• Journal of KIBIM
• /
• v.13 no.4
• /
• pp.96-105
• /
• 2023

This study focuses on developing a building object recognition technology for efficient use in the remodeling of buildings constructed without drawings. In the era of the 4th industrial revolution, smart technologies are being developed. This research contributes to the architectural field by introducing a deep learning-based method for automatic object classification and recognition, utilizing point cloud data. We use a TD3D network with voxels, optimizing its performance through adjustments in voxel size and number of blocks. This technology enables the classification of building objects such as walls, floors, and roofs from 3D scanning data, labeling them in polygonal forms to minimize boundary ambiguities. However, challenges in object boundary classifications were observed. The model facilitates the automatic classification of non-building objects, thereby reducing manual effort in data matching processes. It also distinguishes between elements to be demolished or retained during remodeling. The study minimized data set loss space by labeling using the extremities of the x, y, and z coordinates. The research aims to enhance the efficiency of building object classification and improve the quality of architectural plans by reducing manpower and time during remodeling. The study aligns with its goal of developing an efficient classification technology. Future work can extend to creating classified objects using parametric tools with polygon-labeled datasets, offering meaningful numerical analysis for remodeling processes. Continued research in this direction is anticipated to significantly advance the efficiency of building remodeling techniques.

• Journal of Technologic Dentistry
• /
• v.46 no.2
• /
• pp.36-41
• /
• 2024

Purpose: This study aimed to generate virtual mandibular left first molar teeth using deep convolutional generative adversarial networks (DCGANs) and analyze their matching accuracy with actual tooth morphology to propose a new paradigm for using medical data. Methods: Occlusal surface images of the mandibular left first molar scanned using a dental model scanner were analyzed using DCGANs. Overall, 100 training sets comprising 50 original and 50 background-removed images were created, thus generating 1,000 virtual teeth. These virtual teeth were classified based on the number of cusps and occlusal surface ratio, and subsequently, were analyzed for consistency by expert dental technicians over three rounds of examination. Statistical analysis was conducted using IBM SPSS Statistics ver. 23.0 (IBM), including intraclass correlation coefficient for intrarater reliability, one-way ANOVA, and Tukey's post-hoc analysis. Results: Virtual mandibular left first molars exhibited high consistency in the occlusal surface ratio but varied in other criteria. Moreover, consistency was the highest in the occlusal buccal lingual criteria at 91.9%, whereas discrepancies were observed most in the occusal buccal cusp criteria at 85.5%. Significant differences were observed among all groups (p<0.05). Conclusion: Based on the classification of the virtually generated left mandibular first molar according to several criteria, DCGANs can generate virtual data highly similar to real data. Thus, subsequent research in the dental field, including the development of improved neural network structures, is necessary.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.55-58
• /
• 2024

In this paper, we propose a design and implementation of big data-based fine dust anomaly detection machine learning system. The proposed is system that classifies the fine dust air quality index through meteorological information composed of fine dust and big data. This system classifies fine dust through the design of an anomaly detection algorithm according to the outliers for each air quality index classification categories based on machine learning. Depth data of the image collected from the camera collects images according to the level of fine dust, and then creates a fine dust visibility mask. And, with a learning-based fingerprinting technique through a mono depth estimation algorithm, the fine dust level is derived by inferring the visibility distance of fine dust collected from the monoscope camera. For experimentation and analysis of this method, after creating learning data by matching the fine dust level data and CCTV image data by region and time, a model is created and tested in a real environment.

• Journal of Global Scholars of Marketing Science
• /
• v.18 no.3
• /
• pp.27-57
• /
• 2008

