• STI Policy Review
• /
• v.5 no.2
• /
• pp.1-19
• /
• 2014

The concept of a holistic innovation policy is defined in this article, with discussions of what it is, why it is relevant, and how it can be implemented to enhance product innovation. It is shown that the innovation systems approach has diffused rapidly during the latest decades and has completely replaced the linear view in the field of innovation research. The majority of European countries are striving in the direction of developing a more holistic innovation policy. However, it is concluded that the innovation policies in European countries are still dominantly linear despite the fact that holistic policy seems to be the driving vision. Innovation policy is behindhand. Why innovation policy is still linear is also preliminarily discussed. Policymakers attending conferences on innovation are practically always in favor of holistic (systemic, broad-based, comprehensive, etc) innovation policies, have abandoned the linear view by learning from innovation research. The division between "linear" and "holistic" seems to be located within the community where innovation policies are designed and implemented, a community composed of policymakers (administrators/bureaucrats) and elected politicians. Perhaps the dividing line is between these two groups in that politicians, who actually make the decisions, may still reflexively believe in the linear view. Nevertheless, there seems to be a failure in communication between researchers and politicians in the field of innovation and there is therefore a strong need to involve innovation researchers in policy design and implementation to a much higher degree. Another way to increase the degree of holism could be to separate innovation policy from research policy, since their integration tends to cement the linear character of innovation policy. The empirical results are based on a questionnaire sent to twenty-three EU Member States, out of which nineteen (83%) responded. Part of the work for this article was carried out for the European Research and Innovation Area Committee (ERAC) of the European Commission (DG RTD).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.543-548
• /
• 2021

The higher the resolution of the image, the higher the satisfaction of the viewers of the image, and the super-resolution imaging has a considerable increase in research value among the fields of computer vision and image processing. In this study, the main features of low-resolution image LR are extracted mainly using deep learning super-resolution models. It learns and reconstructs the extracted features, and focuses on reconstruction-based algorithms that generate high-resolution image HR. In this paper, we investigate SRCNN and VDSR in a super-resolution algorithm model based on reconstruction. The structure and algorithm process of the SRCNN and VDSR model are briefly introduced, and the multi-channel and special form are also examined in the improved performance evaluation function, and understand the performance of each algorithm through experiments. In the experiment, an experiment was performed to compare the results of the SRCNN and VDSR models with the peak signal-to-noise ratio and image structure similarity, so that the results can be easily judged.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1224-1232
• /
• 2022

Whole body bone scan is the most frequently performed nuclear medicine imaging to evaluate bone metastasis in cancer patients. We evaluated the performance of a VGG16-based transfer learning classifier for bone scan images in which metastatic bone lesion was present. A total of 1,000 bone scans in 1,000 cancer patients (500 patients with bone metastasis, 500 patients without bone metastasis) were evaluated. Bone scans were labeled with abnormal/normal for bone metastasis using medical reports and image review. Subsequently, gradient-weighted class activation maps (Grad-CAMs) were generated for explainable AI. The proposed model showed AUROC 0.96 and F1-Score 0.90, indicating that it outperforms to VGG16, ResNet50, Xception, DenseNet121 and InceptionV3. Grad-CAM visualized that the proposed model focuses on hot uptakes, which are indicating active bone lesions, for classification of whole body bone scan images with bone metastases.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.335-349
• /
• 2023

