• Journal of Digital Convergence
• /
• v.19 no.5
• /
• pp.307-313
• /
• 2021

In this paper, a lightweight network with fewer parameters compared to the existing object detection method is proposed. In the case of the currently used detection model, the network complexity has been greatly increased to improve accuracy. Therefore, the proposed network uses EfficientNet as a feature extraction network, and the subsequent layers are formed in a pyramid structure to utilize low-level detailed features and high-level semantic features. An attention process was applied between pyramid structures to suppress unnecessary noise for prediction. All computational processes of the network are replaced by depth-wise and point-wise convolutions to minimize the amount of computation. The proposed network was trained and evaluated using the PASCAL VOC dataset. The features fused through the experiment showed robust properties for various objects through a refinement process. Compared with the CNN-based detection model, detection accuracy is improved with a small amount of computation. It is considered necessary to adjust the anchor ratio according to the size of the object as a future study.

• Molecules and Cells
• /
• v.44 no.8
• /
• pp.602-612
• /
• 2021

DNA methylation is an important epigenetic mechanism affecting genome structure, gene regulation, and the silencing of transposable elements. Cell- and tissue-specific methylation patterns are critical for differentiation and development in eukaryotes. Dynamic spatiotemporal methylation data in these cells or tissues is, therefore, of great interest. However, the construction of bisulfite sequencing libraries can be challenging if the starting material is limited or the genome size is small, such as in Arabidopsis. Here, we describe detailed methods for the purification of Arabidopsis embryos at all stages, and the construction of comprehensive bisulfite libraries from small quantities of input. We constructed bisulfite libraries by releasing embryos from intact seeds, using a different approach for each developmental stage, and manually picking single-embryo with microcapillaries. From these libraries, reliable Arabidopsis methylome data were collected allowing, on average, 11-fold coverage of the genome using as few as five globular, heart, and torpedo embryos as raw input material without the need for DNA purification step. On the other hand, purified DNA from as few as eight bending torpedo embryos or a single mature embryo is sufficient for library construction when RNase A is treated before DNA extraction. This method can be broadly applied to cells from different tissues or cells from other model organisms. Methylome construction can be achieved using a minimal amount of input material using our method; thereby, it has the potential to increase our understanding of dynamic spatiotemporal methylation patterns in model organisms.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.10
• /
• pp.225-234
• /
• 2019

The purpose of this paper is to analyze the additional construction cost of G-SEED certification for domestic office building reflecting the latest standard(G-SEED 2016-2), and to derive cost impact by category and level. Therefore, it is intended to provide quantitave cost data according to G-SEED certification at the planning phase of the project, estimate the additional construction cost per level according to G-SEED Certification of similar project to be carried out in the future, and encourage G-SEED certification by supporting the decision of the owners. Method: The Process and method of this study are summarized in five steps, 1) Review of previous research, 2) Selection of target project, 3) Scenario setting by level, 4) Additional construction cost for each evaluation category, 5) Extraction of additional construction cost ratio by level. Result: This paper analyzed the cost impact by deriving the additional construction cost of detailed category for level improvement according to the revised G-SEED certification(G-SEED 2016-2). In conclusion, an additional construction cost(ratio) of G-SEED projects to the reference building is drawn as good level; 157,426,241 KWN(+0.43%), very good level; 321,907,802 KWN(+0.88%), excellent level; 999,371,478 KWN(+2.74%), and outstanding level; 1,467,047,718 KWN(+4.02%).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.3
• /
• pp.384-390
• /
• 2020

The reverberation effect on the sound when producing movies or VR contents is a very important factor in the realism and liveliness. The reverberation time depending the space is recommended in a standard called RT60(Reverberation Time 60 dB). In this paper, we propose a scene recognition technique for automatic reverberation editing. To this end, we devised a classification model that independently trains color images and predicted depth images in the same model. Indoor scene classification is limited only by training color information because of the similarity of internal structure. Deep learning based depth information extraction technology is used to use spatial depth information. Based on RT60, 10 scene classes were constructed and model training and evaluation were conducted. Finally, the proposed SCR + DNet (Scene Classification for Reverb + Depth Net) classifier achieves higher performance than conventional CNN classifiers with 92.4% accuracy.

