• Tunnel and Underground Space
• /
• v.34 no.3
• /
• pp.208-217
• /
• 2024

There is so much research effort for developing and implementing deep learning-based surveillance systems to manage health and safety issues in construction sites. Especially, the development of deep learning-based object detection in various environmental changes has been progressing because those affect decreasing searching performance of the model. Among the various environmental variables, the accuracy of the object detection model is significantly dropped under low illuminance, and consistent object detection accuracy cannot be secured even the model is trained using low-light images. Accordingly, there is a need of low-light enhancement to keep the performance under low illuminance. Therefore, this paper conducts a comparative study of various deep learning-based low-light image enhancement models (GLADNet, KinD, LLFlow, Zero-DCE) using the acquired construction site image data. The low-light enhanced image was visually verified, and it was quantitatively analyzed by adopting image quality evaluation metrics such as PSNR, SSIM, Delta-E. As a result of the experiment, the low-light image enhancement performance of GLADNet showed excellent results in quantitative and qualitative evaluation, and it was analyzed to be suitable as a low-light image enhancement model. If the low-light image enhancement technique is applied as an image preprocessing to the deep learning-based object detection model in the future, it is expected to secure consistent object detection performance in a low-light environment.

• Korean Journal of Cognitive Science
• /
• v.19 no.3
• /
• pp.311-330
• /
• 2008

The effect of spatial attention gradually decreases as a function of the distance between the locus of attention and a target. According to this hypothesis, we tested the spatial-effect profile of visual attention when it operates on perception and memory. Experiment 1 measured accuracy of discriminating the color of a simultaneously masked target after presenting a pre-cue to either at the target location or away from the target (perception-intensive task). Experiment 2 measured accuracy of recognizing the color of several items at and around the pre-cued location (memory-intensive task). In the perception-intensive condition, the accuracy gradually dropped as the distance between the cue and target location increases. However, in the memory-intensive condition, subjects remembered only the item at the cued location. This suggests spatial attention in a memory-intensive process would operate on object-based representations. Experiment 2 showed the object-based effect observed in Experiment 1 can be also present in perception under a special circumstance. The results indicate that spatial attention can operate on object-based representations in a memory-intensive process whereas it flexibly can operate either on location-based or object-based representations in a perception-intensive process.

• Korean Institute of Interior Design Journal
• /
• v.26 no.6
• /
• pp.106-115
• /
• 2017

This study has analysed the characteristics of observation by presence of objects in exhibition spaces through eye-tracking. The area with prevailing observation by the change of observation time with visual-perception response to the presence of objects has been analyzed to figure out the mechanism of sensibility estimation which can take place while the visual-perception is being estimated in spaces. Research results are able to ascertain that First, it was very characteristic that the signs right and left were observed more when there were no objects while when there were objects they were observed prevailingly. Second, the characteristics by section when there were any objects showed that there were more high observation scale I (more than 1000ms). When there were any objects with the number of areas where all the participants prevailed commonly, the scale was more, also. Third, in the process of acquiring any visual-perception information in spaces, the element of landmark can be regarded as a control point for space cognition, where the objects (people) became landmark when there were the objects but the signs became landmark when there were no objects. Fourth, the polynomial trend line of the changes to prevailing observation frequency by observation time shows that there was a gradual average value in general when there were any objects and then after the time range[8] the prevailing observation frequency increased. Without any objects, after a particular time range the value sharply dropped along with the increase of observation time because no objects to be observed prevailingly couln't be seen. The gradual average value means that some elements in the space were prevailingly observed all the time.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.907-917
• /
• 2023

Recently, a self-driving car have applied deep learning technology to advanced driver assistance system can provide convenience to drivers, but it is shown deep that learning technology is vulnerable to adversarial evasion attacks. In this paper, we performed five adversarial evasion attacks, including MI-FGSM(Momentum Iterative-Fast Gradient Sign Method), targeting the object detection algorithm YOLOv5 (You Only Look Once), and measured the object detection performance in terms of mAP(mean Average Precision). In particular, we present a method applying morphology operations for YOLO to detect objects normally by removing noise and extracting boundary. As a result of analyzing its performance through experiments, when an adversarial attack was performed, YOLO's mAP dropped by at least 7.9%. The YOLO applied our proposed method can detect objects up to 87.3% of mAP performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.7
• /
• pp.1585-1609
• /
• 2013

Over recent years, enormous amounts of research in wireless sensor networks (WSNs) have been conducted, due to its multifarious applications such as in environmental monitoring, object tracking, disaster management, manufacturing, monitoring and control. In some of WSN applications dependent the energy-efficient and link reliability are demanded. Hence, this paper presents a routing protocol that considers these two criteria. We propose a new mechanism called Reliable Routing Scheme for Energy-Balanced (RRSEB) to reduce the packets dropped during the data communications. It is based on Swarm Intelligence (SI) using the Ant Colony Optimization (ACO) method. The RRSEB is a self-adaptive method to ensure the high routing reliability in WSNs, if the failures occur due to the movement of the sensor nodes or sensor node's energy depletion. This is done by introducing a new method to create alternative paths together with the data routing obtained during the path discovery stage. The goal of this operation is to update and offer new routing information in order to construct the multiple paths resulting in an increased reliability of the sensor network. From the simulation, we have seen that the proposed method shows better results in terms of packet delivery ratio and energy efficiency.

