• The Journal of Bigdata
• /
• v.4 no.2
• /
• pp.93-103
• /
• 2019

In this paper, we introduce a big data platform and a metadata management technique for fishery science R & D information. The big data platform collects and integrates various types of fisheries science R & D information and suggests how to build it in the form of a data lake. In addition to existing data collected and accumulated in the field of fisheries science, we also propose to build a big data platform that supports diverse analysis by collecting unstructured big data such as satellite image data, research reports, and research data. Next, by collecting and managing metadata during data extraction, preprocessing and storage, systematic management of fisheries science big data is possible. By establishing metadata in a standard form along with the construction of a big data platform, it is meaningful to suggest a systematic and continuous big data management method throughout the data lifecycle such as data collection, storage, utilization and distribution.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.1
• /
• pp.13-19
• /
• 2019

The accuracy of machine learning is greatly affected by amount of learning data and quality of data. Collecting existing Web-based learning data has danger that data unrelated to actual learning can be collected, and it is impossible to secure data transparency. In this paper, we propose a method for collecting data directly in parallel by blocks in a block - chain structure, and comparing the data collected by each block with data in other blocks to select only good data. In the proposed system, each block shares data with each other through a chain of blocks, utilizes the All-reduce structure of Parallel-SGD to select only good quality data through comparison with other block data to construct a learning data set. Also, in order to verify the performance of the proposed architecture, we verify that the original image is only good data among the modulated images using the existing benchmark data set.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.65-73
• /
• 2023

In this paper, we propose a data augmentation method based on CNN(Convolutional Neural Network) learning for efficiently obtaining concrete crack image datasets. Real concrete crack images are not only difficult to obtain due to their unstructured shape and complex patterns, but also may be exposed to dangerous situations when acquiring data. In this paper, we solve the problem of collecting datasets exposed to such situations efficiently in terms of cost and time by using vector and thickness-based data augmentation techniques. To demonstrate the effectiveness of the proposed method, experiments were conducted in various scenes using U-Net-based crack detection, and the performance was improved in all scenes when measured by IoU accuracy. When the concrete crack data was not augmented, the percentage of incorrect predictions was about 25%, but when the data was augmented by our method, the percentage of incorrect predictions was reduced to 3%.

• The Korean Journal of Community Living Science
• /
• v.19 no.3
• /
• pp.431-444
• /
• 2008

Today people obtain much information about appearances through movies. The stars' body images in movies have been created through total fashion as they play the roles of an ideal model to create actual body images. This study examined the relations between the body images in the film media in terms of appearances or total fashion and actual body images based on the simulation theory by Jean Baudrillard. It conducted a literature study by collecting and analyzing film media-related picture images from books, papers, periodicals and the Internet home and abroad. The research scope was limited to the stars who made appearances in movies and the ones who starred in movies in addition to their other lines of work. As a result, the following conclusions were drawn; the stars' body images went hyperreal between their actual body images and the ones of their roles in movies and then were simulated in fashion. To be specific, 1. the stars' actual body images went hyperreal and were simulated in fashion in the following cases; 1) the stars in the heyday of film; 2) fashionistas; and 3) pictures taken by paparazzi. 2. The body images of their roles went hyperreal in the following cases; 1) through brand participation; 2) through changed body images; and 3) through cyber body images. Their body images became hyperreal and were simulated in fashion by the input of fashion designers and the techniques of other areas. The body images in film are the object of desire to the audience. They go hyperreal and become the ideal body image to real people. The various kinds of hyperreal images in the film media create new beauty as a reference and object of comparison for people to change their body images in reality.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.40 no.4
• /
• pp.716-732
• /
• 2016

Focusing on the communication effects of fast fashion visual merchandising (VM), this paper examines the effects of a fast fashion store's VM attributes on consumer's perceptions of store image towards newness and prestige that influence the time spent at stores as well as the frequency of visits. This study was conducted by collecting data online using males and females in their twenties to forties; subsequently, a total of 382 samples were analyzed. The VM communication effect model utilized in fast fashion stores was developed and tested on structural equation modeling. The findings of the study were as follows. First, the show window presentation and ancillary facilities of VM elements have a positive effect on the perception of newness, while merchandise display, layout, and signage have a positive influence on the perception of prestige. Therefore, the VM elements in the fast fashion stores that affect the perception of newness and prestige are unique. Second, the perceptions of newness and prestige have a positive impact on time spent in fast fashion stores; however, only the perception of store's newness has a significant effect on the frequency of visits. Third, show window presentation and facilities are VM elements that directly influence the time spent and frequency of visits. Finally, we confirm that store image partially mediate between VM elements and shopping behavior at a fast fashion store.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.3
• /
• pp.615-622
• /
• 2009

Advances in wireless communication and hardware technology have made it possible to manufacture high-performance tiny sensor nodes. More recently, the availability of inexpensive CMOS cameras that are able to capture multimedia data from the environment has fostered the development of Wireless Multimedia Sensor Networks (WMSNs). WMSN with the CMOS imaging sensor which is cheaper and consumes lower power than the CCD will not only enhance existing sensor network but also enable several new application such as multimedia surveillance sensor network, multimedia environment monitoring. This paper presents the design of a multimedia sensor network with the image sensor mote developed by us using the CMOS. Given new multimeida sensor network, the new image collecting protocol was tested and analyzed.