• Food Science of Animal Resources
• /
• v.42 no.1
• /
• pp.1-17
• /
• 2022

Lipids are one of the major macronutrients essential for adequate growth and maintenance of human health. Their structure is not only complex but also diverse, which makes systematic and holistic analyses challenging; consequently, little is known regarding the relationship between phenotype and mechanism of action. In recent years, rapid advancements have been made in the fields of lipidomics and bioinformatics. In comparison with traditional approaches, mass spectrometry-based lipidomics can rapidly identify as well as quantify >1,000 lipid species at the same time, facilitating comprehensive, robust analyses of lipids in tissues, cells, and body fluids. Accordingly, lipidomics is now being widely applied in various fields, particularly food and nutrition science. In this review, we discuss lipid classification, extraction techniques, and detection and analysis using lipidomics. We also cover how lipidomics is being used to assess food obtained from livestock and poultry. The information included herein should serve as a reference to determine how to characterize lipids in animal food samples, enhancing our understanding of the application of lipidomics in the field in animal husbandry.

• Journal of Multimedia Information System
• /
• v.9 no.2
• /
• pp.113-120
• /
• 2022

Mongolian folk songs are inspired by Mongolian labor songs and are classified into long and short songs. Mongolian long songs have ancient origins, are rich in legends, and are a great source of folklore. So it was inscribed by UNESCO in 2008. Mongolian written literature is formed under the direct influence of oral literature. Mongolian long song has 3 classes: ayzam, suman, and besreg by their lyrics and structure. In ayzam long song, the world perfectly embodies the philosophical nature of world phenomena and the nature of human life. Suman long song has a wide range of topics such as the common way of life, respect for ancestors, respect for fathers, respect for mountains and water, livestock and animal husbandry, as well as the history of Mongolia. Besreg long songs are dominated by commanded and trained characters. In this paper, we proposed a method to classify their 3 types of long songs using machine learning, based on their lyrics structures without semantic information. We collected lyrics of over 80 long songs and extracted 11 features from every single song. The features are the name of a song, number of the verse, number of lines, number of words, general value, double value, elapsed time of verse, elapsed time of 5 words, and the longest elapsed time of 1 word, full text, and type label. In experimental results, our proposed features show on average 78% recognition rates in function type machine learning methods, to classify the ayzam, suman, and besreg classes.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.662-667
• /
• 2022

Semantic role labeling task(SRL) is to extract predicate and arguments such as agent, patient, place, time. In the previously SRL task studies, a pipeline method extracting linguistic features of sentence has been proposed, but in this method, errors of each extraction work in the pipeline affect semantic role labeling performance. Therefore, methods using End-to-End neural network model have recently been proposed. In this paper, we propose a neural network model using the Biaffine Average Attention model for SRL task. The proposed model consists of a structure that can focus on the entire sentence information regardless of the distance between the predicate in the sentence and the arguments, instead of LSTM model that uses the surrounding information for prediction of a specific token proposed in the previous studies. For evaluation, we used F1 scores to compare two models based BERT model that proposed in existing studies using F1 scores, and found that 76.21% performance was higher than comparison models.

• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.

• Journal of the Korean Wood Science and Technology
• /
• v.32 no.2
• /
• pp.26-32
• /
• 2004

Chemical compositions and chemical structure of lignin and alkali cooking condition and fiber length of red pepper were investigated and compared to those of woods. The chemical compositions of red pepper were higher component of extraction than that of wood. The contents of carbon and hydrogen of Klason lignin in red pepper were similar to that of pine and birch wood. On the other hand, the contents of oxygen and nitrogen of Klason lignin in the red pepper were higher than that of wood. The result of nitrobenzene oxidation shows that Klason lignin of red pepper was similar to lignin of softwood. The best alkali cooking condition of red pepper was 0.2%-anthraquinone, active alkali of 20% and liquor ratio of 1:7. The fiber length of red pepper was about 0.47 mm. Therefore, the red pepper fiber will be able to use special purpose of short fiber.

• Journal of the Korean Geotechnical Society
• /
• v.39 no.8
• /
• pp.17-28
• /
• 2023

Laboratory-scale hydraulic fracturing experiments were conducted on granite specimens at various viscosities and injection rates of the fracturing fluid. A series of cross-sectional computed tomography (CT) images of fractured specimens was obtained via a three-dimensional X-ray CT imaging method. Pixel-level fracture segmentation of the CT images was conducted using a convolutional neural network (CNN)-based Nested U-Net model structure. Compared with traditional image processing methods, the CNN-based model showed a better performance in the extraction of thin and complex fractures. These extracted fractures extracted were reconstructed in three dimensions and morphologically analyzed based on their fracture volume, aperture, tortuosity, and surface roughness. The fracture volume and aperture increased with the increase in viscosity of the fracturing fluid, while the tortuosity and roughness of the fracture surface decreased. The findings also confirmed the anisotropic tortuosity and roughness of the fracture surface. In this study, a CNN-based model was used to perform accurate fracture segmentation, and quantitative analysis of hydraulic stimulated fractures was conducted successfully.