• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.356-358
• /
• 2000

본 논문에서는 인간의 시각적인 감각을 멀티미디어 컨텐츠의 UMA 변환 서비스에 적용하여 영상의 다양한 디스플레이 크기의 사용자 단말기에 맞게 효율적으로 변화(해상도 축소나 Cropping) 기법을 제안하고 있다. 즉, 영상의 중요한 객체를 사각형 경계박스로 표시한 후 각각의 객체의 저자의 의도대로 사용자가 지각적으로 인식할 수 있는 최소의 공간 해상도 축소 정보를 정의하여, 영상의 변환 시 각각 객체를 사용자가 충분히 인식할 수 있는 한계치로 사용하여 효율적인 UMA 서비스를 보장하는 사용자 자원 재분배 기법을 제안한다. 또한, 본 논문에서 제안된 알고리즘을 기존의 방식과 비교하여 실험적으로 그 장단점을 비교한다.

• Tunnel and Underground Space
• /
• v.31 no.5
• /
• pp.374-384
• /
• 2021

In this study, we propose a new approach for automatic fracture detection in CT scan images of rock specimens. This approach is built on top of two-stage object detection deep learning algorithm called Faster R-CNN with a major modification of using rotated bounding box. The use of rotated bounding box plays a key role in the future work to overcome several inherent difficulties of fracture segmentation relating to the heterogeneity of uninterested background (i.e., minerals) and the variation in size and shape of fracture. Comparing to the commonly used bounding box (i.e., axis-align bounding box), rotated bounding box shows a greater adaptability to fit with the elongated shape of fracture, such that minimizing the ratio of background within the bounding box. Besides, an additional benefit of rotated bounding box is that it can provide relative information on the orientation and length of fracture without the further segmentation and measurement step. To validate the applicability of the proposed approach, we train and test our approach with a number of CT image sets of fractured granite specimens with highly heterogeneous background and other rocks such as sandstone and shale. The result demonstrates that our approach can lead to the encouraging results on fracture detection with the mean average precision (mAP) up to 0.89 and also outperform the conventional approach in terms of background-to-object ratio within the bounding box.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.23 no.4
• /
• pp.343-351
• /
• 2010

Bending and axial compressive stresses are distributed across the whole upper flange of a box girder bridge which has the span-to-depth ratio (B/L) of below 0.5, according to Korea Bridge Design Specifications (Minister of Land, Transport and Maritime Affairs, 2005). Shear lag phenomenon, however, can take place in the construction phase of cable-stayed bridge, in which stresses combining bending moment due to dead weight and cable vertical compression are induced. This study aims to analyze the effective width of flange over which composite stresses are given, which should be calculated during the construction phase of stiffening girder of single plane cable-stayed box girder bridge. The study results indicate that the full width of stiffening girder can be regarded as the effective flange width when the span-to-depth ratio for the deck is below 0.38. In other words, the area, where shear lag is taken into consideration, is larger than the width of box girder in single plane cable-stayed box girder bridges. Therefore, the current practice of considering the full width as the effective flange width regardless of changes of the span-to-depth ratio during the construction stage can produce an unsafe bridge. If the effective flange width is determined according to the single span structural system in the early stage of construction when the span-to depth ratio for the deck is high and composite stresses of every part expect each end of the bridge are calculated, it can result in a safe structural design. Since the span-to-depth ratio gradually decreases, however, it is appropriate to determine the effective width of flange on the basis of the full width and the cantilever structural system.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.2 no.1
• /
• pp.93-102
• /
• 2006

In order to use a recycled aggregate concrete for precast concrete box structures, the structural behaviors should be analytically and experimentally evaluated. In this study, full-scale precast concrete box structures are manufactured using the recycled aggregate (30% replacement) and natural aggregate. Then, the flexural failure test and shear failure test as well as water leakage lest for the structures arc carried out. First of all, test results of compressive strength show that recycled aggregate concrete is only 4% lower than normal concrete. In the flexural and shear failure test, the structural performances of precast box using the recycled aggregate concrete are 95% of the capacities of normal precast concrete box or more. Especially, the water leakage test shows that leakage pressure of recycled concrete box is more than the critical value, 60kPa, as well as 9% higher than normal precast concrete box. These test results are analyzed and compared with results of finite clement analysis. The comparison shows that test results are more excellent than analytical results. Also, the comparison confirms the applicability of recycled aggregate concrete for the use as practical precast concrete box structure.

