in a series of works emerging as postmodern trends in the 1960s, was transmitted to an interactive aspect of digital art in the late 1990s. The key feature of digital art is the possibility of infinite variations reflecting unpredictable changes based on public participation on the spot. In this process, the importance of computer programs is highlighted. Instead of using the existing program as it is, more and more artists are creating and programming their own algorithms or creating unique algorithms through collaborations with programmers. We live in an era of paradigm shift in which programming itself must be considered as a creative act. Simulation technology and VR technology draw attention as a technique to represent the meaning of reality. Simulation technology helps artists create experimental works. In fact, Baudrillard's concept of Simulation defines the other reality that has nothing to do with our reality, rather than a reality that is extremely representative of our reality. His book Simulacra and Simulation refers to the existence of a reality entirely different from the traditional concept of reality. His argument does not concern the problems of right and wrong. There is no metaphysical meaning. Applying the concept of simulation to algorithmic art, the artist models the complex attributes of reality in the digital system. And it aims to build and integrate internal laws that structure and activate the world (specific or individual), that is to say, simulate the world. If the images of the traditional order correspond to the reproduction of the real world, the synthesized images of algorithmic art and simulated space-time are the forms of art that facilitate the experience. The moment of seeing and listening to the work of Ian Cheng presented in this article is a moment of personal experience and the perception is made at that time. It is not a complete and closed process, but a continuous and changing process. It is this active and situational awareness that is required to the audience for the comprehension of post-representation's forms.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.3
• /
• pp.65-74
• /
• 2021

3D reconstruction means that reconstructing the 3D shape of the object in an image and a video. We proposed a progressive occupancy network architecture that can recover not only the overall shape of the object but also the local details. Unlike the original occupancy network, which uses a feature vector embedding information of the whole image, we extract and utilize the different levels of image features depending on the receptive field size. We also propose a novel network architecture that applies the image features sequentially to the decoder blocks in the decoder and improves the quality of the reconstructed 3D shape progressively. In addition, we design a novel decoder block structure that combines the different levels of image features properly and uses them for updating the input point feature. We trained our progressive occupancy network with ShapeNet. We compare its representation power with two prior methods, including prior occupancy network(ONet) and the recent work(DISN) that used different levels of image features like ours. From the perspective of evaluation metrics, our network shows better performance than ONet for all the metrics, and it achieved a little better or a compatible score with DISN. For visualization results, we found that our method successfully reconstructs the local details that ONet misses. Also, compare with DISN that fails to reconstruct the thin parts or occluded parts of the object, our progressive occupancy network successfully catches the parts. These results validate the usefulness of the proposed network architecture.

• The Journal of Society for e-Business Studies
• /
• v.27 no.1
• /
• pp.15-28
• /
• 2022

Due to recent expansion of online market such as clothing, utilizing customer review has become a major marketing measure. User review has been used as a tool of analyzing sentiment of customers. Sentiment analysis can be largely classified with machine learning-based and lexicon-based method. Machine learning-based method is a learning classification model referring review and labels. As research of sentiment analysis has been developed, multi-modal models learned by images and video data in reviews has been studied. Characteristics of words in reviews are differentiated depending on products' and customers' categories. In this paper, sentiment is analyzed via considering review data and metadata of products and users. Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), Self Attention-based Multi-head Attention models and Bidirectional Encoder Representation from Transformer (BERT) are used in this study. Same Multi-Layer Perceptron (MLP) model is used upon every products information. This paper suggests a multi-modal sentiment analysis model that simultaneously considers user reviews and product meta-information.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.519-524
• /
• 2023

In hashing-based image retrieval, the hash code of a manipulated image is different from the original image, making it difficult to search for the same image. This paper proposes and evaluates a self-supervised deephashing model that generates perceptual hash codes from feature information such as texture, shape, and color of images. The comparison models are autoencoder-based variational inference models, but the encoder is designed with a fully connected layer, convolutional neural network, and transformer modules. The proposed model is a variational inference model that includes a SimAM module of extracting geometric patterns and positional relationships within images. The SimAM module can learn latent vectors highlighting objects or local regions through an energy function using the activation values of neurons and surrounding neurons. The proposed method is a representation learning model that can generate low-dimensional latent vectors from high-dimensional input images, and the latent vectors are binarized into distinguishable hash code. From the experimental results on public datasets such as CIFAR-10, ImageNet, and NUS-WIDE, the proposed model is superior to the comparative model and analyzed to have equivalent performance to the supervised learning-based deephashing model. The proposed model can be used in application systems that require low-dimensional representation of images, such as image search or copyright image determination.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.6
• /
• pp.277-281
• /
• 2006

This paper presents a method for the automatic recognition of pitch accents over syllables. The method that we propose is based on the time-delay recursive neural network (TDRNN). which is a neural network classifier with two different representation of dynamic context: the delayed input nodes allow the representation of an explicit trajectory F0(t) along time. while the recursive nodes provide long-term context information that reflects the characteristics of pitch accentuation in spoken English. We apply the TDRNN to pitch accent recognition in two forms: in the normal TDRNN. all of the prosodic features (pitch. energy, duration) are used as an entire set in a single TDRNN. while in the distributed TDRNN. the network consists of several TDRNNs each taking a single prosodic feature as the input. The final output of the distributed TDRNN is weighted sum of the output of individual TDRNN. We used the Boston Radio News Corpus (BRNC) for the experiments on the speaker-independent pitch accent recognition. π 1e experimental results show that the distributed TDRNN exhibits an average recognition accuracy of 83.64% over both pitch events and non-events.