• Journal of the Korean Society of Radiology
• /
• v.16 no.3
• /
• pp.249-255
• /
• 2022

Positron emission tomography for small animals has very high spatial resolution for imaging very small organs. To achieve good spatial resolution, the system must be constructed using very small scintillation pixels. When a detector is constructed using very small scintillation pixels, the size of the applicable array varies depending on the photosensor pixel. In a previous study, a study was conducted to find the maximum scintillation pixel arrangement according to the size of the photosensor. In this study, a detector with a light guide was designed to configure the detector using a more extended array of scintillation pixels, and try to find the maximum arrangement in which all scintillation pixels are imaged. The detector was designed using DETECT2000, which can simulate a detector made of a scintillator. Simulations were performed by configuring the detectors from an 11 × 11 scintillation pixel array to a 16 × 16 array. After obtaining a flood image by collecting the light generated from the scintillation pixel with a photosensor, the largest arrangement without overlap was found through image analysis. As a result, the largest arrangement in which all scintillation pixels could be distinguished without overlapping was a 15 × 15 arrangement.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.7
• /
• pp.981-988
• /
• 2022

The intelligent production plant called smart factories that apply information and communication technology (ICT) are collecting data in real time through various sensors. Recently, researches that effectively applying to these collected data have gained a lot of attention. This paper proposes a method for the tool condition monitoring based on the sound signal generated in machining process. First, it not only detects a fault tool, but also presents various tool states according to idle and active operation. The second, it's to represent the power spectrum of the sounds as images and apply some transformations on them in order to reveal, expose, and emphasize the health patterns that are hidden inside them. Finally, the contrast-enhanced PSD image obtained is diagnosed by using CNN. The results of the experiments demonstrate the high discrimination potential afforded by the proposed sound PSD image + CNN and show high diagnostic results according to the tool status.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.51-59
• /
• 2024

Collecting a dataset with a corresponding labeled gaze vector requires a high cost in the gaze estimation field. In this paper, we suggest a data augmentation of manipulating the gaze of an original image, which improves the accuracy of the gaze estimation model when the number of given gaze labels is restricted. By conducting multi-class gaze bin classification as an auxiliary task and adjusting the latent variable of the diffusion model, the model semantically edits the gaze from the original image. We manipulate a non-binary attribute, pitch and yaw of gaze vector to a desired range and uses the edited image as an augmented train data. The improved gaze accuracy of the gaze estimation network in the semi-supervised learning validates the effectiveness of our data augmentation, especially when the number of gaze labels is 50k or less.

• Korean Journal of Heritage: History & Science
• /
• v.51 no.4
• /
• pp.120-145
• /
• 2018

This study intended to examine the drainage facility of Gyeongbokgung Palace based on the traces of the relic found during the excavation survey. Historical records indicate that various efforts have been made for smooth drainage facility for the palace since the foundation of the Joseon Dynasty. Although there are no drawings showing the image of early appearance of Gyeongbokgung Palace during the foundation, it is possible to estimate it through the drawings prepared after the King Yeongjo's reign. The image of the palace after reconstruction can be seen through the "Bukgwoldohyeong" and other relevant documents. At this present, since the survey intends to determine the image of Gyeongbokgung Palace during the reign of King Gojong based on the standard restoration plan of the Cultural Heritage Administration, this study also focused on the drainage facility of Gyeongbokgung Palace during King Gojong's reign, particularly on the collecting wells and culverts of six areas including "Chimjeon Hall", "Taewonjeon Hall", "Geoncheongung Hall", "Sojubang Hall", "Hamhwadang - Jipgyeongdang - Yeonghundang Hall", and "Heungbokjeon Hall". Gyeongbokgung Palace is divided into various zones composed of the central halls and surrounding corridors, and the drains also primarily start from each hall and later join the central drain of the zone. The central drain then leads to the "Eo-gu(御溝)" and the water led to the "Eo-gu" is finally discharged through the water gate. It appears that this series of processes were basically devised to coordinate artificial drain with the natural drain using the natural geographical features of the palace. Research showed that the collecting well where the draining begin was installed in the area where a large amount of household sewage was generated but mostly in the corners where corridors met or corridors and wall met. This appears to be an arrangement to handle the water falling from the roof and household sewage. Also, "Ju(廚)" was installed mainly at the end of the corridor to handle household sewage. The installation of these drainage facilities shows the possibility that the drainage of Gyeongbokgung was very compact under a series of plans.