This study presents a truck classification method using panoramic side-view images to meet the Ministry of Land, Infrastructure and Transport’s 12-category standard (types 4–12). The system captures a vehicle’s full side profile via a panoramic imaging device, ensuring complete wheel visibility. A YOLOv12-based deep learning model detects wheels, and image processing extracts their center coordinates. Pixel distances between adjacent wheels are calculated and normalized to determine axle spacing patterns, which, together with wheel count, are applied to a rule-based classifier. Tests on 1,200 real-world panoramic truck images (1,000 for training, 200 for testing) achieved a mean average precision of 96.1% for wheel detection and 90.5% overall classification accuracy. The method offers explainable classification through measurable structural features, supporting applications in smart tolling, road usage billing, overloading enforcement, and autonomous vehicle perception.

Overloaded and improperly loaded trucks cause serious road hazards, such as rollovers and cargo falls. Although automatic enforcement methods are being studied, they face challenges in accuracy and legal application. Thus, a technology for direct tracking and enforcement is needed. This study uses EfficientNet to extract features of vehicles and license plates, and applies cosine similarity to identify the same vehicle. Comparisons were divided into “same vehicle” and “similar vehicle,” with a threshold-based method and five classification types. Results showed that the average similarity of the same vehicle group was 0.11 higher than that of the similar vehicle group. The accuracy of correctly identifying the same vehicle was 84.54%. Integrating OCR or LPR is expected to further improve tracking performance.

Passive acoustic monitoring (PAM) has emerged as an effective tool for studying underwater soundscapes and monitoring marine organisms. In this study, the biological sounds of three fish species that mainly inhabit or occur in the Korean coastal oceans, brown croaker (Miichthys miiuy), Pacific cod (Gadus macrocephalus), and small yellow croaker (Larimichthys polyactis) were recorded using the PAM method. The possibility of automatic classification was evaluated using a deep learning-based convolutional neural network (CNN) model based on the measured data. The biological fish sounds were recorded using hydrophones in the sea cage environments. The three fish species data were converted into spectrogram images and used as input for training and evaluating the CNN model. Gaussian noise was added to the test data to simulate low signal-to-noise ratio (SNR) environments. The model achieved high classification performance, with F1-score of about 96% on raw data and about 77% accuracy under signal-to-noise ratio conditions. These results suggest that CNN-based models be adequate for fish sound classification, even in acoustically complex underwater environments. Applying CNN models to classify and detect fish sounds can improve the automation and efficiency of PAM-based acoustic analysis, thereby improving the monitoring of fish populations, resource assessment, and ecological management in the future.

최근 2년간 중국 정부가 적극적으로 확대해 온 비자면제 정책은 단순한 출입국 절 차 간소화를 넘어 국가 이미지 구축 전략의 핵심 요소로 자리잡고 있다. 비자면제 정책으로 외국인 관광객이 급증하고 관련 영상이 소셜미디어에서 화제가 되는 시점 에서 본 연구는 비자면제 정책 이후 외국인의 중국에 대한 인식의 변화 여부를 검증 하였다. 이를 위해 우선, 중국 정부의 공식 발언 및 발표문의 분석을 통해 중국이 비 자면제 정책을 긍정적인 국가 이미지 구축의 수단으로 활용하고 있음을 밝히고, 실 제로 외국인이 중국 방문을 계기로 긍정적인 방향으로 인식의 변화가 생겼는지, 이 들이 가진 중국의 이미지는 무엇인지 검증하는 연구를 진행하였다. 구체적으로는 중 국이 비자면제 정책을 채택한 2023년 1월 1일부터 2024년 12월 21일까지의 기간 동안 유튜브에 업로드 된 동영상들 중 100개를 선별하여 영상의 내용을 대상으로 주제 적 측면, 표현 형식, 감정적 측면, 인지적 측면이라는 네 가지 차원의 내용분석을 진 행하였다. 분석 결과, 중국을 방문한 외국인들의 중국에 대한 긍정적인 인식이 강화 되었다는 점을 확인하였다. 이는 현재까지 중국 정부의 전략과 실제 외국인의 인식 이 상당 부분 일치한다는 점을 보여준다.

