• Journal of Korean Institute of Industrial Engineers
• /
• v.27 no.1
• /
• pp.47-53
• /
• 2001

In this paper, we compare the classification performances of both ensemble and clustering algorithms (Data Bagging, Variable Selection Bagging, Parameter Combining, Clustering) to logistic regression in consideration of various characteristics of input data. Four factors used to simulate the logistic model are (1) correlation among input variables (2) variance of observation (3) training data size and (4) input-output function. In view of the unknown relationship between input and output function, we use a Taguchi design to improve the practicality of our study results by letting it as a noise factor. Experimental study results indicate the following: When the level of the variance is medium, Bagging & Parameter Combining performs worse than Logistic Regression, Variable Selection Bagging and Clustering. However, classification performances of Logistic Regression, Variable Selection Bagging, Bagging and Clustering are not significantly different when the variance of input data is either small or large. When there is strong correlation in input variables, Variable Selection Bagging outperforms both Logistic Regression and Parameter combining. In general, Parameter Combining algorithm appears to be the worst at our disappointment.

• Journal of Platform Technology
• /
• v.6 no.4
• /
• pp.26-33
• /
• 2018

In this paper, we propose an image retrieval algorithm for real-time processing and design it as hardware. The proposed method is based on the classification of BoWs(Bag of Words) algorithm and proposes an image search algorithm using bit stream. K-fold cross validation is used for the verification of the algorithm. Data is classified into seven classes, each class has seven images and a total of 49 images are tested. The test has two kinds of accuracy measurement and speed measurement. The accuracy of the image classification was 86.2% for the BoWs algorithm and 83.7% the proposed hardware-accelerated software implementation algorithm, and the BoWs algorithm was 2.5% higher. The image retrieval processing speed of BoWs is 7.89s and our algorithm is 1.55s. Our algorithm is 5.09 times faster than BoWs algorithm. The algorithm is largely divided into software and hardware parts. In the software structure, C-language is used. The Scale Invariant Feature Transform algorithm is used to extract feature points that are invariant to size and rotation from the image. Bit streams are generated from the extracted feature point. In the hardware architecture, the proposed image retrieval algorithm is written in Verilog HDL and designed and verified by FPGA and Design Compiler. The generated bit streams are stored, the clustering step is performed, and a searcher image databases or an input image databases are generated and matched. Using the proposed algorithm, we can improve convenience and satisfaction of the user in terms of speed if we search using database matching method which represents each object.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.20 no.4
• /
• pp.235-243
• /
• 2016

Korea is part of a region of low to moderate seismicity located inside the Eurasian plate with bedrock located at depths less than 30 m. However, the spectral acceleration obtained from site response analyses based on the geologic conditions of inland areas of the Korean peninsula are significantly different from the current Korean seismic code. Therefore, suitable site classification scheme and design response spectra based on local site conditions in the Korean peninsula are required to produce reliable estimates of earthquake ground motion. In this study, site-specific response analyses were performed at more than 300 sites with at least 100 sites at each site categories of $S_C$, $S_D$, and $S_E$ as defined in the current seismic code in Korea. The process of creating a huge database of input parameters - such as shear wave velocity profiles, normalized shear modulus reduction curves, damping curves, and input earthquake motions - for site response analyses were described. The response spectra and site coefficients obtained from site response analyses were compared with those proposed for the site categories in the current code. Problems with the current seismic design code were subsequently discussed, and the development and verifications of new site classification system and corresponding design response spectra are detailed in companion papers (II-development of new site categories and design response spectra and III-Verifications)

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.1
• /
• pp.64-69
• /
• 2021

In this paper, a method for improving the defect classification performance in low contrast, ununiformity and featureless steel plate surfaces has been studied based on deep convolution neural network and transfer-learning neural network. The steel plate surface images have low contrast, ununiformity, and featureless, so that the contrast between defect and defect-free regions are not discriminated. These characteristics make it difficult to extract the feature of the surface defect image. A classifier based on a deep convolution neural network is constructed to extract features automatically for effective classification of images with these characteristics. As results of the experiment, AlexNet-based transfer-learning classifier showed excellent classification performance of 99.43% with less than 160 seconds of training time. The proposed classification system showed excellent classification performance for low contrast, ununiformity, and featureless surface images.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.39 no.5
• /
• pp.695-702
• /
• 2015

This study provides basic data that are helpful to prepare a sizing system for dress shirts that improves the wearer's size fitness. The 16 different parts of the dress shirts were measured for 24 pieces of dress shirts with three kinds of size (95, 100 and 105) among the eight different brands on the market. The measurement sizes of the dress shirts analyzed the accuracy of the size information, size classification by size designation, and differences of size by brand. The results of the study were: 1. The size information of dress shirts differed from customer demand. 2. The size increments between size designations differed from each other even though measurement sizes of the dress shirts increased as the size designation increase. 3. Measurement sizes of the dress shirts were different between brands even for dress shirts of the same size designation. It is necessary that manufacturers secure an accurate and standardized sizing system and provide accurate information for the measurement sizes of dress shirts on an online shopping mall.

• Proceedings of the Korea Society of Design Studies Conference
• /
• 2002.05b
• /
• pp.56-57
• /
• 2002

• Proceedings of the Korea Society of Design Studies Conference
• /
• 2001.10a
• /
• pp.63.1-63
• /
• 2001

• Proceedings of the Korea Society of Design Studies Conference
• /
• 2001.10a
• /
• pp.17.2-17
• /
• 2001

• Korean Institute of Interior Design Journal
• /
• v.14 no.5 s.52
• /
• pp.71-79
• /
• 2005

Thinking methods have been widely recognized as phenomenon of problem solving in architectural design process and as one of the bases of creativity. In recent years the study of thinking methods have become a major focus of design research. And the purpose of the paper will understand the phenomenon of characteristic and classification of thinking methods in the architectural design process. Extensive protocols are recorded. In particular, the protocols contain sufficient information to make a detailed picture of the architect's problem-solving processes. A protocol study is reported in which the experimental data by architect's is analyzed through the visual protocol analysis method. These findings will help understand the architectural design nature. And they supply a direction for creative education for architects and the base for CAAD system development through understanding for architect's thinking methods.

• Journal of KIBIM
• /
• v.5 no.4
• /
• pp.37-46
• /
• 2015

In a construction project, collaboration amongst the project participants is a critical factor for high-quality results and successful completion of the project. Owing to the advance of information technologies, web-based systems have become more common in the construction industry, but research and development has been made for only limited areas. For organized and systematic collaboration in various fields, collaboration systems have to be developed in a holistic manner based on diverse needs from the whole construction industry. This study aims to investigate the current status of web-based collaboration systems from structural engineers' perspectives and propose an improvement plan. For a systematic analysis of selected cases, we apply a classification of three developmental stages depending on interoperability and organizational levels: structural design and analysis, collaborative design, and integrated design management. Thereafter, the characteristics of each stage are extracted and comparatively analyzed. Lastly, three functional factors were proposed for the improvement of web-based collaboration systems for structural design.