• Journal of Korea Spatial Information System Society
• /
• v.12 no.1
• /
• pp.66-75
• /
• 2010

The authors argues that the current road updating system based on traditional aerial photograph or multi-spectral satellite image appears to be non-user friendly due to lack of the frequent cartographic representation for the new construction sites. Google Earth are currently being emerged as one of important places to extract road features since the RGB satellite image with high multi-temporal resolution can be accessed freely over large areas. This paper is primarily intended to evaluate optimal threshold of RGB pixel values to extract road features from Google Earth. An empirical study for five experimental sites was conducted to confirm how a RGB picture provided Google Earth can be used to extact the road feature. The results indicate that optimal threshold of RGB pixel values to extract road features was identified as 126, 125, 127 for manual operation which corresponds to 25%, 30%, 19%. Also, it was found that display scale difference of Google Earth was not very influential in tracking required RGB pixel value. As a result the 61cm resolution of Quickbird RGB data has shown the potential to realistically identified the major type of road feature by large scale spatial precision while the typical algorithm revealed successfully the area-wide optimal threshold of RGB pixel for road appeared in the study area.

• The KIPS Transactions:PartB
• /
• v.9B no.5
• /
• pp.625-634
• /
• 2002

In this paper, we conducted various experiments with Bayesian networks in order to analyze clinical data of infertility patients. With these experiments, we tried to find out inter-dependencies among important factors playing the key role in clinical pregnancy, and to compare 3 different kinds of Bayesian network classifiers (including NBN, BAN, GBN) in terms of classification performance. As a result of experiments, we found the fact that the most important features playing the key role in clinical pregnancy (Clin) are indication (IND), stimulation, age of female partner (FA), number of ova (ICT), and use of Wallace (ETM), and then discovered inter-dependencies among these features. And we made sure that BAN and GBN, which are more general Bayesian network classifiers permitting inter-dependencies among features, show higher performance than NBN. By comparing Bayesian classifiers based on probabilistic representation and reasoning with other classifiers such as decision trees and k-nearest neighbor methods, we found that the former show higher performance than the latter due to inherent characteristics of clinical domain. finally, we suggested a feature reduction method in which all features except only some ones within Markov blanket of the class node are removed, and investigated by experiments whether such feature reduction can increase the performance of Bayesian classifiers.

• Journal of IKEEE
• /
• v.13 no.2
• /
• pp.159-166
• /
• 2009

So far, many researches have been done to retrieve music information using static classification descriptors such as genre and mood. Since static classification descriptors are based on diverse content-based musical features, they are effective in retrieving similar music in terms of such features. However, human emotion or mood transition triggered by music enables more effective and sophisticated query in music retrieval. So far, few works have been done to evaluate the effect of human mood transition by music. Using formal representation of such mood transitions, we can provide personalized service more effectively in the new applications such as music recommendation. In this paper, we first propose our Emotion State Transition Model (ESTM) for describing human mood transition by music and then describe a music classification and recommendation scheme based on the ESTM. In the experiment, diverse content-based features were extracted from music clips, dimensionally reduced by NMF (Non-negative Matrix Factorization, and classified by SVM (Support Vector Machine). In the performance analysis, we achieved average accuracy 67.54% and maximum accuracy 87.78%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.4
• /
• pp.1730-1747
• /
• 2018

Multi-view video plus depth (MVD) is a mainstream format of 3D scene representation in free viewpoint video systems. The advanced 3D extension of the high efficiency video coding (3D-HEVC) standard introduces new prediction tools to improve the coding performance of depth video. However, the depth video in 3D-HEVC is time consuming. To reduce the complexity of the depth video inter coding, we propose a fast coding unit (CU) size and mode decision algorithm. First, an off-line trained Bayesian model is built which the feature vector contains the depth levels of the corresponding spatial, temporal, and inter-component (texture-depth) neighboring largest CUs (LCUs). Then, the model is used to predict the depth level of the current LCU, and terminate the CU recursive splitting process. Finally, the CU mode search process is early terminated by making use of the mode correlation of spatial, inter-component (texture-depth), and inter-view neighboring CUs. Compared to the 3D-HEVC reference software HTM-10.0, the proposed algorithm reduces the encoding time of depth video and the total encoding time by 65.03% and 41.04% on average, respectively, with negligible quality degradation of the synthesized virtual view.