Bridges constantly undergo deterioration and damage, the most common ones being concrete damage and exposed rebar. Periodic inspection of bridges to identify damages can aid in their quick remediation. Likewise, identifying components can provide context for damage assessment and help gauge a bridge's state of interaction with its surroundings. Current inspection techniques rely on manual site visits, which can be time-consuming and costly. More recently, robotic inspection assisted by autonomous data analytics based on Computer Vision (CV) and Artificial Intelligence (AI) has been viewed as a suitable alternative to manual inspection because of its efficiency and accuracy. To aid research in this avenue, this study performs a comparative assessment of different architectures, loss functions, and ensembling strategies for the autonomous segmentation of bridge components and damages. The experiments lead to several interesting discoveries. Nested Reg-UNet architecture is found to outperform five other state-of-the-art architectures in both damage and component segmentation tasks. The architecture is built by combining a Nested UNet style dense configuration with a pretrained RegNet encoder. In terms of the mean Intersection over Union (mIoU) metric, the Nested Reg-UNet architecture provides an improvement of 2.86% on the damage segmentation task and 1.66% on the component segmentation task compared to the state-of-the-art UNet architecture. Furthermore, it is demonstrated that incorporating the Lovasz-Softmax loss function to counter class imbalance can boost performance by 3.44% in the component segmentation task over the most employed alternative, weighted Cross Entropy (wCE). Finally, weighted softmax ensembling is found to be quite effective when used synchronously with the Nested Reg-UNet architecture by providing mIoU improvement of 0.74% in the component segmentation task and 1.14% in the damage segmentation task over a single-architecture baseline. Overall, the best mIoU of 92.50% for the component segmentation task and 84.19% for the damage segmentation task validate the feasibility of these techniques for autonomous bridge component and damage segmentation using RGB images.

• International Journal of Computer Science & Network Security
• /
• v.24 no.3
• /
• pp.59-70
• /
• 2024

Background: The COVID-19 pandemic (the form of coronaviruses) developed at the end of 2019 and spread rapidly to almost every corner of the world. It has infected around 25,334,339 of the world population by the end of September 1, 2020 [1] . It has been spreading ever since, and the peak specific to every country has been rising and falling and does not seem to be over yet. Currently, the conventional RT-PCR testing is required to detect COVID-19, but the alternative method for data archiving purposes is certainly another choice for public departments to make. Researchers are trying to use medical images such as X-ray and Computed Tomography (CT) to easily diagnose the virus with the aid of Artificial Intelligence (AI)-based software. Method: This review paper provides an investigation of a newly emerging machine-learning method used to detect COVID-19 from X-ray images instead of using other methods of tests performed by medical experts. The facilities of computer vision enable us to develop an automated model that has clinical abilities of early detection of the disease. We have explored the researchers' focus on the modalities, images of datasets for use by the machine learning methods, and output metrics used to test the research in this field. Finally, the paper concludes by referring to the key problems posed by identifying COVID-19 using machine learning and future work studies. Result: This review's findings can be useful for public and private sectors to utilize the X-ray images and deployment of resources before the pandemic can reach its peaks, enabling the healthcare system with cushion time to bear the impact of the unfavorable circumstances of the pandemic is sure to cause

• Cartoon and Animation Studies
• /
• s.19
• /
• pp.161-181
• /
• 2010

Human beings experience the outside world through senses and have developed various ways of representation to preserve what they've experienced. The rapid progress of digital technology has opened a new era of representation technology, and furthermore, is functioning as a technology which offers new experiences. The sensory experiences through the sense of sight, which humans depend on more than 70% to perceive the outside world, have been becoming the center of representing the reality as the 3-D graphics technology has been growing, and developing by being grafted onto different areas of study. Various technologies to express the sense of reality, such as the technology to reinforce the virtual reality and to represent it in the reality, computer graphic, TUI technology, and five sensory technologies which apply humans' senses, are making advancement based on humans' visual features and sensory elements. In particular, the 3-D technology to display solidness provides not only representation but also new sensory experiences, and is emerging as the key technology to image contents. However, compared to the development of technology of 3-D graphics, there have been few basic studies on the principles of the sense of vision. Therefore, in this study, the principles and elements to sense videos will be examined. The sensory features of 3-D images to represent the sense of reality will be researched into, especially focusing on the experiential and physiological elements to sense 3-D structures, and the physical and psychological elements to sense shapes, which might be hopefully the basic study for producing 3-D contents.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.4
• /
• pp.236-241
• /
• 2023

