• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2001.10a
• /
• pp.156-161
• /
• 2001

Currently, agricultural facilities are evaluated using either basic inspections or detailed analysis. However, conventional analyses as well as methods based on fuzzy logic and rule of thumb have not been very successful in providing a clear relationship between rating and real state of agricultural facilities, because they can't provide exactly acceptable reliability of degraded structures with manager or supervisor. Therefore, in this stage, we must define probabilistic variables for representing degradation of structures being given damages during a survival time. This paper describes the application of neural network system in developing the relation between subjective ratings and parameters of agricultural reservoir as well as that between subjective and analytical ratings. It is shown that neural networks can be trained and used successfully in estimating a rating based on several parameters. The specific application problem for agricultural reservoir in the rural area of Korea is presented and database is constructed to maintain training data set, the information of inspection and facilities. This study showed that a successful training of a neural network could be useful, especially if the input data set for target problem contains parameters with a diverse combination of inter-correlation coefficients. And the networks had a prediction rating of about $^{\ast}^{\ast}^{\ast}%$. The neural network system is expected to show high performance fairly in estimate than statistical method to use equation that is consisted of very lowly interrelated variables.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.6
• /
• pp.2806-2825
• /
• 2018

Unsupervised learning has shown good performance on image, video and audio classification tasks, and much progress has been made so far. It studies how systems can learn to represent particular input patterns in a way that reflects the statistical structure of the overall collection of input patterns. Many promising deep learning systems are commonly trained by the greedy layerwise unsupervised learning manner. The performance of these deep learning architectures benefits from the unsupervised learning ability to disentangling the abstractions and picking out the useful features. However, the existing unsupervised learning algorithms are often difficult to train partly because of the requirement of extensive hyperparameters. The tuning of these hyperparameters is a laborious task that requires expert knowledge, rules of thumb or extensive search. In this paper, we propose a simple and effective unsupervised feature learning algorithm for image classification, which exploits an explicit optimizing way for population and lifetime sparsity. Firstly, a sparse target matrix is built by the competitive rules. Then, the sparse features are optimized by means of minimizing the Euclidean norm ($L_2$) error between the sparse target and the competitive layer outputs. Finally, a classifier is trained using the obtained sparse features. Experimental results show that the proposed method achieves good performance for image classification, and provides discriminative features that generalize well.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.30 no.4 s.247
• /
• pp.364-372
• /
• 2006

In this paper, we present a finger tip tactile sensor which can detect contact normal force as well as slip. The sensor is made up of two different materials, such as polyvinylidene fluoride (PVDF) known as piezoelectric polymer, and pressure variable resistor ink. In order to detect slip on the surface of the object, two PVDF strips are arranged along the normal direction in the robot finger tip and the thumb tip. The surface electrode of the PVDF strip is fabricated using silk-screening technique with silver paste. Also a thin flexible force sensor is fabricated in the form of a matrix using pressure variable resistor ink in order to sense the static force. The developed tactile sensor is physically flexible and it can be deformed three-dimensionally to any shape so that it can be placed on anywhere on the curved surface. In addition, a tactile sensing system is developed, which includes miniaturized charge amplifier to amplify the small signal from the sensor, and the fast signal processing unit. The sensor system is evaluated experimentally and its effectiveness is validated.

• Journal of Korea Multimedia Society
• /
• v.11 no.3
• /
• pp.323-330
• /
• 2008

In this paper, we propose a new fingerprint sensing method that can reduce orientation error. General fingerprint input methods need finger to be put on the surface of a sensor. It can cause of rotation problem and it affects the recognition result significantly. This improved input method can minimize the rotation of a finger by holding double-sided sensor with both thumb and index finger at the same time. Whenever fingerprint is impressed, it has nearly the same orientation because sensors are located between two fingers. As a result, we can get a better performance in fingerprint recognition system, but it may need more hardware cost.

• Archives of Reconstructive Microsurgery
• /
• v.15 no.1
• /
• pp.38-41
• /
• 2006

A 57-year-old male suffered from tingling and paraesthesia on radial side of right hand. We carried out exploration, the process being done, a radial artery passing between the two branches of bifurcated superficial radial nerve was found by accident. Then we performed dissection of the radial artrery a little more distally. It was passing above the extensors of thumb to anstomical snuff box. We treated it with ligating the radial artery, which was done a little above the bifurcated site of superficial radial nerve after angiography of upper extremity. While we were following this procedure, we noticed that symptoms had improved a lot. To our knowlege, neuropathy, by itself of superficial radial nerve by the radial artery with anatomical variation has never been reported and therefore we report this case.

