• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2007.04a
• /
• pp.27-32
• /
• 2007

Determination of crack depth in filed using the self-calibrating surface wave transmission measurement and the cutting frequency in the transmission function (TRF) is very difficult due to variations of the measurement conditions. In this study, it is proposed to use the measured full TRF as a feature for crack depth assessment. A principal component analysis (PCA) is employed to generate a basis of the measured TRFs for various crack cases. The measured TRFs are represented by their projections onto the most significant principal components. Then artificial neural networks (NNs) using the PCA-compressed TRFs is applied to assess the crack in concrete. Experimental study is carried out for five different crack cases to investigate the effectiveness of the proposed method. Results reveal that the proposed method can be effectively used for the crack depth assessment of concrete structures.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.6
• /
• pp.75-82
• /
• 2005

The robust feature vector selection method for the multiple frontal face detection is proposed in this paper. The proposed feature vector for the training and classification are integrated by means, amplitude projections, and its 1D Harr wavelet of the input image. And the statistical modeling is performed both for face and nonface classes. Finally, the estimated probability density functions (PDFs) are applied for the detection of multiple frontal faces in the still image. The proposed method can handle multiple faces, partially occluded faces, and slightly posed-angle faces. And also the proposed method is very effective for low quality face images. Experimental results show that detection rate of the propose method is $98.3\%$ with three false detections on the testing data, SET3 which have 227 faces in 80 images.

• Korean Journal of Cognitive Science
• /
• v.7 no.1
• /
• pp.37-56
• /
• 1996

In this paper,we design a verifier for unconstrained handwritten numerals using structural feature analysis,and use it as a comnination algorithm for multiple recognizers.The existing combination algorithms mainly use learnings,statistical methods,or probabilistic methods without considering structural features of numerals.That is why they cannot recognize some numerals which human can identify clearly.To overcome the shortcomings,we design one-to-one verifiers which compare and analyze the relative structural features between frequently confused numeral pairs,and apply them to combine multiple recongnizers.Structural features for verification consist of contour,direction al chain code,polygonal approximation,and zero crossing number of horizontal/vertical projections. We gained a 97.95% reliability with CENPARMI numeral data,and showed that some misconceived factors generated from typical combination algorithms can be removed.

• International journal of advanced smart convergence
• /
• v.4 no.2
• /
• pp.20-28
• /
• 2015

In this paper, we propose an effective tracking algorithm with an appearance model based on features extracted from a video frame with posture variation and camera view point adaptation by employing the non-adaptive random projections that preserve the structure of the image feature space of objects. The existing online tracking algorithms update models with features from recent video frames and the numerous issues remain to be addressed despite on the improvement in tracking. The data-dependent adaptive appearance models often encounter the drift problems because the online algorithms does not get the required amount of data for online learning. So, we propose an effective tracking algorithm with an appearance model based on features extracted from a video frame.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.4
• /
• pp.534-540
• /
• 2010

High dimensionality and highly correlated features are the major characteristics of hyperspectral data. Linear projections such as LDA and its variants have been used in extracting low-dimensional features from high-dimensional spectral data. Regularization of LDA has been introduced to alleviate the overfitting that often occurs in a small-sized training data set and leads to poor generalization performance. Among them, a smoothness regularized LDA seems to be effective in the feature extraction for hyperspectral data due to its capability of utilizing the high correlatedness. This paper studies the performance of the regularized LDA in hyperspectral data classification experimentally with varying conditions of the training data. In addition, a new dual smoothness regularized LDA is proposed and evaluated that makes use of both the spectral-domain and spatial-domain correlations between neighboring pixels.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.7
• /
• pp.187-195
• /
• 2019

In December 2015, The Paris Agreement was adopted to undertake ambitious efforts to combat climate change. Korean government announced its goal of reducing the country's greenhouse gas emissions by up to 37% below business as usual projections by 2030 in 2015. The purpose of this study was to set up the calculation methodology of GHG emission($CO_{2e}$) in building sector and to estimate the annual GHG emission in building sector based on national energy consumption statistic. The GHG emission from buildings is about 135.8 million ton $CO_{2e}$ as of 2015, taking up about 19.6% of Korea's entire emission and is about 144.7 million ton $CO_{2e}$ in 2017. The GHG emission of building sector is increasing at annual rate of 2.0% from 2001 to 2017. The GHG emission from electricity consumption in buildings is 91.8 million ton $CO_{2e}$ in 2017, is the highest $CO_2$ emission by energy source. The results show that the intensity of GHG emission of residential building sector is $40.6kg-CO_{2e}/m^2{\cdot}yr$ and that of commercial building sector is $68.4kg-CO_{2e}/m^2{\cdot}yr$.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.20 no.3
• /
• pp.247-253
• /
• 2007

Determination of crack depth in filed using the self-calibrating surface wane transmission measurement and the cutting frequency in the transmission function (TRF) is very difficult due to variations of the measurement conditions. In this study, it is proposed to use the measured full TRF as a feature for crack depth assessment. A Principal component analysis (PCA) is employed to generate a basis of the measured TRFs for various crack cases. The measured TRFs are represented by their projections onto the most significant principal components. Then artificial neural networks (NNs) using the PCA-compressed TRFs is applied to assess the crack in concrete. Experimental study is carried out for five different crack cases to investigate the effectiveness of the proposed method. Results reveal that the proposed method can be effectively used for the crack depth assessment of concrete structures.

