• Journal of the Korea Convergence Society
• /
• v.12 no.3
• /
• pp.17-25
• /
• 2021

When a person wears clothing, it is important to configure an outfit appropriate to the intended occasion. Therefore, T.P.O(Time, Place, Occasion) of the outfit is considered in various fashion recommendation systems based on artificial intelligence. However, there are few studies that directly infer the T.P.O from outfit images, as the nature of the problem causes multi-label and class imbalance problems, which makes model training challenging. Therefore, in this study, we propose a model that can infer the T.P.O of outfit images by employing a label-distribution-aware margin(LDAM) loss function. Datasets for the model training and evaluation were collected from fashion shopping malls. As a result of measuring performance, it was confirmed that the proposed model showed balanced performance in all T.P.O classes compared to baselines.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.4
• /
• pp.845-854
• /
• 2016

Sentiment analysis or opinion mining is a technique of text mining employed to identify subjective information or opinions of an individual from documents in blogs, reviews, articles, or social networks. In the literature, only a problem of binary classification of ratings based on review texts in an online review. However, because there can be positive or negative reviews as well as neutral reviews, a multi-class classification will be more appropriate than the binary classification. To this end, we consider the multi-class classification of ratings based on review texts. In the preprocessing stage, we extract words related with ratings using chi-square statistic. Then the extracted words are used as input variables to multi-class classifiers such as support vector machines and proportional odds model to compare their predictive performances.

• Journal of Communications and Networks
• /
• v.15 no.1
• /
• pp.15-24
• /
• 2013

Provisioning of quality of service (QoS) is a key issue in any multi-media system. However, in wireless systems, supporting QoS requirements of different traffic types is a more challenging problem due to the need to simultaneously minimize two performance metrics - the probability of dropping a handover call and the probability of blocking a new call. Since QoS requirements are not as stringent for non-real-time traffic, as opposed to real-time traffic, more calls can be accommodated by releasing some bandwidth from the already admitted non-real-time traffic calls. If the released bandwidth that is used to handle handover calls is larger than the released bandwidth that is used for new calls, then the resulting probability of dropping a handover call is smaller than the probability of blocking a new call. In this paper, we propose an efficient call admission control algorithm that relies on adaptive multi-level bandwidth-allocation scheme for non-realtime calls. The scheme allows reduction of the call dropping probability, along with an increase in the bandwidth utilization. The numerical results show that the proposed scheme is capable of attaining negligible handover call dropping probability without sacrificing bandwidth utilization.

• International Journal of Aeronautical and Space Sciences
• /
• v.12 no.2
• /
• pp.179-189
• /
• 2011

In this paper, robust adaptive control design problem is addressed for a class of parametrically uncertain aeroelastic systems. A full-state robust adaptive controller was designed to suppress aeroelastic vibrations of a nonlinear wing section. The design used leading and trailing edge control actuations. The full state feedback (FSFB) control yielded a global uniformly ultimately bounded result for two-axis vibration suppression. The pitching and plunging displacements were measurable; however, the pitching and plunging rates were not measurable. Thus, a high gain observer was used to modify the FSFB control design to become an output feedback (OFB) design while the stability analysis for the OFB control law was presented. Simulation results demonstrate the efficacy of the multi-input multi-output control toward suppressing aeroelastic vibrations and limit cycle oscillations occurring in pre- and post-flutter velocity regimes.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.31-33
• /
• 2009

Face detection using AdaBoost algorithm is capable of processing images rapidly while having high detection rates. It seemed to be the fastest and the most robust and it is still today. Many improvements or extensions of this method have been proposed. However, previous approaches only deal with upright faces. They suffer from limited discriminant capability for rotated faces as these methods apply the same features for both upright and rotated faces. To solve this problem, it is necessary that we rotate input images or make independently trained detectors. However, this can be slow and can require a lot of training data, since each classifier requires the computation of many different image features. This paper proposes a robust algorithm for finding rotated faces within an image. It reduces the computational and sample complexity, by finding common features that can be shared across the classes. And it will be able to apply with multi-class object detection.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.144-150
• /
• 2017