Many consumers enjoy the challenge of purchasing a brand that matches well with their own values and personalities (for example, Ko et al., 2008; Ko et al., 2006). Therefore, the personalities of consumers can impact on the final selection of a brand and its brand personality in two ways: first, the consumers may incline to purchase a brand or a product that reflects their own personalities; second, consumers tend to choose a company that has similar brand personalities to those brands that are being promoted. Therefore, the objectives of this study are following: 1. Is there any empirical relationship between a consumer's personality and the personality of a brand that he or she chooses? 2. Can a corporate brand be differentiated by the brand personality? In short, consumers are more likely to hold favorable attitudes towards those brands that match their own personality and will most probably purchase those brands matching well with their personality. For example, Matzler et al. (2006) found that extraversion and openness were positively related to hedonic product value; and that the personality traits directly (openness) and indirectly (extraversion, via hedonic value) influenced brand effects, which in turn droved attitudinal and purchase loyalty. Based on the above discussion, the following hypotheses are proposed: Hypothesis 1: the personality of a consumer is related to the brand personality of a product/corporate that he/she purchases. Kuksov (2007) and Wernerfelt (1990) argued that brands as a symbolic language allowed consumers to communicate their types to each other and postulated that consumers had a certain value of communicating their types to each other. Therefore, how brand meanings are established, and how a firm communicate with consumers about the meanings of the brand are interesting topics for research (for example, Escalas and Bettman, 2005; McCracken, 1989; Moon, 2007). Hence, the following hypothesis is proposed: Hypothesis 2: A corporate brand identity is differentiated by the brand personality. And there are significant differences among companies. A questionnaire was developed for collecting empirical measures of the Big-Five personality traits and brand personality variables. A survey was conducted to the online access panel members through the Internet during December 2007 in Korea. In total, 500 respondents completed the questionnaire, and considered as useable. Personality constructs were measured using the Five-factor Inventory (NEO-FFI) scale and a total of 30 items were actually utilized. Brand personality was measured using the five-dimension scale developed by Aaker (1997). A total of 17 items were actually utilized. The seven-point Likert-type scale was the format of responses, for example, from 1 indicating strongly disagreed to 7 for strongly agreed. The Analysis of Moment Structures (AMOS) was used for an empirical testing of the model, and the Maximum Likelihood Estimation (MLE) was applied to estimate numerical values for the components in the model. To diagnose the presence of distribution problems in the data and to gauge their effects on the parameter estimates, bootstapping method was used. The results of the hypothesis-1 test empirically show that there exit certain causality relationship between a consumer's personality and the brand personality of the consumer's choice. Thus, the consumer's personality has an impact on consumer's final selection of a brand that has a brand personality matches well with their own personalities. In other words, the consumers are inclined to purchase a brand that reflects their own personalities and tend to choose a company that has similar brand personalities to those of the brand being promoted. The results of this study further suggest that certain dimensions of the brand personality cause consumers to have preference to certain (corporate) brands. For example, the conscientiousness, neuroticism, and extraversion of the consumer personality have positively related to a selection of "ruggedness" characteristics of the brand personality. Consumers who possess that personality dimension seek for matching with certain brand personality dimensions. Results of the hypothesis-2 test show that the average "ruggedness" attributes of the brand personality differ significantly among Korean automobile manufacturers. However, the result of ANOVA also indicates that there are no significant differences in the mean values among manufacturers for the "sophistication," "excitement," "competence" and "sincerity" attributes of the corporate brand personality. The tight link between what a firm is and its corporate brand means that there is far less room for marketing communications than there is with products and brands. Consequently, successful corporate brand strategies must position the organization within the boundaries of what is acceptable, while at the same time differentiating the organization from its competitors.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.33 no.1
• /
• pp.63-72
• /
• 2020

The purpose of this study is to investigate the effects of the application of various numerical models and frequency contents of earthquakes on the performances of the reactor containment building (RCB) in a nuclear power plant (NPP) equipped with an advanced power reactor 1400. Two kinds of numerical models are developed to perform time-history analyses: a lumped-mass stick model (LMSM) and a full three-dimensional finite element model (3D FEM). The LMSM is constructed in SAP2000 using conventional beam elements with concentrated masses, whereas the 3D FEM is built in ANSYS using solid elements. Two groups of ground motions considering low- and high-frequency contents are applied in time-history analyses. The low-frequency motions are created by matching their response spectra with the Nuclear Regulatory Commission 1.60 design spectrum, whereas the high-frequency motions are artificially generated with a high-frequency range from 10Hz to 100Hz. Seismic responses are measured in terms of floor response spectra (FRS) at the various elevations of the RCB. The numerical results show that the FRS of the structure under low-frequency motions for two numerical models are highly matched. However, under high-frequency motions, the FRS obtained by the LMSM at a high natural frequency range are significantly different from those of the 3D FEM, and the largest difference is found at the lower elevation of the RCB. By assuming that the 3D FEM approximates responses of the structure accurately, it can be concluded that the LMSM produces a moderate discrepancy at the high-frequency range of the FRS of the RCB.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.3 no.2 s.5
• /
• pp.129-140
• /
• 2004

In this paper, we conduct the research about optimal aggregation interval of travel time data on interrupted traffic flow and verify the reliability of AVI collected data by using car plate matching method in RTMS for systematic collection and analysis of link travel time data on interrupted traffic flow rural arterial. We perform Kolmosorov-Smirnov test on AVT collected sample data and on entire population data, and conclude that the sample data does not represent pure random sampling and hence includes sample collection error. We suggest that additional review is necessary to investigate the effectiveness of AVI collected sample data as link representative data. We also develop statistical model by applying two estimation techniques namely point estimation and interval estimation for calculating optimal aggregation interval. We have implemented our model and determine that point estimate is preferable over interval estimate for exactly selecting and deciding optimal aggregation interval. Our final conclusion is that 5-minute aggregation interval is optimal to estimate travel time in RTMS, as is currently being used our investigation is based on AVI data collected from Yang-ji to Yong-in $42^{nd}$ National road.

