• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.38 no.1
• /
• pp.18-25
• /
• 2022

Purpose: This study aimed to evaluate the color stability and antibacterial properties of the surface of polyethyl methacrylate (PEMA) resin with zirconia nanoparticles added. Materials and Methods: The control group was pure PEMA resin, and the experiment group was PEMA resin 15 mm in diameter and 2.5 mm in thickness disk-shaped specimens with 2, 4 and 8 w/v% of zirconia nanoparticles added, which were respectively divided into Group Z2, Group Z4, and Group Z8. After analyzing the surface roughness and color stability of the specimens, their antibacterial properties were evaluated using Porphyromonas gingivalis (P. gingivalis). The Statistical analysis was performed using when normality was met in the Shapiro-Wilk test, one-way ANOVA was used to test parameters, and Tukey's test was used as a post hoc test. When normality was not met, the Kruskal-Wallis test, a non-parametric test was used (P < 0.05). Results: The surface roughness measurement found that there was no significant difference between the experimental and control groups. The color stability evaluation showed that the Z2, Z4, and Z8 groups were within the color range of natural teeth. The adhesion of P. gingivalis was evaluated to be significantly reduced in Group Z2 compared to the control group (P < 0.05). In the Z2 group, Z4 group, and Z8 group, dead cells bacteria than the control group were observed. Conclusion: In conclusion, PEMA resin with zirconia nanoparticles added was within the range of natural teeth in color and reduced the adhesion of P. gingivalis.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.329-352
• /
• 2022

It is reported that particulate matter(PM) penetrates the lungs and blood vessels and causes various heart diseases and respiratory diseases such as lung cancer. The subway is a means of transportation used by an average of 10 million people a day, and although it is important to create a clean and comfortable environment, the level of particulate matter pollution is shown to be high. It is because the subways run through an underground tunnel and the particulate matter trapped in the tunnel moves to the underground station due to the train wind. The Ministry of Environment and the Seoul Metropolitan Government are making various efforts to reduce PM concentration by establishing measures to improve air quality at underground stations. The smart air quality management system is a system that manages air quality in advance by collecting air quality data, analyzing and predicting the PM concentration. The prediction model of the PM concentration is an important component of this system. Various studies on time series data prediction are being conducted, but in relation to the PM prediction in subway stations, it is limited to statistical or recurrent neural network-based deep learning model researches. Therefore, in this study, we propose four transformer-based models including spatiotemporal transformers. As a result of performing PM concentration prediction experiments in the waiting rooms of subway stations in Seoul, it was confirmed that the performance of the transformer-based models was superior to that of the existing ARIMA, LSTM, and Seq2Seq models. Among the transformer-based models, the performance of the spatiotemporal transformers was the best. The smart air quality management system operated through data-based prediction becomes more effective and energy efficient as the accuracy of PM prediction improves. The results of this study are expected to contribute to the efficient operation of the smart air quality management system.

• Korean Journal of Heritage: History & Science
• /
• v.56 no.2
• /
• pp.264-279
• /
• 2023

The 50th anniversary of the UNESCO World Heritage Convention was in 2022. In order to reflect on the present and future of the meaning of World Heritage, this paper examines the development and changes of the UNESCO World Heritage system. After promulgating the convention in 1972, the UNESCO World Heritage system prioritized the protection of heritage sites in the world that were at risk due to armed conflicts and natural disasters to bequeath heritage to the next generation. In addition, the UNESCO World Heritage's emphasis on Outstanding Universal Value represents the particular culture of human beings formed during a certain period of time, and acts as a significant source of soft power in public diplomacy. The UNESCO World Heritage might be perceived as a shared heritage that has not only become a channel to understand various national values, but also an effective medium to convey one of UNESCO's main principles, that is, peacebuilding. However, the UNESCO World Heritage is now at the center of conflicts of heritage interpretation between many stakeholders related to invisible wars, such as cultural wars, memory wars, and history wars as the social, political, and cultural contexts concerning World Heritage have dramatically shifted with the passing of time. Paying attention to such changing contexts, this paper seeks to understand the main developments in UNESCO World Heritage's discourse concerning changes to the World Heritage Operation Guidelines and heritage experts' meetings by dividing its 50-year history into five phases. Next, this paper analyzes the main shifts in keywords related to UNESCO World Heritage through UNESDOC, which is a platform on which all UNESCO publications are available. Finally, this paper discusses three main changes of UNESCO World Heritage: 1) changes in focus in World Heritage inscriptions, 2) changes in perception of World Heritage protection, and 3) changes of view on the role of the stakeholders in World Heritage. It suggests new emerging issues regarding heritage interpretation and ethics, climate change, and human rights.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.6-22
• /
• 2022

