• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2023.05a
• /
• pp.405-406
• /
• 2023

In the era of AI, intelligent construction safety technologies are being introduced to the construction safety environment, but the application of AI has limitations due to the lack of accident images to learn in complex construction sites. In order to overcome this, we will introduce an intelligent risk detection system that dramatically improves risk detection accuracy by combining AI with digital twin technology, and introduce various cases.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.663-672
• /
• 2023

Most vehicle detection methods have poor vehicle feature extraction performance at night, and their robustness is reduced; hence, this study proposes a night vehicle detection method based on style transfer image enhancement. First, a style transfer model is constructed using cycle generative adversarial networks (cycleGANs). The daytime data in the BDD100K dataset were converted into nighttime data to form a style dataset. The dataset was then divided using its labels. Finally, based on a YOLOv5s network, a nighttime vehicle image is detected for the reliable recognition of vehicle information in a complex environment. The experimental results of the proposed method based on the BDD100K dataset show that the transferred night vehicle images are clear and meet the requirements. The precision, recall, mAP@.5, and mAP@.5:.95 reached 0.696, 0.292, 0.761, and 0.454, respectively.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.77-84
• /
• 2024

Aberration is still a problem for making augmented reality displays. The existing methods to solve this problem are either slow and inefficient, consume too much battery, or are too complex for straightforward implementation. There are still some problems with image quality, and users may suffer from eye strain and headaches because the images provided to each eye lack accuracy, causing the brain to receive mismatched cues between the vergence and accommodation of the eyes. In this paper, we implemented a computer simulation of an optical aberration using Zernike polynomials which are defocus, trefoil, coma, and spherical. The research showed that these optical aberrations impact the Point Spread Function (PSF) and Modulation Transfer Function (MTF). We employed the phase conjugate technique to mitigate aberrations. The findings revealed that the most significant impact on the PSF and MTF comes from the influence of spherical aberration and coma aberration.

• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.415-422
• /
• 2024

Recent advances in construction automation have led to increased use of deep learning-based computer vision technology for construction monitoring. However, monitoring systems based on supervised learning struggle with recognizing complex risk factors in construction environments, highlighting the need for adaptable solutions. Large multimodal models, pretrained on extensive image-text datasets, present a promising solution with their capability to recognize diverse objects and extract semantic information. This paper proposes a methodology that generates training data for multimodal models, including safety-centric descriptions using GPT-4V, and fine-tunes the LLaVA model using the LoRA method. Experimental results from seven construction site hazard scenarios show that the fine-tuned model accurately assesses safety status in images. These findings underscore the proposed approach's effectiveness in enhancing construction site safety monitoring and illustrate the potential of large multimodal models to tackle domain-specific challenges.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.11
• /
• pp.2321-2326
• /
• 2011

Cervical vertebrae are a complex structure and an important part of human body connecting the head and the trunk. In this paper, we propose a method to extract sternocleidomastoid muscle from ultrasonography images of cervical vertabrae automatically. In our method, Region of Interests(ROI) is extracted first from an ultrasonography image after removing unnecessary auxiliary information such as metrics. Then we apply Ends-in search stretching algorithm in order to enhance the contrast of brightness. Average binarization is then applied to those pixels which its brightness is sufficiently large. The noise part is removed by image processing algorithms. After extracting fascia encloses sternocleidomastoid muscle, target muscle object is extracted using the location information of fascia according to the number of objects in the fascia. When only one object is to be extracted, we search downward first to extract the target muscle area and then search from right to left to extract the area and merge them. If there are two target objects, we extract first from the upper-bound of higher object to the lower-bound of lower object and then remove the fascia of the target object area. Smearing technique is used to restore possible loss of the fat area in the process. The thickness of sternocleidomastoid muscle is then calculated as the maximum thickness of those extracted objects. In this experiment with 30 real world ultrasonography images, the proposed method verified its efficacy and accuracy by health professionals.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.35 no.1 s.120
• /
• pp.36-47
• /
• 2007

This study intended to provide basic data for the future naming of outdoor areas in apartment complexes by conducting research into the causes, lengths of time, trends, and characteristics of the changes of outdoor space names of apartment complexes and by analyzing current images of outdoor space names. In order to explore the causes and intervals of changes of outdoor spaces in apartment complexes, related literature was surveyed, with on-site inspections being made to examine outdoor space names by time intervals. At the site selected for this study, a case study was conducted to determine the outdoor space names and to understand the trends in the changes of these name in the apartment complexes. Moreover, a questionnaire survey incorporating selected adjectives was administered to investigate the perceived images of outdoor space names and finished construction spaces. The results are as follows: 1. Causes driving changes in apartment outdoor space names included changes in apartment complex planning concepts, changes in the values and demands of apartment buyers, social values, and changes of brand. Intervals of changes were divided into two with the 1997 IMF financial restructuring of the Korean economy as the division. 2. In regards to trends in changes of apartment outdoor space names and their characteristics, various features applied to spaces have caused many changes, which resulted in the phrase $'\bigcirc\;\bigcirc+garden'$, expressions suggesting various images, and Korean names. 3. The results of the analysis of apartment outdoor space names and finished construction images revealed that there was a difference in all the spaces. This is thought to be mainly due to such issues as paved min, the use of ready-made goods, and figurative expressions. The results of the study above indicate that changes in outdoor space names of apartment complexes have been made simply as part of marketing strategies, with no consideration of the residents. Apartments, an important means of housing in Korea, account for more than 50% of Korea's total housing, and the naming of outdoor space serves as an essential element for residents. Therefore, outdoor spaces of apartment complexes should be named in consideration of the final appearance after construction is completed, with effort by the industry being required.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.7 no.1
• /
• pp.115-123
• /
• 2005