• Tunnel and Underground Space
• /
• v.25 no.5
• /
• pp.462-469
• /
• 2015

This study carried out a topographic survey at a small-scale open-pit limestone mine in Korea (the Daesung MDI Seoggyo office) using a popular rotary-wing unmanned aerial vehicle (UAV, Drone, DJI Phantom2 Vision+). 89 sheets of aerial photos could be obtained as a result of performing an automatic flight for 30 minutes under conditions of 100m altitude and 3m/s speed. A total of 34 million cloud points with X, Y, Z-coordinates was extracted from the aerial photos after data processing for correction and matching, then an orthomosaic image and digital surface model with 5m grid spacing could be generated. A comparison of the X, Y, Z-coordinates of 5 ground control points measured by differential global positioning system and those determined by UAV photogrammetry revealed that the root mean squared errors of X, Y, Z-coordinates were around 10cm. Therefore, it is expected that the popular rotary-wing UAV photogrammetry can be effectively utilized in small-scale open-pit mines as a technology that is able to replace or supplement existing topographic surveying equipments.

• Journal of KIISE:Software and Applications
• /
• v.31 no.1
• /
• pp.54-60
• /
• 2004

Face recognition is one of the problems to be solved by appearance based matching technique. However, the appearance of face image is very sensitive to variation in illumination. One of the easiest ways for better performance is to collect more training samples acquired under variable lightings but it is not practical in real world. ]:n object recognition, it is desirable to focus on feature extraction or normalization technique rather than focus on classifier. This paper presents a simple approach to normalization of faces subject to directional illumination. This is one of the significant issues that cause error in the face recognition process. The proposed method, ICR(illumination Compensation based on Multiple Linear Regression), is to find the plane that best fits the intensity distribution of the face image using the multiple linear regression, then use this plane to normalize the face image. The advantages of our method are simple and practical. The planar approximation of a face image is mathematically defined by the simple linear model. We provide experimental results to demonstrate the performance of the proposed ICR method on public face databases and our database. The experimental results show a significant improvement of the recognition accuracy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.3
• /
• pp.568-576
• /
• 2003

This paper proposes a moving object tracking algorithm by using hierarchical search method in dynamic scenes. Proposed algorithm is based on two main steps: generation step of initial model from different pictures, and tracking step of moving object under the time-yawing scenes. With a series of this procedure, tracking process is not only stable under far distance circumstance with respect to the previous frame but also reliable under shape variation from the 3-dimensional(3D) motion and camera sway, and consequently, by correcting position of moving object, tracking time is relatively reduced. Partial Hausdorff distance is also utilized as an estimation function to determine the similarity between model and moving object. In order to testify the performance of proposed method, the extraction and tracking performance have tested using some kinds of moving car in dynamic scenes. Experimental results showed that the proposed algorithm provides higher performance. Namely, matching order is 28.21 times on average, and considering the processing time per frame, it is 53.21ms/frame. Computation result between the tracking position and that of currently real with respect to the root-mean-square(rms) is 1.148. In the occasion of different vehicle in terms of size, color and shape, tracking performance is 98.66%. In such case as background-dependence due to the analogy to road is 95.33%, and total average is 97%.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.6
• /
• pp.49-59
• /
• 2008

SIFT(Scale Invariant Feature Transform) is an algorithm to extract vectors at pixels around keypoints, in which the pixel colors are very different from neighbors, such as vortices and edges of an object. The SIFT algorithm is being actively researched for various image processing applications including 3-D image constructions, and its most computation-intensive stage is a keypoint localization. In this paper, we develope a fixed-point model of the keypoint localization and propose its efficient hardware architecture for embedded applications. The bit-length of key variables are determined based on two performance measures: localization accuracy and error rate. Comparing with the original algorithm (implemented in Matlab), the accuracy and error rate of the proposed fixed point model are 93.57% and 2.72% respectively. In addition, we found that most of missing keypoints appeared at the edges of an object which are not very important in the case of keypoints matching. We estimate that the hardware implementation will give processing speed of $10{\sim}15\;frame/sec$, while its fixed point implementation on Pentium Core2Duo (2.13 GHz) and ARM9 (400 MHz) takes 10 seconds and one hour each to process a frame.