Despite its historical association with oppression, fear, and marginalized femininity, modern reinterpretations of the figure of the witch position it as a symbolic entity combining historical context and literary imagination. This study analyzes contemporary fashion designs that incorporate witch imagery, focusing on the designs' representational methods and aesthetic characteristics. It includes a literature review examining evolving perceptions of witches across different eras as well as visual and formal analyses of post-2010 fashion collections, using image and text materials from Showstudio.com and other online sources. In various realms of visual culture–including art, film and drama, advertising, product design, and halloween costumes–the image of the witch is no longer confined to traditional iconography. Instead, it is continuously reimagined, taking on expanded and multifaceted meanings that attest to its unique and evolving nature. This study identifies two primary expressive types: direct iconographic appropriation, employing traditional motifs such as pointed hats, black cloaks, and pentacles; and metaphorical suggestion, using dark tones, deconstructed silhouettes, and layered styling to imply the witch's essence. The analyses also highlight three aesthetic characteristics: empowered sensuality, expressed through sheer fabrics and dramatic makeup; supernatural mysticism, combining natural and occult symbols; and uncanny playfulness, which subverts traditional motifs through fantasy and humor. Ultimately, the witch in contemporary fashion serves as both a stylistic code and a visual signifier of cultural narrative and identity. This reimagining has contributed to the development of new aesthetic systems in fashion and broadened the visual culture and design innovation discourse.

Fault detection in electromechanical systems plays a significant role in product quality and manufacturing efficiency during the transition to smart manufacturing. Because collecting a sufficient number of datasets under faulty conditions of the system is challenging in practical industrial sites, unsupervised fault detection methods are mainly used. Although fault datasets accumulate during machine operation, it is not straightforward to utilize the information it contains for fault detection after the deep learning model has been trained in an unsupervised manner. However, the information in fault datasets is expected to significantly contribute to fault detection. In this regard, this study aims to validate the effectiveness of the transition from unsupervised to supervised learning as fault datasets gradually accumulate through continuous machine operation. We also focus on experimentally analyzing how changes in the learning paradigm of the deep learning model and the output representation affect fault detection performance. The results demonstrate that, with a small number of fault datasets, a supervised model with continuous outputs as a regression problem showed better fault detection performance than the original model with one-hot encoded outputs (as a classification problem).

This study proposes a real-time content design pipeline optimized for Unreal Engine, integrating generative AI-based image creation with AI-assisted 3D modeling tools. The pipeline aims to streamline the production of high-quality assets for real-time applications, including games and simulations. Two types of subjects were selected: a bust combining organic character features, and a stone slab characterized by planar and symmetrical structure. Multi-angle image data were first synthesized using advanced generative AI models to simulate diverse viewpoints. These were then processed using AI-enhanced photogrammetry and modeling tools to reconstruct detailed 3D meshes and extract base textures. Post-processing steps, including mesh decimation, UV unwrapping, and texture baking, were performed to ensure compatibility with Physically Based Rendering (PBR) workflows used in Unreal Engine. The final assets were successfully imported into Unreal Engine, demonstrating visual fidelity and performance suitability in a real-time environment. The study confirms the pipeline’s potential for accelerating asset development and suggests promising future directions in AI-driven digital content creation.

This study structurally analyzes the algorithmic filtering process by which generative AI images are either selected or discarded before reaching users, and models this process through a visual similarity–based simulation. Images generated by Stable Diffusion are placed on a two-dimensional grid, and a modified version of Conway’s Game of Life algorithm is applied to update the state of each cell. The survival of each cell is determined based on a hybrid visual similarity metric combining CLIP and LPIPS. To prevent the rigidity of the simulation and sustain emergent dynamics, random image injections are periodically introduced. The simulation results reveal that visually similar images repeatedly form clusters, and a visual order gradually converges toward a structurally stabilized state. This suggests that specific visual orders can emerge solely from algorithmic selection criteria, independent of human interpretation. By shifting focus from semantic or symbolic analysis to the experimental conditions for the existence and persistence of images, this study proposes a new analytical perspective for understanding digital image environments.

