• 디지털융복합연구
• /
• 제19권5호
• /
• pp.197-205
• /
• 2021

최근 리테일 산업계에서는 COVID-19 등의 다양한 외부 환경 위협으로부터의 대응과 AI 기술을 활용한 경쟁력을 갖추기 위한 정보기술 융합 및 활용 요구가 증가하고 있으나 리테일 산업에서의 데이터를 활용하기 위한 연구와 응용 서비스의 활용사례가 매우 부족하다. 본 연구는 CCTV 영상 데이터 기반의 AI 활용 응용 서비스 활용 사례연구로 리테일 공간에서의 CCTV 영상 데이터 수집, 객체 탐지 및 추적 AI 모델 활용, 실시간 추적된 객체와 트래킹 데이터를 저장하기 위한 시계열 데이터베이스 활용, 시계열 데이터를 활용한 모니터링, 리테일 공간의 혼잡도와 관심도를 분석하기 위한 히트맵, 리테일 공간에서의 실시간 상황 모니터링, COVID-19와 같은 사회적 위협으로부터의 접촉자 분석과 추적을 위한 사회적 거리 두기 현황, 비인가자의 보안 구역의 접근 모니터링 애플리케이션을 설계하고 이를 실제 구현하여 리테일 공간에서의 CCTV 영상 데이터를 활용한 애플리케이션 설계를 통해 CCTV 영상 데이터 기반의 AI 융합 응용 서비스 활용 모델을 제시하였으며, 실제 구현을 통해 설계된 활용 모델을 검증하였다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제10권5호
• /
• pp.187-194
• /
• 2021

사람의 행동인식(Action Recognition)은 사람의 관절 움직임에 따라 어떤 행동을 하는지 인식하는 것이다. 이를 위해서 영상처리에 활용되는 컴퓨터 비전 태스크를 활용하였다. 사람의 행동인식은 딥러닝과 CCTV를 결합한 안전사고 대응서비스로서 안전관리 현장 내에서도 적용될 수 있다. 기존연구는 딥러닝을 활용하여 사람의 관절 키포인트 추출을 통한 행동인식 연구가 상대적으로 부족한 상태이다. 또한 안전관리 현장에서 작업자를 지속적이고 체계적으로 관리하기 어려운 문제점도 있었다. 본 논문에서는 이러한 문제점들을 해결하기 위해 관절 키포인트와 관절 움직임 정보만을 이용하여 위험 행동을 인식하는 방법을 제안하고자 한다. 자세추정방법(Pose Estimation)의 하나인 AlphaPose를 활용하여 신체 부위의 관절 키포인트를 추출하였다. 추출된 관절 키포인트를 LSTM(Long Short-Term Memory) 모델에 순차적으로 입력하여 연속적인 데이터로 학습을 하였다. 행동인식 정확률을 확인한 결과 "누워있기(Lying Down)" 행동인식 결과의 정확도가 높음을 확인할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권2호
• /
• pp.692-712
• /
• 2022

Buoys used for Aid to Navigation systems are widely used to guide the sea paths and are powered by batteries, requiring continuous battery replacement. However, since human labor is required to replace the batteries, humans can be exposed to dangerous situation, including even collision with shipping vessels. In addition, Maritime sensors are installed on the route signs, so that these are often damaged by collisions with small and medium-sized ships, resulting in significant financial loss. In order to prevent these accidents, maritime object detection technology is essential to alert ships approaching buoys. Existing studies apply a number of filters to eliminate noise and to detect objects within the sea image. For this process, most studies directly access the pixels and process the images. However, this approach typically takes a long time to process because of its complexity and the requirements of significant amounts of computational power. In an emergent situation, it is important to alarm the vessel's rapid approach to buoys in real time to avoid collisions between vessels and route signs, therefore minimizing computation and speeding up processes are critical operations. Therefore, we propose Fast Connected Component Labeling (FCCL) which can reduce computation to minimize the processing time of filter applications, while maintaining the detection performance of existing methods. The results show that the detection performance of the FCCL is close to 30 FPS - approximately 2-5 times faster, when compared to the existing methods - while the average throughput is the same as existing methods.

• Smart Structures and Systems
• /
• 제31권4호
• /
• pp.421-436
• /
• 2023

This paper presents a vibration displacement measurement and damage identification method for a space truss structure from its vibration videos. Features from Accelerated Segment Test (FAST) algorithm is combined with adaptive threshold strategy to detect the feature points of high quality within the Region of Interest (ROI), around each node of the truss structure. Then these points are tracked by Kanade-Lucas-Tomasi (KLT) algorithm along the video frame sequences to obtain the vibration displacement time histories. For some cases with the image plane not parallel to the truss structural plane, the scale factors cannot be applied directly. Therefore, these videos are processed with homography transformation. After scale factor adaptation, tracking results are expressed in physical units and compared with ground truth data. The main operational frequencies and the corresponding mode shapes are identified by using Subspace Stochastic Identification (SSI) from the obtained vibration displacement responses and compared with ground truth data. Structural damages are quantified by elemental stiffness reductions. A Bayesian inference-based objective function is constructed based on natural frequencies to identify the damage by model updating. The Success-History based Adaptive Differential Evolution with Linear Population Size Reduction (L-SHADE) is applied to minimise the objective function by tuning the damage parameter of each element. The locations and severities of damage in each case are then identified. The accuracy and effectiveness are verified by comparison of the identified results with the ground truth data.

