• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권1호
• /
• pp.308-320
• /
• 2016

Although the face almost always has an axisymmetric structure, it is generally not symmetrical image for the face image. However, the mirror image of the face image can reflect possible variation of the poses and illumination opposite to that of the original face image. A robust minimum squared error classification (RMSEC) algorithm is proposed in this paper. Concretely speaking, the original training samples and the mirror images of the original samples are taken to form a new training set, and the generated training set is used to perform the modified minimum sqreared error classification(MMSEC) algorithm. The extensive experiments show that the accuracy rate of the proposed RMSEC is greatly increased, and the the proposed RMSEC is not sensitive to the variations of the parameters.

• 대한화학회지
• /
• 제36권1호
• /
• pp.113-124
• /
• 1992

Chemometrics의 한 분야인 패턴인지(pattern recognition)법을 한국산 고대 유리시료 94종의 중성자방사화분석으로부터 얻은 다변수데이타에 적용하였다. unsupervised learning의 방법인 주성분분석과 비선형도시법으로 시료를 분류한 결과 유리시료는 6개의 군을 형성하였다. 6개의 참조시료셋트와 시험시료셋트에 supervised learning의 SIMCA법을 적용시켰다. 그 결과 참조시료셋트는 주성분분석법 및 비선형도시법의 결과와 일치하였고 시험시료셋트에서 33개의 시료 중 17개 시료에 대해 시료가 속한 군을 판정할 수 있었다.

• 한국측량학회지
• /
• 제28권5호
• /
• pp.493-504
• /
• 2010

This study proposes a classification method based on an automated training extraction procedure that may be used with very high resolution (VHR) images of non-accessible areas. The proposed method overcomes the problem of scale difference between VHR images and geographic information system (GIS) data through filtering and use of a Landsat image. In order to automate maximum likelihood classification (MLC), GIS data were used as an input to the MLC of a Landsat image, and a binary edge and a normalized difference vegetation index (NDVI) were used to increase the purity of the training samples. We identified the thresholds of an NDVI and binary edge appropriate to obtain pure samples of each class. The proposed method was then applied to QuickBird and SPOT-5 images. In order to validate the method, visual interpretation and quantitative assessment of the results were compared with products of a manual method. The results showed that the proposed method could classify VHR images and efficiently update GIS data.

• 수산해양교육연구
• /
• 제27권6호
• /
• pp.1628-1645
• /
• 2015

This study was designed to analyze the present condition of the lifelong education field training program and to implement the need assessment of field training program based on the job model application. In order to achieve these goals, surveys were conducted through questionnaires and in-depth interviews targeting staffs of lifelong education institution in Busan area. The 115 subjects were analyzed by frequency distribution, the paired samples t-test, and the Borich's need analysis. The result showed that the field training program was poorly executed during the year of 2013 due to deficiency of certified practitioners and systematic guidelines. Also, the most highly demanded tasks were promoting change, networking, planning, developing programs and conducting surveys among various roles of lifelong educators. On the other hand, the need for the assisting and supporting programs were relatively low. Finally, this study concluded the need for developing systematic guidelines and for accrediting as a lifelong education field training institution nationwide. The results of this study would ultimately contribute to enhance the quality of lifelong education field training program.

• Nutrition Research and Practice
• /
• 제9권1호
• /
• pp.43-48
• /
• 2015

BACKGROUND/OBJECTIVES: Physical exercise promotes energy producing pathways requiring thiamin and riboflavin as a coenzyme. Therefore, this study investigated the effects of regular exercise training on urinary excretion of thiamin and riboflavin. MATERIALS/METHODS: Fifty rats were randomly assigned to one of two groups: non-exercise training (NT, n = 25) and regular exercise training (ET, n = 25) for 5 weeks. The rats performed moderate exercise on a treadmill (0.5-0.8 km/hour) for 30 min/day, 5 days/week. Twenty-four hour urine samples were collected at the end of the 0 week, $3^{rd}$ week, and $5^{th}$ week of training and thiamin and riboflavin were analyzed. RESULTS: No significant differences in thiamin and riboflavin intakes for each week were observed between the NT and ET groups. Urinary thiamin excretion of each group was the highest at the $5^{th}$ week compared to the levels at 0 and $3^{rd}$ week. Urinary thiamin at the $5^{th}$ week was significantly lower in the ET group than in the NT group. Urinary riboflavin excretion was increased by training duration, however, no difference was observed between NT and ET for each week. At 0 and $3^{rd}$ week, no significant relationships were observed between dietary intake and urinary excretion of thiamin and riboflavin, however, at the $5^{th}$ week, urinary excretion was significantly increased by dietary intake only in the NT group (P < 0.05). Thiamin excretion of both NT and ET groups was significantly increased with riboflavin excretion at the $5^{th}$ week (P < 0.01). CONCLUSION: Regular moderate exercise training increased urinary excretion of thiamin. Dietary intakes and urinary excretions of thiamin and riboflavin showed positive correlation in both the exercise training and non-exercise training groups as the exercise training period went by, while the correlations in the exercise training group were weaker than those in the non-exercise training group. Therefore, regular exercise training can alter the urinary excretion of thiamin and riboflavin in rats.

