• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.338-339
• /
• 2021

In this paper, we proposed a multiclass image scene classification method based on transform learning. The method using the Residual Network (ResNet) model which pre-trained on the large image dataset ImageNet for image classification. Compared with the image classification method of the CNN model, it can greatly improve the classification accuracy and efficiency

• Journal of Biosystems Engineering
• /
• v.35 no.2
• /
• pp.69-76
• /
• 2010

This study was conducted to develop an index of fuel consumption to rate agricultural tractors by their fuel efficiencies. The fuel consumption index consisted of two components: basic and operational indexes. The basic index is to consider an average amount of fuel consumed by engine when it transmits 20 and 100% of the rated power. The operational index is to consider the fuel consumed by tractor for typical field operations: plowing, rotavating, and the remains. The equations and procedures to obtain these indexes were proposed. The method and fuel consumption rate to classify tractors into 5 grades were also proposed. The best 15% of the tractor models were rated as the first grade, 20% as the second grade, 30% as the third grade, 20% as the fourth grade, and 15% as the fifth grade in order of fuel efficiency. Using the fuel consumption index, the classification was conducted on 143 tractor models tested at the National Institute of Agricultural Engineering from 2000 to 2007. The proposed 5-grade system of classification using the fuel consumption index could be used to rate the fuel efficiency of 20-100 kW tractor models produced over past 10 years in Korea.

• Journal of Korea Multimedia Society
• /
• v.14 no.8
• /
• pp.1010-1019
• /
• 2011

This was a preliminary research for the goal of understanding between internal structure of Osteogenesis Imperfecta Murine (OIM) bone and its fragility. 54 hyperspectral bone data sets were captured by using JASCO 2000 Raman spectrometer at UMKC-CRISP (University of Missouri-Kansas City Center for Research on Interfacial Structure and Properties). Each data set consists of 1,091 data points from 9 OIM bones. The original captured hyperspectral data sets were noisy and base-lined ones. We removed the noise and corrected the base-lined data for the final efficient classification. High dimensional Raman hyperspectral data on OIM bones was reduced by Principal Components Analysis (PCA) and Linear Discriminant Analysis (LDA) and efficiently classified for the first time. We confirmed OIM bones could be classified such as strong, middle and weak one by using the coefficients of their PCA or LDA. Through experiment, we investigated the efficiency of classification on the reduced OIM bone data by the Bayesian classifier and K -Nearest Neighbor (K-NN) classifier. As the experimental result, the case of LDA reduction showed higher classification performance than that of PCA reduction in the two classifiers. K-NN classifier represented better classification rate, compared with Bayesian classifier. The classification performance of K-NN was about 92.6% in case of LDA.

• Journal of Sasang Constitutional Medicine
• /
• v.22 no.1
• /
• pp.17-25
• /
• 2010

1. Objectives: Voice diagnosis has been used to classify individuals into the Sasang constitution in SCM(Sasang Constitution Medicine) and to recognize his/her health condition in TKM(Traditional Korean Medicine). In this paper, we purposed a new speech classification algorithm for Sasang constitution. 2. Methods: This algorithm is based on the SVM(Support Vector Machine) technique, which is a classification method to classify two distinct groups by finding voluntary nonlinear boundary in vector space. It showed high performance in classification with a few numbers of trained data set. We designed for this algorithm using 3 SVM classifiers to classify into 4 groups, which are composed of 3 constitutional groups and additional indecision group. 3. Results: For the optimal performance, we found that 32.2% of the voice data were classified into three constitutional groups and 79.8% out of them were grouped correctly. 4. Conclusions: This new classification method including indecision group appears efficient compared to the standard classification algorithm which classifies only into 3 constitutional groups. We find that more thorough investigation on the voice features is required to improve the classification efficiency into Sasang constitution.

• Journal of the Korean Society for Library and Information Science
• /
• v.41 no.3
• /
• pp.127-144
• /
• 2007

The purpose of this study is to design the classification system for christian information resources on the internet in Korea. For this purpose, the study is investigated the divisions of Christianity of (1) library classifications: KDC, DDC, LCC, (2) portal sites: Daum, Empas, Naver, (3) Christianity Portal sites GodPeaple, Kidok, Godpia. And it compared the classification systems of KDC, DDC and GodPeaple. This study selected criteria as follows: comprehension, logicality, definiteness, efficiency and current topics. It suggested the classification system(draft) for christian information resources on internet which are composed of 10 classes.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1329-1331
• /
• 2003

Improvement of the classification accuracy is one of the major concerns in the field of remote sensing application research in recent years. Previous research shows that the accuracy of the conventional classification methods based only on the original spectral information were usually unsatisfied and need to be refined by manual edit. This present paper describes a method of combining the image processing, ancillary data (such as digital elevation model) and expert knowledge (especially the knowledge of local professionals) to improve the efficiency and accuracy of the satellite image classification in hilly land. Firstly, the Landsat TM data were geo-referenced. Secondly, the individual bands of the image were intensitynormalized and the normalized difference vegetation index (NDVI) image was also generated. Thirdly, a set of sample pixels (collected from field survey) were utilized to discover their corresponding DN (digital number) ranges in the NDVI image, and to explore the relationships between land use type and its corresponding spectral features . Then, using the knowledge discovered from previous steps as well as knowledge from local professionals, with the support of GIS technology and the ancillary data, a set of conditional statements were applied to perform the TM imagery classification. The results showed that the integration of image processing and spatial analysis functions in GIS improved the overall classification result if compared with the conventional methods.