In this paper classifies the types of 228 anchor stones discharged from the west and south coasts, assumes a combined method by type, routes through discharge locations, and It attempted to estimate the burial site. Prior to classification of types, the weight, thickness, width, and length of the anchor stone were measured, and the largest Young in the tomb The scent weight was classified into I~V groups, and the shape of the anchor stone was classified into 1-6 types. All of these weight and shape correlations It was classified into 17 types. The combined method by type is 180kg or less depending on the morphological characteristics of the reference value of the extracted anchor stone, and the type An anchor of type 1 or 2 is used in combination with an anchor, and a anchor of type 3 or 6 weighs more than 180kg and is combined with an anchor The dragon was assumed to be an anchor. Along with this, the route and burial site are identified through past records and testimony of local residents It was checked against the data. The route was largely consistent with past records, but the new route was apparent in waters near the island or inland It was also drawn. In the case of burial sites, small and large in Taean Mado Sea and Jindo Byeokpajin Sea, where anchor stones are concentrated, Considering the pattern of the type anchor stones being discharged, it was consistent with the testimony that ships of various sizes were mysterious in the two seas. Based on this type classification, a study on the spatiality of the anchors was conducted. First, a comparison and analysis was done on whether actual real data, such as anchor stones, old ships, and relics, were identified on the Joun-ro route and international trade routes as recorded in the past literature. Where there was no record, the route was estimated based on real data. To this end, routes estimated based on the testimony of local residents and modern ship workers were analyzed as to whether ships traveled there in the past and whether they could actually sail. Next, the location of each seedling was estimated by ship size according to the weight of the anchor stone. In the case of the Taean Mado Sea and Jindo Byeokpajin Sea, both small and large anchor stones were discharged from the coastline and were far away.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.24-36
• /
• 2022

The Bonghwang-dong ruins in Gimhae, the central area of Geumgwan Gaya, is presumed to be the site of the royal palace, and excavations have been in progress at the Gaya National Cultural Heritage Research Institute. According to a research conducted by lowering the level to the base layer on the north side of the site, mostly shell layers composed of oysters were confirmed, and soil composed of different material was alternately filled in to form a site construction. In other words, it can be seen that there was work at the site of the Bonghwang-dong ruins that required large-scale labor, such as building ramparts and embankments. There is stratigraphic confusion such as showing different age values in the same shell layer through a chronological analysis of organic matter and charcoal in the sedimentary layer, and deriving a result value in the upper layer ahead of the lower layer. In addition, open-sea diatoms are observed not only in the sedimentary layers, but also the pits. Therefore, it is judged that the soil constituting the ruins was brought from the outside. The Bonghwang-dong ruins are located inside the commonly called Bonghwang earthen ramparts, where many excavation organizations conducted research within the estimated range of the earthen fortifications. As a result, it was found that it was similar to the sedimentary layers of the ruins of the Three Kingdoms Period, which were investigated along with the ruins of Bonghwang-dong. Through this, the surrounding ruins, including those of Bonghwang-dong, were located close to paleo-Gimhae Bay, so it is believed that the soil brought from the surroundings was used to reinforce the ground. As a result of the excavation research on the Bonghwang-dong ruins conducted so far, it was found by sedimentary layer analysis and soil experiments that the ruins were created on stable land. Relics excavated in the sediments of the ruins and carbon dating data show that Bonghwang-dong carried out large-scale civil construction work in the 4th century to build the site, which clearly shows the status of Geumgwan Gaya.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.56-70
• /
• 2022

