• Proceedings of the Korean Society of Computer Information Conference
• /
• 2009.01a
• /
• pp.169-172
• /
• 2009

본 논문은 SOM을 이용하여 비행슈팅게임을 하는 게이머의 긴장과 이완상태를 학습한다. 학습된 SOM을 이용해 게이머의 새로운 심박데이터가 입력되었을 때 긴장과 이완 상태에서 플레이하는 게이머의 인식을 제안한다. 게이머들은 비행슈팅게임을 플레이하면서 게임 환경들의 패턴들에 익숙해진다. 게이머들은 반복하면서 지루해지면서 자연스럽게 긴장감도 떨어지게 된다. 만약 긴장이완 정도를 알 수 있다면 게이머의 상태에 맞게 게임환경을 조절하여 긴장감을 유지할 수 있을 것이다. 본 연구에서는 비행슈팅게임을 하는 게이머의 심박신호를 이용하여 게이머의 긴장이완상태를 신경망 SOM으로 분류한다. SOM은 주어진 입력패턴에 정확한 답을 정해주지 않고 자기 스스로 학습하여 해답을 찾는 신경망중의 하나이다. 따라서 게이머의 심박신호는 SOM 학습을 통해 게이머의 긴장과 이완상태들을 군집화 할 수 있다. 비행슈팅게임을 20회 반복 플레이하여 SOM으로 게이머의 심박신호를 입력해 본 결과 긴장이완상태를 인식 할 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.08a
• /
• pp.142-145
• /
• 2008

This paper proposes a computationally efficient algorithm for color image segmentation using self-organizing map(SOM) and grassfire algorithm. We reduce a computation time by decreasing the number of input neuron and input data which is used for learning at SOM. First converting input image to CIE $L^*u^*v^*$ color space and run the learning stage with the SOM-input neuron size is three and output neuron structure is 4by4 or 5by5. After learning, compute output value correspondent with input pixel and merge adjacent pixels which have same output value into segment using grassfire algorithm. The experimental results with various images show that proposed method lead to a good segmentation results than others.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.36 no.3
• /
• pp.21-28
• /
• 2008

The management of soil organic matter(SOM) is a key component of golf course green maintenance. As part of a major project examining the sustainable management of SOM on golf course greens, the SOM status of different age greens maintained in the same root zone composition and management were compared. Then the microbial activity, tiller number, bulk density, water content, pH, EC, and T-N in the soil were measured. In the 0${\sim}$5cm depth SOM accumulation showed no significant difference between greens. Below 5cm SOM showed a strong significance between greens and had a positive(+) correlation with year and negative(-) correlation with depth. when regression equations were used to predict SOM accumulation with year and depth, SOM below 5cm tended to increase with a rate of 0.061% . year$^{-1}$ and decrease with a rate of 0.079% . $cm^{-1}$(R2==0.841). Soil microbial activity was investigated with age and depth by using a dehydrogenase assay. Results showed a sharp drop with depth in all greens. The soil microbial activity below 5cm showed no difference between greens. The accumulated SOM below 5cm may be very resistant to decomposition in the long-term. Five years after establishment, the bulk density did not significantly change. The water content, EC, and T-N had a significant correlation with SOM. The pH decreased with the year, which may influence SOM accumulation. Organic matter accumulation was mainly affected by the pH decrase，low soil microbial activity, and high organic matter resistant to decomposition, but the effects of water content, EC, and T-N were obscure.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.5
• /
• pp.280-288
• /
• 2003

In this paper, we propose a design method of the LVQ network using the SOM. The proposed method determines subclasses and initial reference vectors of the LVQ network using the SOM. The efficacy of the proposed method is verified by means of simulations on iris data of Fisher and character recognition. The results show that the proposed method improves considerably on the performance of the conventional LVQ network.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.995-998
• /
• 2013

Aerial imaging is one of the most common and versatile ways of obtaining information from the Earth surface. In this paper, we present an approach by SOM(Self Organization Map) algorithm with 2 phase learning to be applied successfully to aerial images clustering due to its signal-to-noise independency. A comparison with other classical method, such as K-means and traditional SOM, of real-world areal image clustering demonstrates the efficacy of our approach.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.5
• /
• pp.642-646
• /
• 2006