• The Research Journal of the Costume Culture
• /
• v.14 no.2
• /
• pp.271-285
• /
• 2006

The purpose of this study was to develop a new basic skirt pattern for 4 lower body types of Plus-sized women. To reduce a time and economic loss about putting it on, This study applied a 3D sample module in the course to develop a basic skirt pattern which is fit, functional and beautiful. A data analysis used the SPSS 11.0 statistics program. To verify the difference, This study used the variance analysis and Duncan's test for the postmortem verification. The result was as follows. To reflect a feature according to an abdomen projection and hip projection, we added the maximum hip circumference to the waist circumference and hip circumference for necessary for a skirt draft item. The front hip circumference sets to the H/4+1cm(ease)+D/4. The back hip circumference set to the H/4+0.5cm(ease)+D/4. The D is the hip circumference in the maximum hip circumference to subtract. The front waist circumference sets to the W/4+0.5cm(ease)+0.5cm(A front and back's difference)+D/4. The back waist circumference sets to the W/4+0.5cm(ease)-0.5cm(A front and back's difference)-D/4. Compare with 4 lower body types of Plus-sized Women, A type 2 and type 4 abdomen projections are bigger. We raised a front waistline of skirt pattern to a 0.3cm upside, and take about 1.3cm down the skirt length from a developed skirt pattern. Consequently, the appearance of a front skirt length was improved with a side skirt length.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.2
• /
• pp.153-162
• /
• 2012

In this paper, we propose a new shape similarity measure for M:N polygon pairs regardless of different object cardinalities in the pairs. The proposed method compares the projections of two shape functions onto Zernike polynomial basis functions, where the shape functions were obtained from each overall region of objects, thus not being affected by the cardinalities of object pairs. Moments with low-order basis functions describe global shape properties and those with high-order basis functions describe local shape properties. Therefore several moments up to a certain order where the original shapes were similarly reconstructed can efficiently describe the shape properties thus be used for shape comparison. The proposed method was applied for the building objects in the New address digital map and a car navigation map of Seoul area. Comparing to an overlapping ratio method, the proposed method's similarity is more robust to object cardinality.

• Journal of Chest Surgery
• /
• v.12 no.4
• /
• pp.324-335
• /
• 1979

During the period of 4 years from August 1975 to August 1979 one hundred and forty seven cases of lung cancer were seen at Paik Foundation Hospital in Seoul, Korea. Among these 147 cases, 104 patients had primary carcinoma of the lung and the remainder was metastatic carcinoma to the lung. Among these 104 primary carcinoma patients, 73 cases were proven histologically as primary carcinoma of the lung. There were three cases of alveolar cell carcinoma [Table 1 ]. This clinical observation is based on those 73 cases including three case reports of the alveolar cell carcinoma. 1. Peak incidence was observed in the 5th decade of life. Male to female ratio was 2 to 1 [Fig. 1]. 2. Pathological classifications were as follows: epidermoid carcinoma, 24 cases [32.9%]; undifferentiated carcinoma, 20 cases [27.4%]; adenocarcinoma, 15 cases [20.5%]; bronchioloalveolar carcinoma [5.5%] and positive cytology, 10 cases [13.7%] [Fig. 2]. 3. Evidence of inoperability was observed in 55 patients [75% of the 73 cases] [Table 3]. 4. Among those 73 cases, operability was evaluated in 18 patients or 25%. One patient refused operation and 17 patients [23.6%] were explored. In 11 [15%] out of 17 patients, thoracotomies were performed. Six cases were pneumonectomies and 5 cases were lobectomies or bilobectomies [Fig. 3]. 5. First case of alveolar cell carcinoma was a 46 year-old housewife complaining of cough and hemoptysis for one year. The plain chest X-ray and bronchogram showed characteristic pictures as Figures 4 and 5. A pneumonectomy was carried out. Histologically, a beautiful alveolar carcinoma consisted of the characteristic tall columnar epithelial cells, which were lining the alveolar spaces as seen in Figures 6, 7, 8, and 20. 6. In the second case of 41 year old male, predominant clinical feature was single, well defined mass in the right lower lobe [Fig. 10 and 11] on chest X-ray. Bilobectomized specimen showed fragile, soft and hard tissue containing mucoid secretions and focal yellowish necrosis with pigmentation on cut surface [Fig. 12]. Slides showed tumor cells lined up along the alveolar septa with papillary projections [Fig. 13 and 14]. 7. Third case of alveolar cell carcinoma was a 50-year-old housewife with hemoptysis. An outstanding clinical picture was a round to lobulated mass in the right upper lobe [Fig. 16]. She is living now, 2 years and 1 month post-operatively, but has arrived at terminal stage with military nodular disseminations to the contralateral lung [Fig. 19].