• Smart Structures and Systems
• /
• 제31권4호
• /
• pp.365-381
• /
• 2023

The existing vision-based techniques for inspection and condition assessment of civil infrastructure are mostly manual and consequently time-consuming, expensive, subjective, and risky. As a viable alternative, researchers in the past resorted to deep learning-based autonomous damage detection algorithms for expedited post-disaster reconnaissance of structures. Although a number of automatic damage detection algorithms have been proposed, the scarcity of labeled training data remains a major concern. To address this issue, this study proposed a semi-supervised learning (SSL) framework based on consistency regularization and cross-supervision. Image data from post-earthquake reconnaissance, that contains cracks, spalling, and exposed rebars are used to evaluate the proposed solution. Experiments are carried out under different data partition protocols, and it is shown that the proposed SSL method can make use of unlabeled images to enhance the segmentation performance when limited amount of ground truth labels are provided. This study also proposes DeepLab-AASPP and modified versions of U-Net++ based on channel-wise attention mechanism to better segment the components and damage areas from images of reinforced concrete buildings. The channel-wise attention mechanism can effectively improve the performance of the network by dynamically scaling the feature maps so that the networks can focus on more informative feature maps in the concatenation layer. The proposed DeepLab-AASPP achieves the best performance on component segmentation and damage state segmentation tasks with mIoU scores of 0.9850 and 0.7032, respectively. For crack, spalling, and rebar segmentation tasks, modified U-Net++ obtains the best performance with Igou scores (excluding the background pixels) of 0.5449, 0.9375, and 0.5018, respectively. The proposed architectures win the second place in IC-SHM2021 competition in all five tasks of Project 2.

• 대한방사선기술학회지:방사선기술과학
• /
• 제46권3호
• /
• pp.197-205
• /
• 2023

When taking X-rays, various auxiliary tools were used to fix a patient's exact shooting position and posture. In this study, we evaluated the usefulness of carbon fiber reinforced plastics(CFRP) 3K as a material of auxiliary tools by comparing poly methyl metha acrylate(PMMA), polycarbonate(PC), and CFRP 3K each of which has high radiolucency. X-ray radiolucencies were measured by stacking 1 mm panels of each material, and contrast to noise ratio(CNR) and signal to noise ratio(SNR) of images of each material were measured by comparing with None, which stands for images that are taken without any material. All three materials showed over 90% X-ray radiolucencies within 2 ㎜ thickness, and there was no significant difference. PC, PMMA and CFRP 3K had high CNR and SNR in order, and CFRP 3K showed the closest CNR and SNR to those of None. While taking X-rays, by using CFRP 3K material within 2 ㎜ thickness as a material of auxiliary tools, which are used to reduce re-shooting and X-ray exposure by fixing a patient's exact shooting position and posture and improve the quality of medical images, a high X-ray radiolucency of over 90% would be obtained, and the influence on the image could be minimized.

• 한국컴퓨터정보학회논문지
• /
• 제14권11호
• /
• pp.61-71
• /
• 2009

자동차의 번호판은 차량의 등록 정보를 확인할 수 있는 유일한 방법이다. 불법 주정차 단속 및 주차 관리 시스템에 차량의 등록 정보를 확인하기 위해 카메라를 이용한 무인 인식시스템의 개발이 활발히 연구되고 있다. 하지만, 일반 도로상에서 날씨나 주변 장애물들은 자동차 번호판 상에 조명 변화를 일으켜 번호판 문자의 추출을 어렵게 한다. 본 논문은 번호판 영상을 개선하여 조명변화에 강인한 문자 추출 알고리즘을 제안한다. 제안하는 기법은 번호판 영상의 명암 대비도를 높이기 위해 Chi-Square 확률 밀도 함수를 이용한다. 또한, 정확한 문자영역을 추출하기 위해, 적응적인 문턱값을 적용함으로써 고품질의 이진화 영상을 얻는다. 번호판의 문자들을 추출하는 일련의 과정에서 방해가 되는 잡음들을 전처리와 레이블링을 통해 제거한다. 마지막으로 번호판의 문자들은 번호판의 기하학적 특징을 이용한 이진화 영상의 프로파일링으로부터 추출된다.

