• Journal of Korea Multimedia Society
• /
• v.3 no.5
• /
• pp.454-460
• /
• 2000

To retrieve complex data such as images, the content-based retrieval method rather than keyword based method is required. In this paper, we implemented a content-based image retrieval system which retrieves object of user query effectively using invariant moments which have invariant properties about linear transformation like position transition, rotation and scaling. To extract the shape feature of objects in an image, we propose a labeling algorithm that extracts objects from an image and apply invariant moments to each object. Hashing method is also applied to reduce a retrieval time and index images effectively. The experimental results demonstrate the high retrieval efficiency i.e precision 85%, recall 23%. Consequently, our retrieval system shows better performance than the conventional system that cannot express the shale of objects exactly.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.601-604
• /
• 2021

본 논문에서는 모바일 디바이스의 카메라로부터 얻은 RGB이미지를 분석하여 장애물을 안정적으로 탐지할 수 있는 프레임워크를 제안한다. 본 논문에서는 장애물을 안정적으로 찾기 위해 RANSAC(Random Sample Consensus)기반의 다중 평면 방식을 이용한 위험감지 시스템을 제안한다. 우리의 접근 방식은 RGB영상으로부터 특징점(Feature point)을 추출하고, 특징점을 분석(Feature point analysis)하여 영상내의 평면을 감지한다. 복잡한 지형으로 인해 생성되는 다수의 평면을 RANSAC을 통해 단일 평면으로 정규화하고, 이로부터 특징점을 분류하기 위한 기준점을 계산한다. 모바일 디바이스의 위치와 회전 제약 없이 효과적으로 기준평면(Reference plane)을 탐색할 수 있고, 영상 내 특징점을 실시간으로 계산한다. 다양한 실험을 통해 기준평면과 장애물과의 거리를 파악하여 장애물을 효과적으로 분류하는 결과를 얻었다. 우리의 기법은 실세계에서의 위험요소를 감지하고 모바일 디바이스 사용자의 안전성 확보에 활용할 수 있을 거라 기대한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.723-726
• /
• 2021

현재 단일 이미지에서 Object Detection 성능은 매우 좋은 편이다. 하지만 동영상에서는 처리 속도가 너무 느리고 임베디드 시스템에서는 real-time이 힘든 상황이다. 연구 논문에서는 하이엔드 GPU에서 다른 기능 없이 YOLO만 구동했을 때 real-time이 가능하다고 하지만 실제 사용자들은 상대적으로 낮은 사양의 GPU를 사용하거나 CPU를 사용하기 때문에 일반적으로는 자연스러운 real-time을 하기가 힘들다. 본 논문에서는 이러한 제한점을 해결하고자 계산량이 많은 Object Detection model 사용을 줄이는 방안은 제시하였다. 현재 Video영상에서 Object Detection을 수행할 때 매 frame마다 YOLO모델을 구동하는 것에서 YOLO 사용을 줄임으로써 계산 효율을 높였다. 본 논문의 알고리즘은 카메라가 움직이거나 배경이 바뀌는 상황에서도 사용이 가능하다. 속도는 최소2배에서 ~10배이상까지 개선되었다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.8A
• /
• pp.820-829
• /
• 2006

IP-based design methodology has been popularly employed for SoC design to reduce design complexity and to cope with time-to-market pressure. Due to the request for high performance of current mobile systems, embedded SoC design needs a multi-processor to manage problems of high complexity and the data processing such as multimedia, DMB and image processing in real time. Interface module for communication between system buses and processors are required, since many IPs employ different protocols. High performance processors require interface module to minimize the latency of data transmission during read-write operation and to enhance the performance of a top level system. This paper proposes an automatic interface generation system based on FSM generated from the common protocol description sequence of a bus and an IP. The proposed interface does not use a buffer which stores data temporally causing the data transmission latency. Experimental results show that the area of the interface circuits generated by the proposed system is reduced by 48.5% on the average, when comparing to buffer-based interface circuits. Data transmission latency is reduced by 59.1% for single data transfer and by 13.3% for burst mode data transfer. By using the proposed system, it becomes possible to generate a high performance interface circuit automatically.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.469-472
• /
• 2022

With the recent development of intelligent transportation systems, various technologies applying deep learning technology are being used. To crackdown on illegal vehicles and criminal vehicles driving on the road, a vehicle type classification system capable of accurately determining the type of vehicle is required. This study proposes a vehicle type classification system optimized for mobile traffic control systems using YOLO(You Only Look Once). The system uses a one-stage object detection algorithm YOLOv5 to detect vehicles into six classes: passenger cars, subcompact, compact, and midsize vans, full-size vans, trucks, motorcycles, special vehicles, and construction machinery. About 5,000 pieces of domestic vehicle image data built by the Korea Institute of Science and Technology for the development of artificial intelligence technology were used as learning data. It proposes a lane designation control system that applies a vehicle type classification algorithm capable of recognizing both front and side angles with one camera.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.825-834
• /
• 2023

