• Journal of Korean Tunnelling and Underground Space Association
• /
• v.2 no.3
• /
• pp.13-24
• /
• 2000

The collapse shapes and causes of tunnel in the highway were analyzed and reinforced methods of tunnel were investigated in the paper. Collapse shapes of tunnel are divided into three types such as subsurface failure, small scale wedge failure and slickenside strata failure. These three shapes consist of 35%, 50%, and 15%, respectively. The 85% of collapse was located near the entrance and exit of tunnel. The 15% was located at the intersection of emergency laybys. When tunnel collapses are analyzed by the failure concept, sliding failure amounts to more than 83%.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.1
• /
• pp.117-124
• /
• 2008

We suggest CBIR(Content Based Image Retrieval) method using color and shape information. Using just one feature information may cause inaccuracy compared with using more than two feature information. Therefore many image retrieval system use many feature informations like color, shape and other features. We use two feature, HSI color information especially Hue value and CSS(Curvature Scale Space) as shape information. We search candidate image form DB which include feature information of many images. When we use two features, we could approach better result.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1005-1008
• /
• 1993

Supervised learnmg 01 recurrent neural networks (RNNs) is discussed. First, we review the present state of art, featuring their major properties in contrast of those of the multilayer neural networks. Then, we concisely describe one of the most practical learning algorithms, i.e. backpropagation through time. Revising the basic formulation of the learning algorithms, we derive a general formula to solve for the exact solution(s) of the whole connection weights w of RNNs. On this basis we introduce a novel interpretation of the supervised learning. Namely, we define a multidimensional Euclidean space, by assigning the cost function E(w) and every component of w to each coordinate axis. Since E=E(w) turns up as a hyper surface in this space, we refer to the surface as learning surface. We see that topological features of the learning surface are valleys and hills. Finally, after explicating that the numerical procedures of learning are equivalent to descending slopes of the learning surface along the steepest gradient, we show that a minimal value of E(w) is the intersection of curved valleys.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.33 no.3
• /
• pp.10-17
• /
• 2005

An experimental study was conducted at a Mach number of 2.0 to investigate the buzz suppression method on an axisymmetric, external compression supersonic inlet. The inlet model has a fixed geometry with no internal contraction. The inlet configuration was altered by changing the cowling. Results show that source of buzz has been related to the existence in the flow field of velocity discontinuity across a vortex sheet which originates from a shock intersection point. With external compression inlet, buzz can be suppressed by positioning the oblique shock slightly inside or outside of the cowl.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.11
• /
• pp.197-203
• /
• 1999

Increasing demands on precision machining have necessitated the tool to move not only position error as small as possible, but also with smoothly varying feedrates. Because of the lack of accurate and efficient algorithms for generation of 3-dimensional lines and circles, a full accomlishment for available machine tool resolution is generally unavailable. In this paper, a reference-pulse type 3-axis PC_NC milling system is developed for the precision machining of complex shapes in 3-dimensional space. Three AC servomotors are used as the actuator instead of the hand wheel to operate a 3-axis milling machine under the same mechanical structure. A PC is used to handle the control signal calculation for various types of motion command. To achieve the synchronous 3-axis motion, a real-time reference-pulse 3-dimensional linear and circular interpolator based on the intersection criteria is developed in software. The performance test via computer simulation and actual machining have shown that the PC-NC milling system is useful for the machining of arbitrary lines and circles in 3-dimensional space.

• Journal of Korean Foot and Ankle Society
• /
• v.21 no.4
• /
• pp.128-134
• /
• 2017

Purpose: To introduce reliable and newly developed radiographic measures based on a lateral ankle radiograph to assess a syndesmotic reduction after screw fixation and to compare with the radiographic measures based on the anteroposterior (AP) and mortise radiographs. Materials and Methods: The postoperative ankle radiographs of 34 ankle fracture cases after screw fixation for concurrent syndesmosis injury were reviewed. Two radiographic parameters were measured on each AP and mortise radiograph; tibiofibular clear space (TFCS) and tibiofibular overlap (TFO). Five radiographic parameters were measured on the true lateral radiographs; the anteroposterior tibiofibular (APTF) ratio, anterior tibiofibular ratio (ATFR), posterior tibiofibular ratio (PTFR), distances of intersection of the anterior fibular border and the tibial plafond to anterior cortex of the tibia (AA'), and the intersection of posterior fibular border and tibial plafond to the tip of the posterior malleolus (BB'). In addition, the distance (XP) between the fibular posterior margin (X) crossing tibial plafond or the posterior malleolus and posterior articular margin (P) of the tibial plafond was measured on the lateral view. Results: Using TFCS and TFO in the AP and mortise radiographs, malreductions of syndesmosis were estimated in 17 of 34 cases (50.0%). Using the introduced and developed radiographic measures in the lateral radiographs, syndesmotic malreductions were estimated in 16 out of 34 cases (47.1%). Seventeen cases (50.0%) showed no evidence of postoperative diastasis using the radiographic criteria on the AP and mortise view, 10 cases (58.8%) of whom showed evidence of a malreduction on the lateral radiograph. The newly developed measurements, XP, were measured 0 in 11 out of 34 cases (32.4%). Conclusion: The reduction of syndemosis after screw fixation can be accurately assessed intraoperatively with a combination of several reliable radiographic measurements of the lateral radiograph and traditional radiographic measurements of the AP and mortise radiograph.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.9B
• /
• pp.1713-1720
• /
• 1999