Quantitative understanding of spatial characteristics of the study site is a prerequisite to investigate water and carbon cycles in agricultural and forest ecosystems, particularly with complex, heterogeneous landscapes. The spatial characteristics of variables related with topography, vegetation and soil in Gwangneung forest watershed are quantified in this study. To characterize topography, information on elevation, slope and aspect extracted from DEM is analyzed. For vegetation and soil, a land-cover map classified from LANDSAT TM images is used. Four satellite images are selected to represent different seasons (30 June 1999, 4 September 2000, 23 September 2001 and 14 February 2002). As a flux index for CO₂ and water vapor, normalized difference vegetation index (NDVI) is calculated from satellite images for three different grid sizes: MODIS grid (7km x 7km), intensive observation grid (3km x 3km), and unit grid (1km x 1km). Then, these data are analyzed to quantify the spatial scale of heterogeneity based on semivariogram analysis. As expected, the scale of heterogeneity decreases as the grid size decreases and are sensitive to seasonal changes in vegetation. For the two unit grids where the two 40 m flux towers are located, the spatial scale of heterogeneity ranges from 200 to 1,000m, which correspond well to the climatology of the computed tower flux footprint.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.4
• /
• pp.43-48
• /
• 2018

This study was conducted to reproduce digital colour based on the lightness of the background and size of the colour stimulus so that colour can be similarly perceived under different conditions. With the evolution of display technologies, display devices of various sizes can now reproduce more accurate colour and enhanced images, thus affecting the overall quality of display images. This study reproduced digital colour by considering the visual characteristics of the digital media environment. To accomplish this, we developed a colour appearance model which distinguishes the properties of foveal and peripheral vision. The proposed model is based on existing research on the lightness of the background and size of the colour stimulus. Based on experimental results, an analysis of variance was performed in order to develop the colour appearance model. The algorithm and modelling were verified based on the proposed model. In addition, to apply this model to display technologies, a practical colour control system and a method for handling complex input images were developed. Through this research, colour conversion errors which might occur when the input image is converted to fit a specific display size are resolved from the perspective of the human visual system. As a result, more accurate colour can be displayed and enhanced images can be reproduced.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.7-12
• /
• 2012

Recently the medical field to efficiently process the vast amounts of information to decision trees, neural networks, Bayesian Networks, including the application method of various data mining techniques are investigated. In addition, the basic personal information or patient history, family history, in addition to information such as MRI, HRCT images and additional information to collect and leverage in the diagnosis of disease, improved diagnostic accuracy is to promote a common status. But in real world situations that affect the results much because of the variable exists for a particular data mining techniques to obtain information through the enemy can be seen fairly limited. Medical images were taken as well as a minor can not give a positive impact on the diagnosis, but the proportion increased subjective judgments by the automated system is to deal with difficult issues. As a result of a complex reality, the situation is more advantageous to deal with the relative probability of the multivariate model based on Bayesian network, or TAN in the K2 search algorithm improves due to expansion model has been proposed. At this point, depending on the type of search algorithm applied significantly influenced the performance characteristics of the extended Bayesian network, the performance and suitability of each technique for evaluation of the facts is required. In this paper, we extend the Bayesian network for diagnosis of diseases using the same data were carried out, K2, TAN and changes in search algorithms such as classification accuracy was measured. In the 10-fold cross-validation experiment was performed to compare the performance evaluation based on the analysis and the onset of high-risk classification for patients with HRCT images could be possible to identify high-risk data.

• Journal of the Korean Society of Costume
• /
• v.43
• /
• pp.203-224
• /
• 1999

As androgny presented by feminism still appear in contemporary fashion since mid 1980's feminism as a social phenomenon can be considered to habe influenced fashion history. This paper surveys the sex images of androgynous look that appeared in contemporary fashion into bisexual image and neutral images based on the feminist aspects of changes in contemporary fashion. As stated above bisexual images and neutral images of androgynous look shown in clothing are an attempt to reach one single integrated complete being by overacoming 'femininity' or 'masculinity' and combining the element of the two sexes. Consequently androgynous look creates new attraction by harmoniously coordinating characteristics of the two sexes instead of totally differentiating between masculinity and femininity that are represented in the form of clothing that has conceptual visual characteristics. Following are the conclusions of the reseach of this paper. First 'androgynous' as a feminist sex concept reflects the spirit of the times. Since the sex roles are divided and diversified and the concept of femininity is changing the concept of androgyny requiring both masculine and feminine characteristics in an essential sex concept for sucessful conduct of roles in a complex modern society. Second the integrated human image of androgynous look has led to a new culture with fashion trends that has been ahead of times by curing the functions of fashion to differentiate between men and women. Third androgynous look at the turn-of-the-centuy will not cease to exist but be a lasting fashion phenomenon while the sex concept presented by feminism has become the core to destrov the dichotomous fashion style of the 20th century. Fourth androgynous look is expressed in various manners in contemporary fashion since it shown how human beings accipt the adaptation method desperately required by the contemporary times. Androgynous look can be viewd as the desire for ultimate freedom that can be acquired by destroying the dichotomous sex concept and as the will to become a subject of the world as integrated human being. As we look into fashion culture creation of fashion is not the result of an accidental cause or an individual's work but the reflection of the spirit of the times. In this sense androgynous look introduced by a new sex concept to contemporary fashion has reflected the spirit of the times and led the cultural atmosphere and moreover it is an important fashion style that can characterize the contemporary times while lasting in the future.