The fuzzy TAM(Topographical Attentive Mapping) network is a supervised method of pattern analysis which is composed of input layer, category layer, and output layer. But if we don't know the target value of the pattern, the network can not be trained. In this case, the target value can be replaced by a result induced by using an unsupervised neural network as the SOM (Self-organizing Map). In this paper, we apply the results of SOM to fuzzy TAM network and show its usefulness through the case study.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.5
• /
• pp.1052-1061
• /
• 2003

In this paper, we propose the car plate recognition using SOM algorithm and describe the parking control system using the proposed car plate recognition. The recognition of car plate was investigated by means of the SOM algorithm. The morphological information of horizontal and vertical edges was used to extract a plate area from a car image. In addition, the 4-direction contour tracking algorithm was applied to extract the specific area, which includes characters from an extracted plate area. The extracted characteristic area was recognized by using the SOM algorithm. In this paper, 50 car images were tested. The extraction rate obtained by the proposed extraction method showed better results than that from the color information of RGB and HSI, respectively. And the car plate recognition using SOM algorithm was very efficient. We develop the parking control system using the proposed car plate recognition that shows performance improvement by the experimental results.

• The Korean Journal of Physiology and Pharmacology
• /
• v.12 no.5
• /
• pp.253-258
• /
• 2008

Somatostatin (SOM) is a widely distributed peptide in the central nervous system and exerts a variety of hormonal and neural actions. Although SOM is assumed to play an important role in spinal nociceptive processing, its exact function remains unclear. In fact, earlier pharmacological studies have provided results that support either a facilitatory or inhibitory role for SOM in nociception. In the current study, the effects of SOM were investigated using anesthetized cats. Specifically, the responses of rostrally projecting spinal dorsal horn neurons (RPSDH neurons) to different kinds of noxious stimuli (i.e., heat, mechanical and cold stimuli) and to the $A{\delta}$ -and C-fiber activation of the sciatic nerve were studied. Iontophoretically applied SOM suppressed the responses of RPSDH neurons to noxious heat and mechanical stimuli as well as to C-fiber activation. Conversely, it enhanced these responses to noxious cold stimulus and $A{\delta}$-fiber activation. In addition, SOM suppressed glutamate-evoked activities of RPSDH neurons. The effects of SOM were blocked by the SOM receptor antagonist cyclo-SOM. These findings suggest that SOM has a dual effect on the activities of RPSDH neurons; that is, facilitation and inhibition, depending on the modality of pain signaled through them and its action site.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.24 no.12
• /
• pp.1167-1172
• /
• 2013

This paper presents a X-band doppler radar with high conversion gain using a self-oscillating-mixer(SOM) that oscillation and frequency mixing is realized at the same time. To improve phase noise of the SOM oscillator, a ${\lambda}/2$ slotted square patch resonator(SSPR) was proposed, which shows high Q-factor of 175.4 and the 50 % reduced circuit area compared to the conventional resonator. To implement the low power system, low biasing voltage of 1.7 V was supplied. To enhance the conversion gain of the SOM, bias circuit is configured near the pinch-off region of transistor, and the conversion gain was optimized. The output power of the proposed SOM was -3.16 dBm at 10.65 GHz. A high conversion gain of 9.48 dB was obtained whereas DC Power consumption is relatively low about 7.65 mW. The phase noise is -90.91 dBc/Hz at 100 kHz offset. The figure-of-merit(FOM) of the proposed SOM was measured as -181.8 dBc/Hz, which is supplier to other SOMs by more than about 7 dB.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.381-384
• /
• 2006

Self Organizing Map(SOM) is a powerful neural network model for unsupervised loaming. In many clustering works with exploratory data analysis, it has been popularly used. But it has a weakness which is the poorly theoretical base. A lot more researches for settling the problem have been published. Also, our paper proposes a method to overcome the drawback of SOM. As compared with the presented researches, our method has a different approach to solve the problem. So, a hybrid SOM is proposed in this paper. Using Monte Carlo computing, a hybrid SOM improves the performance of clustering. We verify the improved performance of a hybrid SOM according to the experimental results using UCI machine loaming repository. In addition to, the number of clusters is determined by our hybrid SOM.