OSMU 의 원천 소스로써 문학 자원의 효용이 널리 인정받은 이래로 문학진흥법 시행 및 국립한국 문학관 건립 추진 등의 이벤트가 잇따르며 지역의 역사․문화 및 경제 발전을 일으키는 문학 테마 공간에 대한 관심이 고취되고 있다. 본 연구는 문학관, 문학공원 등 문학을 테마로 조성된 조경 공간 의 테마가 의도에 부합하게 전달되고 있는지 알아보기 위해 스토리 전달의 핵심 매개물인 환경조형 물의 이미지텔링 현황을 종합적으로 검토하였다. 이를 위해 문헌 조사와 현장 조사를 실시하여 황 순원문학촌 소나기마을, 박경리문학공원, 김유정문학촌의 3곳에 설치된 문학촌 조형물의 이미지 텔링 현황을 평가하고 선행 연구를 바탕으로 작성한 분석의 틀을 적용하여 평가의 결과와 개선 방 향을 도출하였다. 본 연구는 향후 문학 테마 공간에 설치된 조형물의 이미지텔링 이해를 돕고 이미 지텔링의 개념으로 수용되는 환경조형물의 설치 계획 수립 및 심의와 평가에 참고할 수 있는 기초 자료를 제공하고자 한다.

선박에는 단열을 위한 발포제가 적용된다. 기존의 발포제에는 지구온난화물질인 수소불화탄소(HFC)를 다량 포함하고 있는 문제점이 있으며, 우리나라는 몬트리올 의정서의 ‘키칼리 개정서’를 채택함에 따라 HFC를 ‘24년부터 ’45년까지 기준 수량의 80% 감 축하기로 결정되었다. 이에, 메틸포메이트 원료는 지구온난화지수가 0(HFC는 960~1,430)으로 향후 친환경발포제로 높은 기대를 갖고 있다. 하지만, 메틸포메이트 발포제의 성능은 원료의 순도 및 주변환경에 높은 영향을 받음으로 각 공정환경에 대한 정확한 분류가 필요하다. 이에, 본 논문에서는 주변환경(온도)과 메틸포메이트 순도에 따라, 총 4개의 케이스를 만들었다. 각 케이스에 대해서 10,010 장의 이미지를 학습하고, 이를 구글넷(GoogLeNet)알고리즘을 이용하여 분류하였다. 분류결과 정확도는 96.8%를 갖고, F1-Score는 0.969 를 갖는 것으로 계산하였다.

This study analyzes the visual image effects regarding the design changes of sleeves and shoulders on plus-size women in their 20s and 30s to propose ideas for plus-size women’s clothing production and design. Based on a plus-size women’s basic raglan sleeve shirt, six types of raglan sleeve shirts with various design elements were developed using a 3D virtual imaging program to analyze the visual image effect of each pattern’s features. The results showed that the A-shaped sleeve silhouette, exemplified by Pattern A2, makes the upper body look longer and slimmer as a whole. It also conveyed a sense of refinement and attractiveness, displaying positive effects regarding body shape and aesthetic perception. Meanwhile, O-shaped silhouettes, such as Pattern B1 and Pattern B2, showed positive effects regarding comfort and mobility, while Y-shaped silhouettes, such as Pattern C1 and Pattern C2, created a visually confident appearance. Furthermore, the research examined the effects of static images and dynamic videos on evaluation outcomes. The results showed that dynamic images enhanced immersion and vibrancy, thereby increasing focus on the apparel evaluated. The results of the visual image evaluation can be used as a basis for designing not only raglan sleeves but also overall sleeve shapes.