• Quality Improvement in Health Care
• /
• v.10 no.2
• /
• pp.204-215
• /
• 2003

Background : ICD-10 Classification, which is used domestically as well as internationally, has limited use in the clinical practice since it is developed for at disease statistics and epidemiology. Therefore, the purposes of this study were to improve the quality of diagnosis by constructing a new disease classification based on the diagnoses doctors currently make in the clinical setting and connecting this classification with OCS and EMR, and to meet the demands of doctors for high quality medical study data in medical research. Methods : The specialists in each ophthalmic subfield collected clinical diagnoses and abbreviations based on the ophthalmology textbooks and confirmed the classifications. Total number of clinical diagnoses collected was totaled 672, for which ideal diagnoses had been selected and a new model of disease classification model in connection with ICD-10 was constructed. The constructed classification of clinical diagnoses consisted of six steps: the first step was the classification by ophthalmic subspecialty field; the second to fifth steps were the detailed classification by each specialty field; the sixth step was the classification by site. Results : After introducing the new disease classification, research on the use and a pre-post comparison was conducted. The result from the research on the use of the clinical diagnoses in inpatient and outpatient care has shown a gradually increasing tendency. From the pre-post comparison of EMR discharge summary diagnoses, the result demonstrated that the diagnosis was stated correctly and in detail. Since the diagnosis was stated correctly, code classification became correct as well, which makes it possible to construct high quality medical DB. Conclusion : This construction of clinical diagnoses provides the medical team with high quality medical information. It is also expected to increase the accuracy and efficiency of service in the department of medical record and department of insurance investigation. In the future, if hospitals wish to construct a classification of clinical diagnosis and a standard proposal of clinical diagnosis is presented by a medical society, the standardization of diagnosis seems to be possible.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1995.06a
• /
• pp.225-227
• /
• 1995

This paper compared waveform, cepstrum, and spline wavelet features with nonlinear discriminant analysis. This measure shows efficiency of speech parametrization better than old linear separability criteria and can be used to measure the efficiency of each layer of certain system. Spline wavelet transform has larger gap among classes and cepstrum is clustered better than the spline wavelet feature. Both features do not have good property for classification and we will compare Gabor wavelet transform, Mel cepstrum, delta cepstrum, etc.

• The korean journal of orthodontics
• /
• v.47 no.3
• /
• pp.151-157
• /
• 2017

Objective: The objective of this study was to investigate the differences in masticatory efficiency among patients with different Angle's classes of malocclusion and to assess the correlation between masticatory efficiency and the occlusal contact area. Methods: The mixing ability index (MAI) was calculated for measuring masticatory efficiency of 61 adult patients according to Angle's classifications of malocclusion. The study included 25, 15, and 21 patients with Angle's Class I, II, and III malocclusions, respectively. Silicone interocclusal recording material was used to measure the occlusal contact area. Results: Both the MAI and occlusal contact area showed the highest average values in the Class I malocclusion group, followed by the Class II and Class III malocclusion groups. No significant difference was observed in the MAI values between the Class I and Class II malocclusion groups (p > 0.05), whereas a significant difference was observed between the Class I and Class III malocclusion groups (p < 0.01) and between the Class II and Class III malocclusion groups (p < 0.05). A weak positive correlation was also observed between the MAI and occlusal contact area (p < 0.01, $r^2=0.13$). Conclusions: The results of this study indicated that masticatory efficiency was the highest in patients with Angle's Class I malocclusion, followed by those with Angle's Class II and Angle's Class III malocclusions. Moreover, a weak positive correlation was observed between masticatory efficiency and the occlusal contact area.

• Journal of Korea Multimedia Society
• /
• v.13 no.4
• /
• pp.547-559
• /
• 2010

Delaunay triangulation is suitable for fingerprint matching because of its robustness to rotation and translation. However, missing and spurious minutiae degrade the performance and computational efficiency. In this paper, we propose a method of combining local quality assessment and 4-category minutiae classification to improve accuracy and decrease computational complexity in matching process. Experimental results suggest that removing low quality areas from matching candidate areas and classifying minutiae improve computational efficiency without degrading performance. The results proved that the proposed algorithm outperforms the matching algorithm (BOZORTH3) provided by NIST.