• Advances in Computational Design
• /
• v.1 no.3
• /
• pp.253-264
• /
• 2016

Based on the variation of strain along the cross section, any region in a structural member can be classified into two regions namely, Bernoulli's region (B-region) and Disturbed region (D-region). Since the variation of strain along the cross section for a B-region is linear, well-developed theories are available for their analysis and design. On the other hand, the design of D-region is carried out based on thumb rules and past experience due to the presence of nonlinear strain distribution. Strut-and-Tie method is a novel approach that can be used for the analysis and design of both B-region as well as D-region with equal importance. The strut efficiency factor (${\beta}_s$) is needed for the design and analysis of concrete members using Strut and Tie method. In this paper, equations for finding ${\beta}_s$ for bottle shaped struts in concrete deep beams (a D-region) with and without steel fibres are developed. The effects of transverse reinforcement on ${\beta}_s$ are also considered. Numerical studies using commercially available finite element software along with limited amount of experimental studies were used to find ${\beta}_s$.

• Journal of Electrodiagnosis and Neuromuscular Diseases
• /
• v.20 no.2
• /
• pp.153-158
• /
• 2018

Posterior interosseus (PI) nerve compression is a rare form of compressive neuropathy. It can cause pain along the radial nerve course and weakness of radial nerve innervating muscles without sensory symptoms. A 65-year-old man visited our institution complaining weakness of finger extension and thumb abduction after 2 times of injections at the right elbow in local clinic. The patient's clinical history and physical examination implied an iatrogenic radial nerve injury caused by the injection. The electrophysiologic study revealed of posterior interosseus neuropathy (PIN) with incomplete conduction block. However, the ultrasound study showed that the PI nerve was compressed by an anechoic cyst. The magnetic resonance imaging also confirmed of a ganglion cyst, not a hematoma. After repeated aspirations and a steroid injection, the electrophysiologic study showed recovery of motor weakness. Despite of the clue which implying an iatrogenic injury, clinician should consider other possibilities such as ganglion cysts and ultrasound guided aspiration and steroid injection could be an effective option for conservative management.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.10a
• /
• pp.139-145
• /
• 1996

A forecast on the past output data sets of small bar steels is very important information to make a decision on the future production quantities. In many cases, however, it has been mainly determined by experience (or rule of thumb). In this paper, past basic data sets of each small bar steels are statistically analyzed by some graphical and statistical forecasting methods. This work is mainly done by SAS. Among various quantitative forecasting methods in SAS, STEPAR forecasting method was best performed to the above data sets. By the method, the future production quantities of each small bar steels are forecasted. As a result of this statistical analysis, 95% confidence intervals for future forecast quantities are very wide. To improve this problem, a suitable systematic database system, integrated management system of demand-production-inventory and integrated computer system should be required.

• Journal of Yeungnam Medical Science
• /
• v.20 no.2
• /
• pp.223-228
• /
• 2003

Calcifying aponeurotic fibroma is a rare benign soft tissue tumor that usually involves distal extremities in children and adolescents, especially the hands and feet. We report a case of calcifying aponeurotic fibroma arising in a 14-year-old boy who complained of right thumb mass. Surgical excision was performed. The resected specimen showed a $2.0{\times}1.5cm$ grayish white, fibrotic tissue. Histologic examination showed proliferation of fibroblastic cells with infiltrative growth pattern. Foci of calcification and chondroid differentiation were present.

• Journal of Microbiology and Biotechnology
• /
• v.18 no.2
• /
• pp.328-333
• /
• 2008

A series of pep tides binding to the HCV NS5B polymerase was selected from phage display peptide libraries. A conserved motif of Ser-Arg-X-Arg/Leu was identified among the selected peptides, and Pep2 (Trp-Ser-Arg-Pro-Arg-Ser-Leu) was chosen for further characterization. The binding of Pep2 to HCV NS5B in vivo was shown by a yeast two-hybrid assay and by subcellular colocalization analysis using immunofluorescence confocal microscopy. The in vitro interaction was also confirmed by GST pulldown assay. The replication of the HCV 1b subgenomic replicon was efficiently inhibited by the presence of the peptide. By using a subtractive biopanning against Pep2, the binding site of the peptide was mapped at the pocket of Pro388 to Pro391 in the thumb subdomain of the polymerase. A yeast two-hybrid analysis using Pro388Ala and Pro391Ala mutants of NS5B confirmed the binding.