Today, as the social demand for tap water safety and the need for an ICT-based intelligent integrated control system increase, K-water (Korea Water Resources Corporation) is building and operating a ‘Water pipe monitoring CCTV system’ to quickly respond to crises in the event of a water leak. However, in the case of the existing system, when the CCTV rotates, the image information and the mapped water pipe image do net match, so the operator has the limitation that the water pipe image must be mapped anew every time. In this paper, in odert to solve the above problems, we propose an improved system that can extract feature points from CCTV images, detect changes in the coordinate values of the feature points, and automatically transform the location of the water pipe image by utilizing LoFTR (Detector-Free Local Feature Matching with Transformer), a type of deep learning image matching algorithm that is actively being studied in th field of the latest computer vision, and examine its effectiveness.

This study developed and implemented an elementary English writing program using image-generative AI to examine its effects on students’ writing abilities and affective domains. The program was designed following the ADDIE model and aligned with the 2022 Revised English National Curriculum. It used Padlet’s ‘I Can’t Draw,’ allowing students to create images based on their prompts. Eighty sixth-grade students participated, with data collected through pre- and post-writing tests, affective domain surveys, student work, reflections, and teacher interviews. A mixed-methods analysis, combining ANCOVA and content analysis, revealed the following: (1) The program improved the syntactic and lexical complexity and fluency of students’ writing, but not accuracy. (2) Students’ situational interest significantly increased, but individual interest did not. Additionally, students’ engagement and motivation in learning English improved. These findings indicate that image-generative AI is an effective tool for enhancing English writing lessons and providing feedback. The study also emphasizes the critical role of teacher competence in successfully integrating AI into language instruction.

이 연구는 직장 여성의 속눈썹 연장 시술 동기 및 이미지 변화 자신감과 시술 효과성에 미치는 영향의 필요성에 대해 검증하고자 하였다. 이 연구는 서울 및 지방도시 지역에 거주하는 20세 이상의 직장 여성을 대상으로 2024년 5월 10일부터 2024년 6월 5일까지 최종 286부를 SPSS 25.0 프 로그램을 활용하였다. 분석결과 속눈썹 연장 시술 동기의 하위요인별 외모 단점 보완, 이미지 개선은 심리적 효과성에 영향을 미치는 것으로 나타났고, 속눈썹 연장 시술 동기의 하위요인별 외모 단점 보완, 이미지 개선, 대인 지향은 사회적 효과 성 영향을 미치는 것으로 나타났다. 분석결과 직장 여성들의 삶의 질 향상에 속눈썹 연장 시술이 전체적인 이미지 변화에 매우 효과적이며 한국 여성들의 추구하는 세련된 이미지 변화에 만족도를 높일 수 있는 미용산업의 한 분야 로 판단된다. 갈수록 심화 되는 미용산업의 경쟁구도 속에 자리 잡고 있는 속눈썹미용 이 더욱 체계화되고 전문화시킬 수 있는 교육과 전문가 양성, 프랜차이즈 개 발, 프로그램 개발이 필요하다고 생각된다.

본 논문에서는 히니가 어떻게 저장소로서의 의미와 아일랜드의 집단적 기억이 퇴적된 공간으로서의 의미를 가지고 있는 늪의 이미지를 통해 북아일랜드에서 일어나는 폭력에 대해 사유했는지 보여주고 있다. 늪 시편에서 시인은 고고학자가 되 어 늪을 발굴하면서 늪에 있는 희생자를 다시 끄집어내어 이들이 자신의 행위에 대한 심판을 받게 한다. 이러한 발견의 과정은 또한 히니가 1974년 에세이 “Feeling into Words”에서 썼듯 시적 창작 과정에 대한 은유이기도 하다. “The Tollund Man,” “Punishment,”와 “Kinship”이라는 시에서 히니는 과거와 현재의 폭력을 나란히 비교하 면서 어떻게 두 시대에서 (더 이상 유효하지 않은) 믿음에 기반한 폭력이 자행되는지 보여준다.