Water quality problems in rivers, lakes, and estuaries have become serious in Korea. In order to overcome eutrophication of freshwater lakes and river basins, systematic management of water quality is necessary. To manage water quality in freshwater lakes and basins, apply hydrological models suitable for the basin and water quality models such as rivers and lakes to reduce water pollution based on the prediction results of these models. Improvement measures must be presented. In order to apply appropriate water pollution improvement measures in the watershed, accurate pollution sources must be identified and pollution loads must be predicted and presented. Based on GIS, the connection between the pollutant database and the hydrological and water quality prediction model will be integrated based on spatial location, making it possible to provide systematic support to improve watershed water quality by comprehensively including the water quality modeling process. In this paper, in order to accurately predict water pollution in freshwater lakes and river basins, a water quality model system is established using GIS-based spatial information to present a comprehensive water quality management method for freshwater lake basins in the future, and to systematically manage pollution sources through water quality modeling. This study was conducted to easily and efficiently operate hydrological and water quality models using automated spatial information.

• Journal of the HCI Society of Korea
• /
• v.1 no.2
• /
• pp.43-50
• /
• 2006

A mirror is a familiar tool for human beings who have been seeing themselves through it for a long time since it was created. As evolving Digital Technology, many approaches about digital mirrors which reflect not only the light, but also the information have been studied. Traditional mirrors on the wall do not need any special control to perform their automatic visual feedbacks, reflecting lights. On the contrary, digital mirrors can actively provide more information to the user than the traditional ones. In this paper, we propose an active digital mirror system of which functions are changed according to the user-mirror distance. First of all, we investigated users' behaviors on mirrors and categorized the interactions by user-mirror distance. Based on the previous result, we designed the user interface of the mirror, and developed a prototype which has three recognition modules: a distance measuring module using infrared sensor arrays, a user recognition module by computer vision technique, and a control perception module using infrared sensor grid. In addition, the next steps for improving the user-centered digital mirror system, and the possibility for developing a mirror-shaped computer system were suggested.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.34 no.1
• /
• pp.53-62
• /
• 2016

In recent years, UAV Photogrammetry based on an ultra-light UAS(Unmanned Aerial System) installed with a low-cost compact navigation device and a camera has attracted great attention through fast and accurate acquirement of geo-spatial data. In particular, UAV Photogrammetry do gradually replace the traditional aerial photogrammetry because it is able to produce DEMs(Digital Elevation Models) and Orthophotos rapidly owing to large amounts of high resolution image collection by a low-cost camera and image processing software combined with computer vision technique. With these advantages, UAV-Photogrammetry has therefore been applying to a large scale mapping and cadastral surveying that require accurate position information. This paper presents experimental results of an accuracy performance test with images of 4cm GSD from a fixed wing UAS to demarcate parcel boundaries in agricultural area. Consequently, the accuracy of boundary point extracted from UAS orthoimage has shown less than 8cm compared with that of terrestrial cadastral surveying. This means that UAV images satisfy the tolerance limit of distance error in cadastral surveying for the scale of 1: 500. And also, the area deviation is negligible small, about 0.2%(3.3m²), against true area of 1,969m² by cadastral surveying. UAV-Photogrammetry is therefore as a promising technology to demarcate parcel boundaries.

• Knowledge Management Research
• /
• v.23 no.2
• /
• pp.123-142
• /
• 2022

The non-face-to-face environment accelerated by COVID-19 has speeded up the dissemination of digital virtual ecosystems and metaverse. In order for the metaverse to be sustainable, digital twins that are compatible with the real world are key, and critical technology for that is AR (Augmented Reality). In this study, we examined research trends about AR, and will propose the directions for future AR research. We conducted LDA based topic modeling on 11,049 abstracts of published domestic and foreign AR related papers from 2009 to Mar 2022, and then looked into AR that was comprehensive research trends, comparison of domestic and foreign research trends, and research trends before and after the popularity of metaverse concepts. As a result, the topics of AR related research were deduced from 11 topics such as device, network communication, surgery, digital twin, education, serious game, camera/vision, color application, therapy, location accuracy, and interface design. After popularity of metaverse, 6 topics were deduced such as camera/vision, training, digital twin, surgical/surgical, interaction performance, and network communication. We will expect, through this study, to encourage active research on metaverse AR with convergent characteristics in multidisciplinary fields and contribute to giving useful implications to practitioners.