The demands of people who purchase fashion products on Internet shopping are gradually increasing, and attempts are being made to provide user-friendly images with 3D contents and web 3D software instead of pictures and videos of products provided. As a reason for this issue, which has emerged as the most important aspect in the fashion web shopping industry, complaints that the product is different when the product is received and the image at the time of purchase has been heightened. As a way to solve this problem, various image processing technologies have been introduced, but there is a limit to the quality of 2D images. In this study, we proposed an automatic conversion technology that converts 2D images into 3D and grafts them to web 3D technology that allows customers to identify products in various locations and reduces the cost and calculation time required for conversion. We developed a system that shoots a mannequin by placing it on a rotating turntable using only 8 cameras. In order to extract only the clothing part from the image taken by this system, markers are removed using U-net, and an algorithm that extracts only the clothing area by identifying the color feature information of the background area and mannequin area is proposed. Using this algorithm, the time taken to extract only the clothes area after taking an image is 2.25 seconds per image, and it takes a total of 144 seconds (2 minutes and 4 seconds) when taking 64 images of one piece of clothing. It can extract 3D objects with very good performance compared to the system.

• Explosives and Blasting
• /
• v.38 no.2
• /
• pp.1-12
• /
• 2020

Digital Image Correlation is a well-established method for displacements, strains and shape measurements of engineering objects. Stereo-camera 3D Digital Image Correlation (3D-DIC) systems have been developed to match the specific requirements for measurements posed by material and mechanical industries. Although DIC method provides the capabilities of scaling a field-of-view(FOV), dimensions of Geotechnical structure objects in many cases are too big to be measured with DIC based on a single camera pair. It can be the most important issue with applying 3D DIC to the measurement of Geotechnical structures. In this paper, We were present stereo vision conditions in a 3D DIC system that can be measured for large FOV(30×20m) and high precisions(z-displacement 0.5mm) of the ground objects with Stereo-camera DIC systems.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.539-554
• /
• 2018

Effective interaction between user and device is considered an important ability of IoT devices. For some applications, it is necessary to recognize human facial expressions in real time and make accurate judgments in order to respond to situations correctly. Therefore, many researches on facial image analysis have been preceded in order to construct a more accurate and faster recognition system. In this study, we constructed an automatic recognition system for facial expressions through two steps - a facial recognition step and a classification step. We compared various models with different sets of data with pixel information, landmark coordinates, Euclidean distances among landmark points, and arctangent angles. We found a fast and efficient prediction model with only 30 principal components of face landmark information. We applied several prediction models, that included linear discriminant analysis (LDA), random forests, support vector machine (SVM), and bagging; consequently, an SVM model gives the best result. The LDA model gives the second best prediction accuracy but it can fit and predict data faster than SVM and other methods. Finally, we compared our method to Microsoft Azure Emotion API and Convolution Neural Network (CNN). Our method gives a very competitive result.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.854-857
• /
• 2011

전자 결재 시스템은 오프라인 문서 결재보다 효율적인 문서 관리가 가능 하고 예산 낭비를 방지할 수 있으며, 작업 능률을 극대화 할 수 있다는 장점 때문에 현재 많은 기업에서 도입하여 사용하고 있다. 하지만 상용화되어 있는 전자 결재 시스템에서 사용되는 이미지 서명은 복제 및 위조가 가능하다는 위협이 존재한다. 이러한 문제를 해결하기 위해서 단순 서명 방식을 적용한 전자 결재 시스템이 제안되었다. 하지만 단순 서명 방식을 이용한 전자 결제 시스템은 단 한 명의 결재자만을 고려한 시스템으로 실제 기업에는 다양한 직책의 결재자가 존재하기 때문에 적용하기가 어렵다. 이에 본 논문에서는 C.Boyd 의 순차 다중서명 방식을 사용하여 다수의 결제자가 서명할 수 있는 전자 결제 시스템을 제안한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.3B
• /
• pp.254-259
• /
• 2011

This paper attempts to propose a method to improve the snapshot boot method, one of the methods to enhance the booting speed of the embedded Linux based system and to increase its efficiency. Snapshot boot is a method of using suspend/resume that is a method of saving and restoring the entire information of the current tasks, and it enhances the booting speed by processing a restoration of previous tasks from the bootloader instead of processing the Linux booting. The facing problem of snapshot boot is that it takes a long time to create images and save them to the storage device during the suspend process since it targets every pages. Additionally, if the switch is turned off while creating an image by some extraordinary circumstances, then the image is not created and thus, the resume process of snapshot boot cannot be executed. The suggest method in this paper creates the snapshot image for once only and utilize the image to enhance the speed and stability on booting.