• Journal of The Korean Association For Science Education
• /
• v.19 no.3
• /
• pp.389-399
• /
• 1999

The purpose of this study was to identify children's mental models of the free-fall objects. This study examined children's prediction and observation about the free-fall of objects. The experimental procedure involved conducting interviews with first-, third-, fifth-, and seventh grade students. The interview had three phases: Prediction, explanation, and observation. During the prediction phase, the object pairs which varied on the dimensions of size, weight, shape, color were presented to students. The students were asked to predict what would happen if the objects were dropped simultaneously. During the explanation phase, the students were asked to explain how they arrived at their answers. During the observation phase, the students observed the free-fall of the object pairs and were asked to describe what they saw. The results showed as follows. (1) Fifth-and seventh grade students made more correct predictions than first- and third grade students. (2) The conflict problems, object pairs involving the dimensions of size and weight, were the most difficult for students to accurately predict. (3) With regard to observations, there was a non-significant effect of grade, indicating equivalence in the number of correct observations made by first-, third-, fifth-, and seventh graders. (4) The conflict problems were the most difficult for students to correctly observe. (5) First- and third grade students showed a significant difference between prediction and observation about the free-fall of objects. However. no difference was found in the fifth- and seventh grade students.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.8A
• /
• pp.859-867
• /
• 2004

To provide high-quality media streaming service over the best-effort Internet, a streaming methodology is required to response to the dynamic fluctuation of underlying networks. In this paper, we implement the MPEG-4 streaming system with adaptive transport based on priorities of media packets. The implemented system is composed of the common MPEG-4 streaming components such as elementary stream provider, sync and DMIF layer, and adaptive transport module including data prioritization and FEC control. More specifically, the prioritized sync layer packets (based on object level) are delivered to the transport module and then are encoded by an adaptive FEC encoder to help reliable transport. The FEC combination is dynamically adjusted by the feedback information from the receiver. In addition, low priority packets are selectively dropped to meet the limitation of available bandwidth. The experimental results over the emulation-based testbed show that the Proposed system can mitigate the impact of network fluctuation and thus improve the quality of streaming.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.3
• /
• pp.2158-2164
• /
• 2015

In u-GIS environments, various load shedding techniques have been researched in order to balance loads caused by input spatial data streams. However, typical load shedding methods on aspatial data lack regard for characteristics of spatial data, also previous load shedding approaches on spatial, which still lack regard for spatial data density or dynamic input data stream, give rise to troubles on spatial query processing performance and accuracy. Therefore, dynamic load shedding scheme over spatial data stream is proposed through stored spatial data deviation and load ratio of input data stream in order to improve spatial continuous query accuracy and performance in u-GIS environment. In proposed scheme, input data which are a big probability related to spatial continuous query may be a strong chance to be dropped relatively.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.3
• /
• pp.282-291
• /
• 2001

In this paper, an efficient VLSI architecture of Shape Adaptive Discrete Cosine Transform(SADCT) based on systolic array is proposed. Since transform size in SADCT is varied according to the shape of object in each block, it are dropped that both usability of processing elements(PE´s) and throughput rate in time-recursive SADCT structure. To overcome these disadvantages, it is proposed that the architecture based on a systolic way structure which doesn´t need memory. In the proposed architecture, throughput rate is improved by consecutive processing of one-dimensional SADCT without memory and PE´s in the first column are connected to that in the last one for improvement of usability of PE. And input data are put into each column of PE in parallel according to the maximum data number in each rearranged block. The proposed architecture is described by VHDL. Also, its function is evaluated by MentorTM. Even though the hardware complexity is somewhat increased, the throughput rate is improved about twofold.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.4 no.1
• /
• pp.17-24
• /
• 2006

A demonstration facility for advanced spent fuel conditioning process (ACP) is under construction in KAERI. In this hot cell facility, all process equipments and materials are taken in and out only through the rear door. The working table in front of the process rear door is specially designed to be openable for the efficient use of the space. This paper presents the structural safety analysis of the openable working table, for the normal operational load condition and accidential drop condition of heavy object. Both cases are investigated through static and dynamic finite element analyses. The analysis results show that structural safety of the working table is sufficiently assured and the working table is not collapsed even when an object of 500 kg is dropped from the height of 50 cm.