• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.10
• /
• pp.1010-1014
• /
• 2010

Article-related product recommender system is an emerging e-commerce service which recommends items based on association in contexts between items and articles. Current services recommend based on the similarity between tags of articles and items, which is deficient not only due to the high cost in manual tagging but also low accuracies in recommendation. As a component of novel article-related item recommender system, we propose a new method for tagging item images based on pre-defined categories. We suggest a hypernetwork-based algorithm for learning association between images, which is represented by visual words, and categories of products. Learned hypernetwork are used to assign multiple tags to unlabeled item images. We show the ability of our method with a product set of real-world online shopping-mall including 1,251 product images with 10 categories. Experimental results not only show that the proposed method has competitive tagging performance compared with other classifiers but also present that the proposed multi-tagging method based on hypernetworks improves the accuracy of tagging.

• Korean Journal of Remote Sensing
• /
• v.22 no.4
• /
• pp.275-284
• /
• 2006

The noise sources of a Hyperion image are mainly due to the atmospheric effects, the sensor's instrumental errors, and A/D conversion. Though uncalibrated, overlapping, and all deep water absorption bands generally are removed, there still exist noisy bands. The visual inspection for selecting clean and stable processing bands is a simple practice, but is a manual, inefficient, and subjective process. In this paper, we propose that the extrema ratio be used for noise estimation and unsupervised band selection. The extrema ratio was compared with existing SNR and entropy measures. First, Gaussian, salt and pepper, and Speckle noises were added to ALI (Advanced Land Imager) images with relatively low noises, and the relation of noise level and those measures was explored. Second, the unsupervised band selection was performed through the EM (Expectation-Maximization) algorithm of the measures which were extracted from a Hyperion images. The Hyperion data were classified into 5 categories according to the image quality by visual inspection, and used as the reference data. The experimental result showed that the extrema ratio could be used effectively for band selection of Hyperion images.

• Current Optics and Photonics
• /
• v.2 no.5
• /
• pp.474-481
• /
• 2018

We propose a polarization phase-shifting technique to investigate the thickness of $Ta_2O_5$ thin films deposited on BK7 substrates, using a modified Sagnac interferometer. Incident light is split by a polarizing beam splitter into two orthogonal linearly polarized beams traveling in opposite directions, and a quarter-wave plate is inserted into the common path to create an unbalanced phase condition. The linearly polarized light beams are transformed into two circularly polarized beams by transmission through a quarter-wave plate placed at the output of the interferometer. The proposed setup, therefore, yields rotating polarized light that can be used to extract a relative phase via the self-reference system. A thin-film sample inserted into the cyclic path modifies the output signal, in terms of the phase retardation. This technique utilizes three phase-shifted intensities to evaluate the phase retardation via simple signal processing, without manual adjustment of the output polarizer, which subsequently allows the thin film's thickness to be determined. Experimental results show that the thicknesses obtained from the proposed setup are in good agreement with those acquired by a field-emission scanning electron microscope and a spectroscopic ellipsometer. Thus, the proposed interferometric arrangement can be utilized reliably for non-contact thickness measurements of transparent thin films and characterization of optical devices.

• Healthcare Informatics Research
• /
• v.24 no.4
• /
• pp.309-316
• /
• 2018

Objectives: Both the valence and arousal components of affect are important considerations when managing mental healthcare because they are associated with affective and physiological responses. Research on arousal and valence analysis, which uses images, texts, and physiological signals that employ deep learning, is actively underway; research investigating how to improve the recognition rate is needed. The goal of this research was to design a deep learning framework and model to classify arousal and valence, indicating positive and negative degrees of emotion as high or low. Methods: The proposed arousal and valence classification model to analyze the affective state was tested using data from 40 channels provided by a dataset for emotion analysis using electrocardiography (EEG), physiological, and video signals (the DEAP dataset). Experiments were based on 10 selected featured central and peripheral nervous system data points, using long short-term memory (LSTM) as a deep learning method. Results: The arousal and valence were classified and visualized on a two-dimensional coordinate plane. Profiles were designed depending on the number of hidden layers, nodes, and hyperparameters according to the error rate. The experimental results show an arousal and valence classification model accuracy of 74.65 and 78%, respectively. The proposed model performed better than previous other models. Conclusions: The proposed model appears to be effective in analyzing arousal and valence; specifically, it is expected that affective analysis using physiological signals based on LSTM will be possible without manual feature extraction. In a future study, the classification model will be adopted in mental healthcare management systems.