This paper presents a new key frame extraction technique, for scene change detection, using the proposed AHIM (Accumulative Histogram Intersection Measure) from the DC image constructed by DCT DC coefficients in the compressed video sequence that is video compression standard such as MPEG. For fast content-based browsing and video retrieval in a video database, we also provide a novel coarse-to-fine video indexing scheme. In the extracted key frame, we perform the region segmentation as a preprocessing. First, the segmented image is projected with the horizontal direction, then we transform the result into a histogram, which is saved as a database index. In the second step, we calculate the moments and change them into a distance value. From the simulation results, the proposed method clearly shows the validity and superiority in respect of computation time and memory space, and that in conjunction with other techniques for indexing, such as color, can provide a powerful framework for image indexing and retrieval.

• Journal of Forest and Environmental Science
• /
• v.37 no.1
• /
• pp.62-68
• /
• 2021

Despite numerous efforts on reducing road-kill worldwide, the collisions have been occurring continuously. Many factors are affecting road-kill occurrences and the effect is various by species. We investigated Hydropotes inermis argyropus road-kill characteristics on the national highway. We examined 9,099 H. i. argyropus road-kill points with distance to the gaps on road (interchange and intersection) and distance to six natural land-cover types as explanatory variables. We also examined the number of road-kill occurrences according to temporal variation using chi-square test with 9,658 events. In general, H. i. argyropus road-kill location tended to occur close to the gaps on road, agricultural lands and forests. The average distance from road-kill to the gap was 694.7 m and 78.6% of the collisions were occurred within 1 km from the gaps. In addition, Kruskal-Wallis test showed the distance between road-kill points and each land cover and the gaps was significantly different. The temporal analyses showed that the differences of the H. i. argyropus road-kill frequency are significant in both month and season. Our results implies H. i. argyropus road-kill location tended to occur close to the gaps on road, agricultural lands and forests in general, especially during May and June, according to their seasonal behavior. Thus, we suggest there is a need of concentrated management on the roads with specific characteristics for both wildlife and human safety.

• Journal of Positioning, Navigation, and Timing
• /
• v.12 no.2
• /
• pp.91-100
• /
• 2023

The existing indoor localization method using Wi-Fi fingerprinting has a high collection cost and relatively low accuracy, thus requiring integrated correction of convergence with other technologies. This paper proposes a new method that significantly reduces collection costs compared to existing methods using Wi-Fi fingerprinting. Furthermore, it does not require labeling of data at collection and can estimate pedestrian travel paths even in large indoor spaces. The proposed pedestrian movement path estimation process is as follows. Data collection is accomplished by setting up a feature area near an indoor space intersection, moving through the set feature areas, and then collecting data without labels. The collected data are processed using Kernel Linear Discriminant Analysis (KLDA) and the valley point of the Euclidean distance value between two data is obtained within the feature space of the data. We build learning data by labeling data corresponding to valley points and some nearby data by feature area numbers, and labeling data between valley points and other valley points as path data between each corresponding feature area. Finally, for testing, data are collected randomly through indoor space, KLDA is applied as previous data to build test data, the K-Nearest Neighbor (K-NN) algorithm is applied, and the path of movement of test data is estimated by applying a correction algorithm to estimate only routes that can be reached from the most recently estimated location. The estimation results verified the accuracy by comparing the true paths in indoor space with those estimated by the proposed method and achieved approximately 90.8% and 81.4% accuracy in two experimental spaces, respectively.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.1
• /
• pp.47-55
• /
• 2024

We spend a lot of time in indoor space, and the space has a huge impact on our lives. Interior design plays a significant role to make an indoor space attractive and functional. However, it should consider a lot of complex elements such as color, pattern, and material etc. With the increasing demand for interior design, there is a growing need for technologies that analyze these design elements accurately and efficiently. To address this need, this study suggests a deep learning-based design analysis system. The proposed system consists of a semantic segmentation model that classifies spatial components and an image classification model that classifies attributes such as color, pattern, and material from the segmented components. Semantic segmentation model was trained using a dataset of 30000 personal indoor interior images collected for research, and during inference, the model separate the input image pixel into 34 categories. And experiments were conducted with various backbones in order to obtain the optimal performance of the deep learning model for the collected interior dataset. Finally, the model achieved good performance of 89.05% and 0.5768 in terms of accuracy and mean intersection over union (mIoU). In classification part convolutional neural network (CNN) model which has recorded high performance in other image recognition tasks was used. To improve the performance of the classification model we suggests an approach that how to handle data that has data imbalance and vulnerable to light intensity. Using our methods, we achieve satisfactory results in classifying interior design component attributes. In this paper, we propose indoor space design analysis system that automatically analyzes and classifies the attributes of indoor images using a deep learning-based model. This analysis system, used as a core module in the A.I interior recommendation service, can help users pursuing self-interior design to complete their designs more easily and efficiently.