Recently, deep learning is used for intelligent processing and accuracy improvement of data. It is formed calculation model composed of multi data processing layer that train the data representation through an abstraction of the various levels. A category of deep learning, convolution neural network is utilized in various research fields, which are human pose estimation, face recognition, image classification, speech recognition. When using the deep layer and lots of class, CNN that show a good performance on image classification obtain higher classification rate but occur the overfitting problem, when using a few data. So, we design the training network based on convolution neural network and trained our image data set for object classification in few class problem. The experiment show the higher classification rate of 7.06% in average than the previous networks designed to classify the object in 1000 class problem.

• Journal of Korean Institute of Industrial Engineers
• /
• v.24 no.1
• /
• pp.145-156
• /
• 1998

In a classical clustering problem, grouping is done on the basis of similarities or distances (dissimilarities) among the elements. Therefore, the objective is to minimize the variance within each group while maximizing the between-group variance among all groups. In this paper, however, a new class of clustering problem is introduced. We call this a laydown grouping problem (LGP). In LGP, the objective is to minimize both the within-group and between-group variances. Furthermore, the problem is expanded to a multi-dimensional case where the two-way minimization process must be considered for each dimension simultaneously for all measurement characteristics. At first, the problem is assessed by analyzing its variance structures and their complexities by conjecturing that LGP is NP-complete. Then, the simulated annealing (SA) algorithm is applied and the results are compared against that from others.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.4 no.3
• /
• pp.217-224
• /
• 2011

This paper proposes DDPT(Dynamic Data Protection Technique) which solves the problem of private information exposure occurring in a dynamic database environment. The DDPT in this paper generates the MAG(Multi-Attribute Generalization) rules using multi-attributes generalization algorithm, and the EC(equivalence class) satisfying the k-anonymity according to the MAG rules. Whenever data is changed, it reconstructs the EC according to the MAC rules, and protects the identification exposure which is caused by the EC change. Also, it measures the information loss rates of the EC which satisfies the ${\ell}$-diversity. It keeps data accuracy by selecting the EC which is less than critical value and enhances private information protection.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.785-804
• /
• 2017

Due to various factors such as postures, facial expressions and illuminations, face recognition by videos often suffer from poor recognition accuracy and generalization ability, since the within-class scatter might even be higher than the between-class one. Herein we address this problem by proposing a hierarchical cascaded classifier for video face recognition, which is a multi-layer algorithm and accounts for the misclassified samples plus their similar samples. Specifically, it can be decomposed into single classifier construction and multi-layer classifier design stages. In single classifier construction stage, classifier is created by clustering and the number of classes is computed by analyzing distance tree. In multi-layer classifier design stage, the next layer is created for the misclassified samples and similar ones, then cascaded to a hierarchical classifier. The experiments on the database collected by ourselves show that the recognition accuracy of the proposed classifier outperforms the compared recognition algorithms, such as neural network and sparse representation.

• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1083-1094
• /
• 2020

Recognizing plant species based on leaf images is challenging because of the large inter-class variation and inter-class similarities among different plant species. The effective extraction of leaf descriptors constitutes the most important problem in plant leaf recognition. In this paper, a multi-scale angular description method is proposed for fast and accurate leaf recognition and retrieval tasks. The proposed method uses a novel scale-generation rule to develop an angular description of leaf contours. It is parameter-free and can capture leaf features from coarse to fine at multiple scales. A fast Fourier transform is used to make the descriptor compact and is effective in matching samples. Both support vector machine and k-nearest neighbors are used to classify leaves. Leaf recognition and retrieval experiments were conducted on three challenging datasets, namely Swedish leaf, Flavia leaf, and ImageCLEF2012 leaf. The results are evaluated with the widely used standard metrics and compared with several state-of-the-art methods. The results and comparisons show that the proposed method not only requires a low computational time, but also achieves good recognition and retrieval accuracies on challenging datasets.