• Clinical and Experimental Pediatrics
• /
• v.62 no.9
• /
• pp.360-366
• /
• 2019

Background: Attention deficit-hyperactivity disorder (ADHD) is one of the most common chronic behavioral disorders in school-aged children. Purpose: This study aimed to evaluate the effect of omega-3 supplementation as an alternative therapy for ADHD, which can be caused by vitamin and mineral deficiencies. Methods: This was a double-blinded clinical trial study. Sixty-six children with ADHD (aged 6-12 years) referred to our child and adolescent psychiatric educational and therapeutic clinic were selected based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision criteria. Instruments including the Parent ADHD Rating Scale were used to assess ADHD at 0, 2, 4, and 8 weeks during the study. Results: The results showed no statistically significant difference between the methylphenidate with omega-3 group and methylphenidate with placebo group based on the Parents ADHD Rating Scale between week 0 ($P{\geq}0.96$) and week 8 ($P{\geq}0.75$). There were no significant intergroup differences between the Inattention ($P{\geq}0.48$) and hyperactivity/impulsivity ($P{\geq}0.80$) subscale scores on the Parents ADHD Rating Scale. The most common drug complications in the methylphenidate with placebo and methylphenidate with omega-3 groups were anorexia (27 [54%] vs. 41 [60.29%], respectively) and diarrhea (10 [20%] vs. 8 [11.76%], respectively), but the differences were not statistically significant (P>0.05). Conclusion: Our results demonstrate that a specific dose of omega-3 for 8 weeks had no effect on ADHD.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.4
• /
• pp.531-540
• /
• 2019

Existing dozer productivity computation models use different input variables, formulas, productivity correction factors, and experimental data source. This paper presents a method that characterizes the productivity outputs obtained by the PLS model and the Caterpillar model that are accepted as industry standards. The method identifies the input variables to be collected from the site, the performance charts to be referenced, and the formulas and implements them in a single computational tool. This study verifies that the PLS model may replace the manual computational process of Caterpillar model by eliminating reliance on graphics manipulation. Replacing the Caterpillar model with the PLS model and implementing the process as a function contributes to assess the productivity of a dozer timely by encouraging to utilize real-time information collected directly from the site. This study allows researchers and practitioners to effectively deal with the values of productivity correction factors collected from the job site and to control the productivity. The practicality and effectiveness of the method have been validated by applying to a project case.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.3
• /
• pp.341-348
• /
• 2020

The ability to extract useful information from an image, such as the human eye, is an interface technology essential for AI computer implementation. The building recognition technology has a lower recognition rate than other image recognition technologies due to the various building shapes, the ambient noise images according to the season, and the distortion by angle and distance. The computer vision based building recognition algorithms presented so far has limitations in discernment and expandability due to manual definition of building characteristics. This paper introduces the deep learning CNN (Convolutional Neural Network) model, and proposes new method to improve the recognition rate even by changes of building images caused by season, illumination, angle and perspective. This paper introduces the partial images that characterize the building, such as windows or wall images, and executes the training with whole building images. Experimental results show that the building recognition rate is improved by about 14% compared to the general CNN model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1761-1777
• /
• 2021

An image with infrared features and visible details is obtained by processing infrared and visible images. In this paper, a fusion method based on Laplacian pyramid and generative adversarial network is proposed to obtain high quality fusion images, termed as Laplacian-GAN. Firstly, the base and detail layers are obtained by decomposing the source images. Secondly, we utilize the Laplacian pyramid-based method to fuse these base layers to obtain more information of the base layer. Thirdly, the detail part is fused by a generative adversarial network. In addition, generative adversarial network avoids the manual design complicated fusion rules. Finally, the fused base layer and fused detail layer are reconstructed to obtain the fused image. Experimental results demonstrate that the proposed method can obtain state-of-the-art fusion performance in both visual quality and objective assessment. In terms of visual observation, the fusion image obtained by Laplacian-GAN algorithm in this paper is clearer in detail. At the same time, in the six metrics of MI, AG, EI, MS_SSIM, Q_abf and SCD, the algorithm presented in this paper has improved by 0.62%, 7.10%, 14.53%, 12.18%, 34.33% and 12.23%, respectively, compared with the best of the other three algorithms.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.26 no.7
• /
• pp.759-766
• /
• 2020

In recent years, the maritime traffic environment has been changing in various ways, and the traffic volume has been increasing constantly. Accordingly, the requirements for maritime traffic analysis have become diversified. To this end, traffic characteristics must first be analyzed using vessel trajectory data. However, as the conventional method is mostly manual, it requires a considerable amount of time and effort, and errors may occur during data processing. In addition, ensuring the reliability of the analysis results is difficult, because this method considers the subjective opinion of analysts. Therefore, in this paper, we propose an automated method of traffic network generation for maritime traffic analysis. In the experiment, spatiotemporal features are analyzed using data collected at Mokpo Harbor over six months. The proposed method can automatically generate a traffic network reflecting the traffic characteristics of the experimental area. In addition, it can be applied to a large amount of trajectory data. Finally, as the spatiotemporal characteristics can be analyzed using the traffic network, the proposed method is expected to be used in various maritime traffic analyses.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.12
• /
• pp.143-150
• /
• 2021

In the recent power industry, a change is taking place from manual meter reading to remote meter reading using AMI(Advanced Metering Infrastructure). If such the power data generated from the AMI is recorded on the blockchain, integrity is guaranteed by preventing forgery and tampering. As data sharing becomes transparent, new business can be created. However, Ethereum blockchain is not suitable for processing large amounts of transactions due to the limitation of processing speed. As a solution to overcome such the limitation, various On/Off-Chain methods are being investigated. In this paper, we propose a interface server using data sharding as a solution for storing large amounts of power data in Etherium blockchain environments. Experimental results show that our power-data processing system with sharding method lessen the data omission rate to 0% that occurs when the transactions are transmitted to Ethereum and enhance the processing speed approximately 9 times.