• Smart Structures and Systems
• /
• 제31권5호
• /
• pp.469-483
• /
• 2023

The dynamic characteristics of wind turbine blades are usually monitored by contact sensors with the disadvantages of high cost, difficult installation, easy damage to the structure, and difficult signal transmission. In view of the above problems, based on computer vision technology and the improved YOLOv5 (You Only Look Once v5) deep learning model, a non-contact dynamic characteristic monitoring method for wind turbine blade is proposed. First, the original YOLOv5l model of the CSP (Cross Stage Partial) structure is improved by introducing the CSP2_2 structure, which reduce the number of residual components to better the network training speed. On this basis, combined with the Deep sort algorithm, the accuracy of structural displacement monitoring is mended. Secondly, for the disadvantage that the deep learning sample dataset is difficult to collect, the blender software is used to model the wind turbine structure with conditions, illuminations and other practical engineering similar environments changed. In addition, incorporated with the image expansion technology, a modeling-based dataset augmentation method is proposed. Finally, the feasibility of the proposed algorithm is verified by experiments followed by the analytical procedure about the influence of YOLOv5 models, lighting conditions and angles on the recognition results. The results show that the improved YOLOv5 deep learning model not only perform well compared with many other YOLOv5 models, but also has high accuracy in vibration monitoring in different environments. The method can accurately identify the dynamic characteristics of wind turbine blades, and therefore can provide a reference for evaluating the condition of wind turbine blades.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2022년도 학술발표회
• /
• pp.279-279
• /
• 2022

최근 이미지분석 기술은 하드웨어 및 소프트웨어 기술의 급격한 발전으로 인해 의학, 생물학, 지리학, 재료공학 등에서 수많은 연구 분야에서 광범위하게 활용되고 있으며, 이미지분석은 다량의 토사에 대하여 입경을 포함한 형상학적 특성을 간편하게 정량화 할 수 있기 때문에 매우 효과적인 분석 방법으로 판단된다. 현재 모래의 입도분석 방법으로는 신뢰성 있는 체가름 시험법(KSF2302) 등이 있으나, 번거로운 처리과정과 많은 시간이 소요된다. 또한 입자형상은 입경이 세립 할수록 직접 측정이 어렵기 때문에, 최근에는 이미지 분석을 이용하는 방법이 시도되고 있다. 본 연구에서는 75㎛ 이상의 조립질 하상 토사 이미지를 취득하여, 입자들의 장·축단 길이, 면적, 둘레, 공칭직경 및 종횡비 등의 형상학적 특성인자를 자동으로 측정하는 프로그램 개발을 수행하였다. 프로그램은 이미지 분석에 특화된 라이브러리인 OpenCV(Open Source Computer Vision)를 적용하였다. 이미지 분석 절차는 크게 이미지 취득, 기하보정, 노이즈제거, 객체추출 및 형상인자 측정 단계로 구성되며, 이미지 취득시 패널의 하단에 Back light를 부착해 시료에 의해 발생되는 음영을 제거하였다. 기하보정은 원근변환(perspective transform)을 적용했으며, 노이즈 제거는 모폴로지 연산과 입자간의 중첩으로 인한 뭉침을 제거하기 위해 watershed 알고리즘을 적용하였다. 최종적으로 객체의 외곽선 추출하여 입자들의 다양한 정보(장축, 단축, 둘레, 면적, 공칭직경, 종횡비)를 산출하고, 분포형으로 제시하였다. 본 연구에서 제안하는 이미지분석을 적용한 토사의 형상학적 특성 측정 방법은 시간과 비용의 측면에서 보다 효율적으로 하상 토사에 대한 다양한 정보를 획득 할 수 있을 것으로 기대한다.

• Asia Marketing Journal
• /
• 제16권1호
• /
• pp.215-228
• /
• 2014

The automotive industry plays a significant role in the global economy. One of the reasons is that this industry compasses every aspects of the value chain - from raw materials to design and development, manufacturing, sales and services, and even disposal. Thus, the industry needs significant upfront capital investment and requires years of R&D and market development. As a result, this industry is dominated by a handful of global players and it is not easy for a new entrant to enter this industry. Furthermore, success is even more difficult to achieve. How did Hyundai Motor make it in this tough marketplace? Can it continue against all odds? The CAGR for last 5 years is 12% and it stands at 6th in the world. Compared to other global brands, Hyundai has geographically well-balanced sales portfolio. The quality improvement is outstanding. The brand performance follows these quality and sales improvements. Yet, the global competition is ever intensifying. Now, it is the time to step up once more. The next strategic goal needs fundamental shift toward brand and marketing-focus. In constructing global marketing strategy, Hyundai Motor's vision is "Lifetime partner in mobility and beyond" and its goal is global top 3 brand by year 2015 through modern premium brand image and selling 5 million vehicles. The target brand positioning of Hyundai Motor is the leading position in premium dimension and stylish/modern dimension. The global brand strategy framework is based on the brand direction of "Modern Premium" and is designed to deliver core brand identity (i.e., Simple, Creative, Caring) to customers. In order to manage brand performance, Hyundai's marketing platformalso includes marketing performance management, brand performance management, and market driven organization. From this diagnosis, Hyundai Motor is well posed to build a strong brand. Nevertheless, there are still challenges ahead from consumer, technology, competitor, and macro-environment perspectives. To overcome these threats, the bases of competition for all successful automotive brands are various differentiation factors, including technology, performance, value proposition, or heritage. Hyundai Motor is well prepared so far. However, it is not tested against time yet whether Hyundai can overcome these unforeseeable major threats. Hyundai is trying to find the solution from a strong brand, while believing in "New Thinking. New Possibilities."