• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.201-207
• /
• 2021

Edge Computing enables image-based inference in close proximity to end users and real-world objects. However, since edge servers have limited computational and accelerator resources, efficient resource management is essential. In this paper, we present HeteroAccel system that performs optimal scheduling in Kubernetes platform based on available node and accelerator information for various inference requests. Our experiments showed 25.3% improvement in overall inference performance over the default scheduling scheme in edge computing environment in which four types of inference services are requested.

• Korean Journal of Remote Sensing
• /
• v.28 no.1
• /
• pp.79-93
• /
• 2012

This paper suggests a system architecture for the advanced meteorological satellite image processing system that is developing by NMSC(National Meteorological Satellite Center). The meteorology satellite image processing system has basically the image acquisition, the image processing, the data management, and the data dissemination functionalities. However, the existing system has some problems with respect to system integration as well as maintenance to accommodate new satellites and/or the new image processing systems for them which will be developed in the near future. To cope with these problems we propose a new system architecture for the advanced meteorology satellite image processing system. To do this we select as the architectural drivers the quality attributes such as modifiability, inter-operability, extendability, reusability, and platform independency and design the architecture to achieve such quality attributes. We expect that the new approach will solve current issues such as system integration, system dependency, or data management problems and will provide easy ways to incorporate new systems and to maintain them.

• Journal of the Korean Society for information Management
• /
• v.24 no.3
• /
• pp.321-341
• /
• 2007

It will be possible to prevent the infringement of the trademarks and the insueing disputes regarding the originality of the trademarks by using an efficient content-based trademark image retrieval system. In this paper, we describe a content-based image retrieval system using the Self Organizing Map(SOM) algorithm. The SOM algorithm utilizes the visual features, which were derived from the gray histogram representation of the images. In addition, we made the objective effectiveness evaluation possible by coming up with a quantitative measure to gauge the effectiveness of the content-based image retrieval system.

• Journal of Korea Multimedia Society
• /
• v.20 no.10
• /
• pp.1619-1627
• /
• 2017

Image hashing has been successfully applied for the problems associated with the protection of intellectual property, management of large database and indexation of content. For a reliable hashing system, improving hash matching accuracy is crucial. In order to improve the hash matching performance, we propose an asymmetric hash matching method using the psychovisual threshold, which is the maximum amount of distortion that still allows the human visual system to identity an image. A performance evaluation over sets of image distortions shows that the proposed asymmetric matching method effectively improves the hash matching performance as compared with the conventional Hamming distance.

• Korean Journal of Remote Sensing
• /
• v.5 no.1
• /
• pp.29-39
• /
• 1989

A study to develop generalized and systematically designed satellite image processing software system on mainframe computer was successfully carried out. Commercially available softwares such as LARSYS were analyzed and modified, and well known satellite data processing algorithms were implemented into comprehensive software. New algorithms were also presented and developed. The contents of developed softwere system may be divided into 8 major sections: menu and user interface, data file management, preprocessing, enhancement in monochrome image, multi-dimension image analysis, scene classification, image display/hardcopy, image handle utility software. Some additional software such as GIS and DBMS will make this software more comprehensive and generalized one for the satellite data processing.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.1243-1244
• /
• 2022

In recent years, the growing interest in off-site construction has led to factories scaling up their manufacturing and production processes in the construction sector. Consequently, continuous large-scale site monitoring in low-variability environments, such as prefabricated components production plants (precast concrete production), has gained increasing importance. Although many studies on computer vision-based site monitoring have been conducted, challenges for deploying this technology for large-scale field applications still remain. One of the issues is collecting and transmitting vast amounts of video data. Continuous site monitoring systems are based on real-time video data collection and analysis, which requires excessive computational resources and network traffic. In addition, it is difficult to integrate various object information with different sizes and scales into a single scene. Various sizes and types of objects (e.g., workers, heavy equipment, and materials) exist in a plant production environment, and these objects should be detected simultaneously for effective site monitoring. However, with the existing object detection algorithms, it is difficult to simultaneously detect objects with significant differences in size because collecting and training massive amounts of object image data with various scales is necessary. This study thus developed a large-scale site monitoring system using edge computing and a small-object detection system to solve these problems. Edge computing is a distributed information technology architecture wherein the image or video data is processed near the originating source, not on a centralized server or cloud. By inferring information from the AI computing module equipped with CCTVs and communicating only the processed information with the server, it is possible to reduce excessive network traffic. Small-object detection is an innovative method to detect different-sized objects by cropping the raw image and setting the appropriate number of rows and columns for image splitting based on the target object size. This enables the detection of small objects from cropped and magnified images. The detected small objects can then be expressed in the original image. In the inference process, this study used the YOLO-v5 algorithm, known for its fast processing speed and widely used for real-time object detection. This method could effectively detect large and even small objects that were difficult to detect with the existing object detection algorithms. When the large-scale site monitoring system was tested, it performed well in detecting small objects, such as workers in a large-scale view of construction sites, which were inaccurately detected by the existing algorithms. Our next goal is to incorporate various safety monitoring and risk analysis algorithms into this system, such as collision risk estimation, based on the time-to-collision concept, enabling the optimization of safety routes by accumulating workers' paths and inferring the risky areas based on workers' trajectory patterns. Through such developments, this continuous large-scale site monitoring system can guide a construction plant's safety management system more effectively.