• Journal of the Korean Geotechnical Society
• /
• v.16 no.1
• /
• pp.243-250
• /
• 2000

The mechanical behavior of the underground box culvert constructed by the open cut method depends mainly on the earth pressure acting on it. In this study, the earth pressure on the underground box culverts constructed by the open cut method during and after the construction sequence was numerically analysed by using FLAC. The results are compared with those of the Marston-Spangler's theory, silo theory, and the model tests. The results showed that the vertical earth pressure on the upper slab of the box structure was not uniform. It was as large as the overburden in the middle part of the slab but was smaller or larger than that at its end part depending on the slope of the excavation, the depth of the cover, and the width of the side refill. The horizontal earth pressure on the side wail was much smaller than the earth pressure at rest and grew nonlinearly with the depth.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.1
• /
• pp.30-40
• /
• 2010

The hard disk is one of the most frequently used storage in IPTV sep-top box. It has large storage capacity and provides fast I/O speed compared to its price whereas it causes high power consumption due to mechanical characteristics of spindle motor. In order to play streaming data in the set-top box, spindle motor of hard disk keeps active mode and it causes high power consumption. In this paper, We propose an offset-buffering and multi-mode spin-down method to reduce power consumption for streaming data playback. The offset-buffering inspects the user's viewing pattern and performs buffering based on the analysis of viewing pattern. So, it can maintain the status of spindle motor as idle mode for long time. Besides, it can reduce power consumption by spinning down according to offset-buffer size. The experimental result shows that proposed offset-buffering and multi mode spin-down method is about 28.3% and 12.5% lower than the full-Buffering method in terms of the power consumption and spin-down frequency, respectively.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1141-1147
• /
• 2020

The hornet species are so similar in shape that they are difficult for non-experts to classify, and because the size of the objects is small and move fast, it is more difficult to detect and classify the species in real time. In this paper, we developed a system that classifies hornets species in real time based on a deep learning algorithm using a boundary box. In order to minimize the background area included in the bounding box when labeling the training image, we propose a method of selecting only the head and body of the hornet. It also experimentally compares existing boundary box-based object recognition algorithms to find the best algorithms that can detect wasps in real time and classify their species. As a result of the experiment, when the mish function was applied as the activation function of the convolution layer and the hornet images were tested using the YOLOv4 model with the Spatial Attention Module (SAM) applied before the object detection block, the average precision was 97.89% and the average recall was 98.69%.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.1
• /
• pp.93-101
• /
• 2023

In this paper, we propose a method of creating a 3D bounding box for an object using a vanishing point to increase the accuracy of object recognition in an image when recognizing an traffic object using a video camera. Recently, when vehicles captured by a traffic video camera is to be detected using artificial intelligence, this 3D bounding box generation algorithm is applied. The vertical vanishing point (VP1) and horizontal vanishing point (VP2) are derived by analyzing the camera installation angle and the direction of the image captured by the camera, and based on this, the moving object in the video subject to analysis is specified. If this algorithm is applied, it is easy to detect object information such as the location, type, and size of the detected object, and when applied to a moving type such as a car, it is tracked to determine the location, coordinates, movement speed, and direction of each object by tracking it. Able to know. As a result of application to actual roads, tracking improved by 10%, in particular, the recognition rate and tracking of shaded areas (extremely small vehicle parts hidden by large cars) improved by 100%, and traffic data analysis accuracy was improved.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.77-84
• /
• 2022

In this paper, we propose a Label Reassignment Strategy to improve the performance of an object detection algorithm. Our approach involves two stages: an inference stage and an assignment stage. In the inference stage, we perform multi-scale inference with predefined scale sizes on a trained model and re-infer masked images to obtain robust classification results. In the assignment stage, we calculate the IoU between bounding boxes to remove duplicates. We also check box and class occurrence between the detection result and annotation label to re-assign the dominant class type. We trained the YOLOX-L model with the re-annotated dataset to validate our strategy. The model achieved a 3.9% improvement in mAP and 3x better performance on AP_S compared to the model trained with the original dataset. Our results demonstrate that the proposed Label Reassignment Strategy can effectively improve the performance of an object detection model.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.3
• /
• pp.82-88
• /
• 2015

This paper presents a design of the enable switch circuit, which is consist of discrete device at smart junction box(SJB) board. The Enable switch circuit, which receives ignition signal (IG) for input, sends a drive signal to linear regulator and other elements. The circuit design is carried out in a BCDMOS $0.18{\mu}m$ technology, and the performances are verified through simulations according to AEC-Q100 and ISO 7637-2. Die area of the designed Enable switch circuit is $1.67mm{\times}0.54mm$ in layout, and it is shown that the die can be housed in $3mm{\times}3mm$ HVSON8 package. The designed enable switch circuit is expected to be widely adopted in various automotive SJB's since it can significantly reduce the overall printed circuit board form factor.