Tombs No. 39 and No. 63 in Gyo-dong, Changnyeong, are unique in the Changnyeong area in that they do not have additional burials because they are hoenggu-style tombs with an entrance on the north side. This study tried to understand the nature and meaning of the two separate stone walls at the entrance of the tomb along with the burial process of the tomb. These two stone walls mark small tombs built independently within a large tomb, with stone wall No. 39 stone wall No. 3 (No.39-3) and No. 63 stone wall No. 3 (No. 63-3). Both units are located in the middle of the northern wall of the burial body part and share one wall with the burial body part wall stone. All animal fluids inside the stone wall were identified. In particular, it was estimated that at least three dogs were buried as a result of identification of animal fluids No. 63-3. Above all, these animals have their heads facing outward with their backs to the main occupants, and do not overlap in a limited space and are placed side by side. Changnyeong Gyo-dong No. 39-3 and 63-3 were created in the process of building the burial body, and although they are independent relics, they form a subordinate relationship in that they were built along the main burial within one tomb. In addition, it is coercive in that it is placed in an orderly manner according to a certain direction in a state that has not been dismantled after killing an animal. Therefore, It is understood to be the SHUNJANG of dogs. Studies on animal fluids excavated from tombs in the Three Kingdoms period are interpreted as animal stewardship, sacrificial collection, and animal sacrifice depending on their location, and this is known as a series of animal sacrifice rites, namely, animal stewardship and sacrifice. This recognition is based on material objectification of animals, such as food or sacrifices. However, Changnyeong Gyo-dong No. 39-3 and 63-3 are different in that they recognize animals as spiritual beings in the process of funeral rites and are closely related to the ideology that there is life after death. In addition, analysis of the location and directionality of the remains is also required from multiple angles. These two SHUNJANG correspond to the entrance to the tomb, and the location is the most open space at the entrance. The appearance of a dog looking outward, etc., can also be interpreted as the meaning of protecting the tombs and byeoksa. This appearance can be compared with the dog depicted in a mural in a Goguryeo tomb that reflects the ancient world's thought and stone figures excavated from the tomb of King Muryeong of Baekje, and it is also consistent with the meaning of the JINMYOSU protecting the ancient tombs. This suggests that a multifaceted study on animal fluid burial remains is needed in the future.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.72-94
• /
• 2022

This study is about the architectural changes over time in Dongchun-gotaek, the house of Song Jun-gil, one of the representative figures of the Hoseosarim(湖西士林), also an architectural cultural asset representing Daejeon. Data related to Sangryang(上樑) found in the restoration process of Dongchundang(同春堂) and DongchundangJongtaek(同春堂 宗宅), the state designated heritage application report written by Daejeon Metropolitan City, Deokeun-gaseung(德恩家乘), a book which has been handed down from generation to generation from Dongchundang Munjeonggongpa(同春堂 文正公派) of Eunjin Song's Clan and the results of partial excavation surveys respectively conducted in 2010 and 2020 were used as basic research data, and these data were compared and analyzed to examine the changes of the arrangement of Dongchundang, Jeongchim(正寢), and ancestral shrine buildings. Dongchundang was built by Song Jun-Gil. Rather than a new building, it was a building that was basically relocated to its current area when Cheongjwawa(淸坐窩), which was built by his father, dilapidated, and the timing of its construction can be clarified through Sangryangmun(上樑文). However, in the estimated area of Cheongjwawa, no exact site was found in two surveys of buried cultural heritages. In the case of Jeongchim, it was possible to confirm that it had been relocated two times, and it can be said that the biggest achievement of this study was to confirm that the first relocation was outside the current fence. In addition, one of the building sites which was identified in the excavation survey for confirming the servants' quarters was estimated to be the first construction at the site of Dongchun-gotaek. In the shrine area, there were the first constructed Gamyo(家廟), including Byeolmyo(別廟) dedicated to Bulcheonwi(不遷位), and Jomyo(祧廟) dedicated to Checheon(遞遷), and it can be seen that it was a space where many changes such as new construction, demolition, or mutual exchange of location occurred over time. The present buildings arrangement through these processes was not far from the original plan of Song Joon-gil. Therefore, the name of 'Dongchun-gotaek' is appropriate.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.96-116
• /
• 2022