Evaluating the performance of asphalt concrete using CT scanning has become an essential area of research due to its potential to revolutionize the way we assess road materials. Traditional methods often require destructive sampling, which can damage infrastructure and offer limited insight into the material's internal structure. In contrast, CT scanning provides a non-destructive, highly detailed analysis of asphalt's internal features, such as air voids, aggregate distribution, and binder coverage, all of which are critical to its durability and performance. Additionally, the ability to create 3D models from CT scans allows for deeper insights into factors like void connectivity and aggregate bonding, which directly affect the lifespan of pavements. By combining CT imaging with advanced data processing techniques, such as deep learning, this research offers more accurate and reliable methods for optimizing asphalt mix designs, ultimately leading to longer-lasting roads, reduced maintenance costs, and more sustainable construction practices.

This study was carried out to establish various physiological changes according to soil water stress and to compare the degree of water stress between two species of grapevines (‘Jinok’ as a new breeding cultivar and ‘Campbell Early’ as a control) using thermography. Soil water potentials were treated at -70, -30, and -5 kPa with waterlogging for 7 days. Regarding the photosynthetic rates (A) of the two cultivars, they showed an order of –30 kPA > -5 kPa > -70 kPa in order. With -70 kPa and waterlogging treatments, a decrease of photosynthetic rate was observed at 3 days after treatment, with a more significant decrease accumulating over time. At 7 days after treatment, photosynthetic rates of ‘Campbell Early’ (33.3, 45.6%) and ‘Jinok’ (56.6, 57.3%) grapes decreased compared to those with -30 kPa treatment. H2O2 and proline synthesis were the highest with the waterlogging treatment. In terms of proline synthesis, ‘Campbell Early’ had a relatively higher rate than ‘Jinok’. Leaf and stem water potential were the lowest with the -70 kPa treatment and the highest with the - 30 kPa treatment f or both cultivars. Crop water stress index (CWSI) showed the following order: waterlogging > -70 kPa > -5 kPa > -30 kPa, which was the opposite result of water vapor transfer (IG). As a result of correlation analysis between factors, photosynthetic rate showed negative correlations with the water potential of leaf and stem and crop water stress index but a positive correlation with the relative water content of leaves. Thus, tolerance to water stress of ‘Campbell Early’ was relatively stronger than that of ‘Jinok’ grape. It is possible to compare water stress using infrared imaging.

미국 메트로폴리탄 미술관에는 청 황제 건륭제(乾隆帝, 재위 1735~1796)가 발문한 사 슴뿔 그림인 <녹각도(鹿角圖)>가 소장되어 있다. 기존 연구에서 이 그림은 건륭제가 서양의 회화 주제인 사슴뿔을 서양 화법을 적용하여 그림으로써 서양풍을 모방하고 자신의 목적에 맞게 전용하고자 한 시도로 이해되었다. 본고는 관점을 달리하여 <녹각도>의 상서적(祥瑞 的) 요소를 발견하고 청 황제들의 대표적인 사냥감이었던 사슴의 뿔이 사냥 전리품에서 산령 (山靈)이 보낸 상서로운 징표, 즉 부서(符瑞)로 변모한 과정을 고찰한다. 이를 위해 우선 북 경ㆍ심양고궁박물원에 소장된 홍타이지, 강희제, 건륭제가 제작한 녹각의(鹿角椅)를 통해 청 황제들에게 있어 사슴뿔이 만주족의 전통과 용맹함을 보여주는 사냥 전리품이었음을 확인 한다. 또한 <녹각도>에 실린 건륭제의 발문인 ‘녹각기(鹿角記)’를 분석하여 이 사슴뿔이 어떻게 부서가 되었는지 밝힌다. <녹각도>에서 건륭제는 사냥감보다 서수(瑞獸)로서의 사슴 을 강조함으로써 청 황실의 사냥 의식이 중시한 만주적 무용(武勇) 위에 한족의 전통 사상인 천인감응설(天人感應設)에 의거한 부명(符命)의 가치를 덧입혔다. 이처럼 <녹각도>에서는 사냥을 자연의 정복이 아닌 자연과의 교감으로 해석하여 만주 황권을 성군의 경지로 끌어올 리고자 한 건륭제의 의도가 읽힌다.