• International Journal of Quality Innovation
• /
• 제7권3호
• /
• pp.58-69
• /
• 2006

This paper presents neural network-based recognition system for automatic light emitting diode (LED) inspection. The back-propagation neural network (BPNN) is proposed and tested. The current-voltage (I-V) characteristic data of LED from the inspection process is used for the network training and testing. This study selects 300 random samples as network training and employs 100 samples as network testing. The experimental results show that if the classification work is done well, the accuracy of recognition is 100%, and the testing speed of the proposed recognition system is almost one half faster than the traditional inspection system does. The proposed neural-network approach is successfully demonstrated by real data sets and can be effectively developed as a recognition system for a practical application purpose.

• Asian-Australasian Journal of Animal Sciences
• /
• 제7권4호
• /
• pp.481-486
• /
• 1994

The effect of neutralization of urea-treated rice straw with sulfuric acid was investigated. Long-cut (15-20 cm) and short-cut (2-3 cm) rice straw were treated with 6% urea for 21 days, and the treated straw was mixed with an acid-molasses solution to neutralize free ammonia and kept airtightly in a plastic bag for 24 hours. The neutralized and non-neutralized straw were dried and subjected to chemical analysis and in vitro dry matter (DM) digestibility determination. The in vitro DM digestibility as well as crude protein (CP) content were remarkably improved by neutralization. Short-cutting of the straw before treatment gave a better result than the long-cut samples. Neutralization with sulfuric acid also affected the chemical composition and increased sulfur content of samples. The CP thus fixed by neutralization was proven to be kept stable for 3 months, and in vitro DM digestibility was not affected by the storing period.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1992년도 하계학술대회 논문집 A
• /
• pp.473-475
• /
• 1992

This paper compares performance of neural and fuzzy truck backer-upper control systems. Conventional controllers require a mathematical model of how outputs depend on inputs. Neural and fuzzy control systems offer a key advantage over conventional control systems. They are model-free controllers. Neural networks learn a control process by examples (training samples). Fuzzy systems directly encode designer's experience as IF-THEN rules. For robustness test, we gradually removed training samples for the neural controller, and fuzzy rules for the fuzzy system. The errors increased laster in the neural controller than in the fuzzy system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권6호
• /
• pp.2709-2729
• /
• 2016

Representation based classification, kernel method and sparse representation have received much attention in the field of face recognition. In this paper, we proposed an improved kernel principal component analysis method based on sparse representation to improve the accuracy and robustness for face recognition. First, the distances between the test sample and all training samples in kernel space are estimated based on collaborative representation. Second, S training samples with the smallest distances are selected, and Kernel Principal Component Analysis (KPCA) is used to extract the features that are exploited for classification. The proposed method implements the sparse representation under ℓ₂ regularization and performs feature extraction twice to improve the robustness. Also, we investigate the relationship between the accuracy and the sparseness coefficient, the relationship between the accuracy and the dimensionality respectively. The comparative experiments are conducted on the ORL, the GT and the UMIST face database. The experimental results show that the proposed method is more effective and robust than several state-of-the-art methods including Sparse Representation based Classification (SRC), Collaborative Representation based Classification (CRC), KCRC and Two Phase Test samples Sparse Representation (TPTSR).

• 한국동물위생학회지
• /
• 제20권1호
• /
• pp.95-102
• /
• 1997

Author investigated internal parasitism for the feces of dog's training center, breeding-dog farm, dairy farm, home by 120 indoor breeding dog and 566 outdoor one in Inchon area. This survey was done from February in 1994 to December in 1995. 1. As a result of total 686 samples, positives were 373(54.4%). Among them, indoor and outdoor breeding dogs were 21(3.1%) 352(51.3%), respectively. 2. According to breeding, it was manifested that 21samples (17.5%) of 120 indoor breeding dogs were positive, and 352 samples(62.2%) of outdoor breeding were positive. 3. The infection rate of dogs for food in dairy farm, breeding dogs in the farm, dog of training center and dog of house is high in order. 4. Infection rate of parasites in 24 dogs breeds, Mongrel dogs were 81.3%, Shepherds were 80.0%, Tosas were 78.4%, Akida and Siberian huskys were 76.2%, Jindos were 55.5%, Pointers were 50.0%, although Afghan hound, Spanial, Shin-tzu, Maltis and Buldog were examined as aparasites negative. 5. After administration with vermicide parasites infection rate were 43.0% in two months. In four months, it were 66.7%, and dogs without vermicide were 87.0%. It seemed like that further research about dosage of vermicide is needed. 6. The rate of single-infection was 37.6% and that of mixed-infection was 16.8%. Among classified 13 types, Ancylostoma caninum 35.6%, Toxocara cams 11.2%, Isospora sp 9.3%, Toxascaris leonina 5.1%, Trychuris vulpis 4.4% were investigated.