The Daebodan was an altar, which held a memorial service for emperors of the Ming dynasty. This alter, which was referred to as Hwangdan, was first constructed in 1704. When the Japanese Invasion of Korea commenced in 1592, Shinjong, the emperor of the Ming dynasty, sent reinforcements to Josun to help. This alter was made to repay Shinjong's kindness. Before this, Song-siyeol(宋時烈), Leader of Noron(老論), made a shrine at Hwayangdong to hold memorial services for Shinjong, and after some time, this developed into a national ceremony. Construction of the Daebodan largely changed the backyard of Changdukgung-palace. However considering the construction process, the meaning of the Daebodan was not a big deal. At first, the optimal place for the Daebodan was selected at the site of a inner icehouse. But the inner icehouse could not be transferred to other site due to the circumstances. After all, the Daebodan was constructed at the site of Byeoldaeyeong(別隊營) which was located in the outside of palace. Most of the stones for the new Daebodan were used ones. And the annexe of Byeoldaeyeong was used for Daebodan without any changes being made. The scale of the construction was not particularly grand. After the construction, Sukjong, who made the Daebodan, showed barely any interest in it. But the conception of the Daebodan was back again in the history by Youngjo. He was also not interested in the Daebodan during his early years of ruling time. However, in the 1740's, he started to become interested in the ceremony of Daebodan, and carried out a large-scale reconstruction of the Daebodan. Jegigo(祭器庫) was rebuilt In 1739. And Jaesil(齋室), staying one night before the ceremonial day, was added in 1745. In 1749, the Daebodan was greatly changed by enshrining Uijong and Taejo, emperors of the Ming dynasty. The shape of alter was changed. Moreover this alter was made by newly quarried stones. And several buildings, Junsachung(典祀廳), Jaesaengchung(宰牲廳) and Akgongchung(樂工廳), were added to the site. In 1762, meritorious retainers were enshrined to the Daebodan. After all the Daebodan became an important part of the backyard of Changdukgung-palace. During the reign of Jungjo, the Daebodan also was an important part of backyard of Changdukgung-palace. But significant changes were not made at that time. The only change was the moving of Kyungbonggak(敬奉閣) in 1799. Afterward the Daebodan remained unchanged. The ceremonies at the Daebodan stopped in 1908. And the Daebodan disappeared into the mist of history.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.118-135
• /
• 2022

The primary purpose of this study is to estimate the structure of Byeongpungsadaseok (屛風莎臺石) and Nanganseok (欄干石) in Geonwonneung (建元陵) and Heonneung (獻陵), which were built in the early of 15th century, based on the Annals of King Sejong. In addition, the ultimate purpose of this study is to reveal structural changes and their significance by comparing the differences with the contents of the dismantlement survey. Geonwonneung, Jereung(齊陵), and Heonneung were repaired at the same time in 1442, and the structural changes were the similar. The purpose of the repair in 1442 was to prevent water from flowing into the underground palace with smooth drainage. As a result of estimating the structure of Geonwonneung and Heonneung according to the records of the Annals of King Sejong, it was created in a very similar form to the Hyeonneung and Jeongneung of Goryeo. And it was clearly recognized that the Royal Tomb of Goryeo was followed. However, as the structure was improved in 1442, the unique characteristics of the Royal Tomb of the Joseon Dynasty were formed. First is the appearance of the Bokbuhyeong lime (覆釜形石灰, which is a convex roof on the Byeongpungsadaseok that serves to prevent rainwater from penetrating into the burial mound. It also plays a role in connecting and fixing the Manseok (滿石) and the Inseok (引石), which are the upper structures of the Sadaseok (莎臺石). Second, the Bakseok (薄石) between the nanganseok and the sadaseok has been transformed into the Sangseok (裳石) with a slope. This plays a role in protecting the inner stone chamber by expanding the length of the bakseok, which forms an overall slope, like the eaves of the roof. After both of these features were first attempted in 1442, they were applied to all Royal Tombs of the Joseon Dynasty and became unique features of these Royal Tombs.