• The Journal of Korean Society for Radiation Therapy
• /
• v.18 no.2
• /
• pp.67-73
• /
• 2006

Purpose: The number of patients receiving radiotherapy has increased every year and will keep increasing in the future. Therefore, the technique of radiotherapy is developing from day to day, as a result of it, the quantities of image and data used for radiotherapy are also considerably increasing. Therefore, there have been many difficulties in storing, keeping and managing them. Then, we developed and applied this system for improving complicated work process as well as solving these problems with the collaboration Medical Information Team. Materials and Methods: We exported its image at R & V (Record and Verify: Varis vision, Varian, USA) system and planning system after giving some code to be able to access from management system(RO) for department of radiation oncology to PACS. And, we programmed their information by using necessary information among many information included in DICOM head. Results: All images and data generated by our working environment (Simulation CT, L-gram image and internal body structure, DRR, does distribution )were realized at PACS and it became to be possible for clear image to be printed from any computer in department of radiation oncology. Conclusion: It was inevitable to use film during radiotherapy for patients in the past, however, due to the development of this system, film-less system became to be possible. Therefore, the darkroom space and its management cost in relation to the development process disappeared and it became to be unnecessary for spending tangible and intangible financial expense including human resources, time needed for finding film storing space and film and purchasing separate storing equipment for storing images. Finally, we think this system would be very helpful to handle ail complicated processes for radiotherapy and increasing efficiency of overall working conditions.

• Journal of Korea Multimedia Society
• /
• v.22 no.10
• /
• pp.1199-1207
• /
• 2019

Video surveillance data, which is used for preemptive or post-emptive action against any event or accident, is required for monitoring the location, but is reducing the capacity of the image data by removing intervals for cost reduction and system persistence. Such a video surveillance system is fixed in a certain position and monitors the area only within a limited angle, or monitors only the fixed area without changing the angle. At this time, the video surveillance system that is monitored only within a limited angle shows that the variation object such as the floating population shows different status in the image, and the background of the image maintains a generally constant appearance. The static objects in the image do not need to be stored in all the images, unlike the dynamic objects that must be continuously shot, and occupy a storage space other than the necessary ones. In this paper, we propose a mechanism to analyze the image, store only the small size image for the fixed background, and store it as image data only for variable objects.

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.4
• /
• pp.47-53
• /
• 2011

As global warming and has caused the number of abnormal changes, much damage has occurred recently in Korean Peninsula from natural disaster. To prevent and cope with these annually repeated natural hazards, scientific analysis are needed in Disaster Management, including systematic management. In this study, As part of scientific disaster prevention measures, digital image based hazard map for solution to overcome the existing hazard map with schematic representation was constructed. which was combined digital image, digital map and hazard information, and than hazard information system was developed using the hazard map. The hazard information map and hazard information system are able to provide great service to the public and to support quick decision making for hazard management and analysis of natural hazard damage.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.48 no.4
• /
• pp.469-478
• /
• 2012

To manage, sort, and grade fishery resources, it is necessary to measure their morphometric characteristics. This labor-intensive task involves performing repetitive operations on land and on a research vessel. To reduce the amount of labor required, a vision-based automatic measurement system (VAMS) for the measurement of morphometric characteristics of flatfish, such as total length (TL), body width (BW), and body height (BH), has been developed as part of a database management system for fishery resources management. This system can also measure the mass (M) of flatfish. In the present study, we describe a morphological image processing algorithm for the measurement of certain characteristics of flatfish. This algorithm, which involves preprocessing, edge pattern matching, and edge point detection, is effective in cases where the flatfish being measured has a deformed tail and is randomly oriented. The satisfactory performance of the proposed algorithm is also demonstrated by means of experiments involving the measurement of the BW, TL and BH of a flatfish when it is straightened (BW : 117mm, TL : 329mm, BH : 24.5mm), when its tail is deformed, and when it is randomly oriented.